Rosenthal & Wong Before And Beyond Divergence Before And

rosenthal & wong before and beyond divergence before and beyond divergence: the politics of economic change in china and europe jea

Rosenthal & Wong Before and Beyond Divergence
Before and Beyond Divergence: The Politics of Economic Change in China
and Europe
Jean-Laurent Rosenthal (Caltech)
and
R. Bin Wong (UCLA)
4/25/2010
Table of Contents
Preface
Unrevised
1
Introduction: Politics and Economic Change in the Making of the
Modern World; Miracles, Myths and Explanations in Economic
History
Revised
5
Chapter 1: Space and Politics
Revised
18
Chapter 2: Population Resources and Economic Growth
Revised
45
Chapter 3: Formal and Informal Mechanisms for Market Development
Revised
87
Chapter 4: Warfare, Location of Manufacturing, and Economic
Growth in China and Europe
Revised
130
Chapter 5: Credit Markets and Economic Change
Revised
167
Chapter 6: Autocrats, War, Taxes, and Public Goods
Revised
214
Chapter 7: Political Economies of Growth, 1500-1950: Chinese
Empire and European State
Revised
258
Conclusion: Historical Perspectives on Politics and Economic
Change in China and Europe: Findings, Methods and Implications
Revised
276
Bibliography
Revised
302
Why did China decline between 1400 and 1980 only to re-establish a
major presence in the global economy? Why did Europe, a region torn by
strife and suffering and economic collapse after the fall of the Roman
Empire, become the birthplace of modern economic growth? These two
questions are at the forefront of research in economic history and
important elements of public discussions. This book answers both
questions with a new explanation for the distinctive patterns of
economic change in China and Europe. As specialists on these two
regions of the world, we make specific comparisons of similar
processes. We pose, whenever possible, falsifiable propositions so
that our explanations of particular phenomena can be challenged,
qualified or confirmed by future research. We begin with a review of
some conventional arguments offered for both China’s failures and
Europe’s successes. Some of these we reject based on their inability
to explain known facts. Others we accept but place into a larger
framework of explanation that allies price theory and political
economy. We contend that this approach provides a more satisfying
discussion of the issues and formulates better answers to the big
questions than do the conventional narratives. Our collaboration
suggests that the alliance of economic theory with expertise in the
history of both China and Europe makes for better economic history.
The introduction lays out our approach. While the problem we tackle is
a major issue in economic history we eschew any grand theory. Rather
we seek to develop a series of accounts that are in accord with what
we know and that could be falsified (or affirmed) by future research.
Chapter One examines the long history of political fragmentation in
Europe and unity in China. We argue that political equilibria were not
foreordained consequences of geography and culture. China’s empires
collapsed numerous times and fragmentation persisted for centuries.
The Roman Empire was about as successful as any dynasty in China. The
peculiar process in Europe is the failure to reform an integrated
polity at any time between the sixth and the twenty-first centuries.
The rest of the volume explores in detail the consequences of
political fragmentation for five key areas of the economy.
In Chapter Two we consider demography and labor markets. We argue that
neither differences in demographic regimes nor the degree of
efficiency in labor markets explain many of the economic differences
that emerged between China and Europe after 1300. More generally, we
emphasize that not all variations in institutions necessarily have
important economic effects. Chapter Three considers formal and
informal institutions for market development. The Chinese and
Europeans relied on both types of institutions; the relative intensity
of their use varied over time and space. We suggest that for both
those specific institutions and formal and informal institutions
generally, the geographical scale of the polities explains much of the
variation in their relative importance. Yet these contrasts, like
those presented in Chapter Two, do not seem crucial to long-run
economic growth. Chapter Four identifies some factors we think were
very important to long-run growth possibilities. We develop this
chapter’s arguments from elements of Chapters Two and Three: (1)
relative prices matter to how economies evolve; (2) institutions can
have large effects on relative prices; (3) and the spatial scale of
polities can affect the choice of institutions over the long run. Our
thesis is that warfare biased European manufacturing towards urban
location and in the long run towards capital intensive methods.
Europe’s eighteenth century breakthrough was an unintended consequence
of political economy.
Chapter Five moves our argument to credit markets. We review arguments
that differences in interest rates favored European economic
development. While this is no doubt true, we suggest that the
differences were neither as extreme as sometimes proposed, nor as
consequential as many analyses of credit markets would have. Capital
markets grew in Europe earlier and on a larger scale than they did in
China as a result of differences that were more political and social
than economic in origin and the advantages of well-developed capital
markets did not become economically important until the nineteenth
century. Hence differences in credit markets were another unintended
consequence of political structure. Chapter Six looks at the public
finance aspect of political structures more closely and finds that
fiscal expenditures differed in China and Europe; before 1800 those
made in China were both intended to, and in fact did, facilitate
economic growth more than those made in Europe. Chapter Seven reprises
the problems raised in Chapter 1, to consider the evolution of
political structures and economies after 1500 with an emphasis on the
period after the French Revolution. Then Europe’s colonial expansion
was joined with the spread of representative institutions. In China,
at the same time, the Qing dynasty began to confront domestic
disturbances and not so long thereafter external invaders. We argue
that China’s dismal economic performance in the nineteenth century
owes more to these political travails than to an intrinsic incapacity
to adopt or adapt Western technologies. The conclusion bring us to the
present by arguing that the political structure inherited from the
first millennium AD still have importance consequences for how China
and Europe meet today’s challenges.
Upon finishing the book, we hope readers understand that political
economy matters to economic history in basic ways. Second, we should
have demonstrated that the kind of history we explain matters to
understanding present practices and future possibilities. Third, the
political economy of earlier periods of Chinese and European economic
history makes clear the distinction between the intentions of actors
and the significance of their actions, including unintended outcomes.
Appreciating this distinction can help us better plan our desired
futures and be more modest about our expected successes.
Introduction
Politics and Economic Change in the Making of the Modern World:
Miracles, Myths and Explanations in Economic History
In the last three decades scholars have reconsidered what set Europe
off from the rest of the world as the site of state formation and
economic changes leading to modern national states and industrialized
economies. The themes are of course far older. They recur in the
enquiries of great social thinkers from Montesquieu, to Smith through
Karl Marx and Max Weber. Europe’s, and in particular England’s,
success moved scholars to assess other societies from a European
benchmark. In particular, quite diverse social science scholarship
presumed that there was unique and European-defined path to
modernization and prosperity. But beginning in the 1980s doubts about
the intrinsic superiority of Western political and economic practices
crept into public discussions as Japan’s rise to prominence as the
world’s second largest economy was followed by economic
transformations in South Korea, Taiwan, Hong Kong and Singapore. The
political and economic evolution of East Asia raised serious questions
about the Western origins of contemporary political and economic ideas
and institutions.
With China’s persistent high rates of economic growth, East Asia has
once again become a region of fundamental importance to the
contemporary world economy. Thirty years of rapid growth with few
legal changes and little political change more generally (either
towards decentralization or democratization) also forces us to
reconsider the extent to which we can account for the development of
China with paradigms deduced from European history. If in fact
European ideas and institutions are inadequate to explain China
successful growth, can we nevertheless put forward a method of
comparison to evaluate the significance of similarities and
differences between the two ends of Eurasia?
This book argues that this can and should be done. Rather than focus
on the recent past, we consider the process of divergence that
preceded the onset of modern economic growth in the eighteenth
century. In writing such historical analyses scholars often seek to
encompass all institutional details with little distinction between
significant and trivial institutional differences.{NEED CITATION} Our
premise is that social scientists should integrate the legacies of
history into falsifiable theories of historical change. As such our
enterprise is both more modest and more ambitious than most. It is
more modest because we must focus on specific institutions and develop
frameworks for making comparisons across societies and over time. It
is more ambitious because at the end of the process we arrive at a
sharper understanding of the linkages between politics and economics
in China and Europe.
Ours is not the first attempt at such an analysis, in fact,
comparative economic history lies at the core of efforts to understand
why some places are prosperous and others poor. Obviously, Europe and
North America were the first places to experience modern economic
growth, and they have also provided the heart of the evidence upon
which models of development have been based. The relative dearth of
evidence on other parts of the globe has led comparative economic
history to proceeds in two steps. First scholars find some trait that
has been associated with success (e.g. representative government, the
nuclear family or Christianity); second, they seek to classify other
societies based on how close their institutions are to the favored
one. Scholars have proffered many features to explain either Britain
or Europe’s early success. These range from broadly cultural to more
specifically social, political and economic factors. Douglass C. North
has led the way in stressing the importance of institutions to
economic growth. Good institutions provide the rules and the sanctions
to encourage productive behavior. People who enjoy secure property
rights are more likely to engage in production and trade with others.
Thus Good government is crucial because only the state can provide
laws and courts to make and enforce contracts. These maxims work well
to give an account of how and why England succeeded economically in
the seventeenth and eighteenth centuries in ways that Spain or
Portugal did not. Variations across Europe in early modern economic
development line up quite well with the security of property rights
and effectiveness of law and courts in enforcing claims stated in
contracts (North 1981). Since England was the first industrial nation,
it makes apparent sense to consider the institutions found in England
during the eighteenth and nineteenth centuries to be important factors
explaining the onset of modern economic growth. England’s virtues
extend well beyond improvements to production and exchange. The logic
of private property was intimately tied to ideas of “liberties,” which
elites vocally defended. In England, propertied elites initially
gained a voice to negotiate with the king, culminating in the rise of
parliament. On the continent, elites also negotiated with their kings
and invoked similar political ideas even if the institutions giving
them voice were not as effective as those forged in England. The
political and economic institutions forged in early modern Europe were
thus intimately connected to economic outcomes after 1750.
The gains and pitfall of this approach are clearly in evidence in
North’s recent work with John Wallis and Barry Weingast, Violence and
Social Orders (2009). While they stress the importance of politics to
economic performance they generalize the linkages between political
and economic practices found in European and American history. They
identify a grand historical arc leading to modern societies
characterized by the replacement of polities with limited access by
polities with open access. Limited access orders (societies where
privileged elites limit the use of violence), have elites who capture
wealth and power. Open access orders, by contract allow everyone to
enjoy economic opportunities and political voice. In the open access
society political and economic competition prevails because the cost
of forming either political or economic organizations is small and
equal for everyone. At the heart of their analysis are institutions
which North has previously referred to as the “rules of the game”
(North 1990: 3-4). In this work, the authors explain, “Institutions
include formal rules, written laws, formal social conventions,
informal norms of behavior, and shared beliefs about the world, as
well as the means of enforcement.”(North et al 2009 p 15) The
trajectory of change from “limited access orders” to “open access
orders” is complex and contingent because it involves dramatic changes
to the political coalitions that insure civil peace and to the
structure of the economy. In particular, success depends on the
emergence of increasing numbers of economic and political
organizations that can realize peaceful economic and political
competition. Empirically European history exemplifies the process they
are reconstructing and within Europe, England figures most prominently
and positively.
China, our focus for comparison with Europe, presents a case far less
easy to fit into their framework. China’s recent experiences of
industrialization in the 1980s and early 1990s, for example, did not
depend very much on the formal institutions of property rights,
contracts, and third-party enforcement by the state, as Douglass
North’s approach would have predicted. Recently, Avner Greif has
rehabilitated informal institutions, showing that they play a critical
role in sustaining trade, and arguing that the social structures
behind these informal institutions can powerfully inflect the path of
economic change (Greif 2006). In this book, we suggest ways in which
quite different institutions can perform similar functions. In fact,
many of the arguments deployed in the past fall by the wayside as soon
as one realizes that neither China nor Europe is homogeneous. In fact
in most situations before Industrialization, Europeans and Chinese
individuals confronted a similar menu of institutions. The
distribution of what institutions were used responded to simple
economic logic. It was not until much later that the institutional
menus began to diverge. In the case of contracts, many of the
differences among Chinese and European institutions can be understood
in terms of the degree and substance of formality versus informality.
Chinese and Europeans deployed both formal and informal institutions;
what differed was their relative importance. When economic conditions
changed, both societies altered their relative reliance on formal
institutions, sometime they increased it, at other they reduced it.
That both societies responded to circumstances encourages us to
consider that some early modern Chinese and European economic
practices were different simply because of circumstances. Moreover in
this comparison, the degree of formality of transactions is not an
indicator of efficiency.
Putting institutions into spatial as well as temporal contexts also
matters to us. The national units that North, Wallis and Weingast
favor as their units of comparison are nation states that compete with
each other. Their focus is heavily on domestic politics and war is
little more than another form of public spending. This approach is
inadequate for Europe because international relations cannot be
separated from domestic politics. It is even more inadequate for China
because after 1000 CE, the Middle Kingdom was always larger
territorially and demographically than many European polities put
together. To this day much of what passes for international relations
in Europe is domestic politics in China.
The need to specify geographical units of analysis as part of the
research process (rather than simply relying on political boundaries
at a point in time) is well understood in social science, but most
often ignored for practical reasons. We cannot do so because
differences in spatial scale and in the resulting intensity of armed
conflict is central to our analysis of China and Europe. While the
division between domestic and international make sense for many
twentieth-century subjects, it is certainly extremely problematic when
making comparisons between China and Europe in the past. We therefore
re-evaluate conventional political contrasts of European fragmentation
versus Chinese unity to identify the advantages of a large domestic
unit to economic activity. In this comparative light, Europe’s
competitive political system was extremely costly. The costs that
accrued to Europe from fragmentation were quite visible because they
involved violent political strife; the advantages were unintended and
indeed unanticipated.
We do not propose an equally bold but alternative framework. Indeed,
it is history and its uncomfortable facts that force us to depart in
significant ways from the model proposed by North, Wallis, and
Weingast. Rather our claims are more limited because they are more
precise. It may well be that in considering change in Africa, South
America or Southeast Asia, a researcher can take the structure of
political institutions as given. If so, neither the lessons Chinese or
European history would export well. Even that negative finding would
be worth establishing. We suspect, however, that the study of long-run
economic change is informed everywhere encompassing ‘domestic’ and
‘international’ political competition.
Our finding that European national units are too limiting a spatial
focus brings us to a second major inspiration. Kenneth Pomeranz has
led the way in thinking carefully about the interaction between space
and economic history. In The Great Divergence (2006) Pomeranz draws on
frameworks that have long touted Europe’s geographic advantages both
in terms of intra-European trade and in terms of access to New World
resources (Jones 1981, Wrigley 1988, Allen 2009). In contrast China
did not have these spatial advantages and as others have emphasized it
also did not possess a political system that sought maritime expansion
(Wallerstein 1974). Yet his analysis also rests on a much deeper
understanding of China and thus offers the richest comparison of China
and Europe in the economic history literature.
Pomeranz’s argument that in terms of markets, consumption, and life
expectancy China appears strikingly similar to Europe well into the
eighteenth century extends a line of inquiry one of us began some
years ago (Wong 1997: 9-52). Pomeranz’s work has made very clear the
challenge of explaining how economies similar in so many ways during
the eighteenth century ended up diverging dramatically in the
nineteenth century. His explanation of economic divergence stresses
two sets of reasons for the break between Europe and China, both of
them based on environmental differences. First, the location of coal
in Europe closer to cities hungry for energy than was the case in
China gave the English in particular a critical advantage over the
Chinese when timber became scare. The argument expands on prior work
from Wrigley and has received substantial support from Allen (2009).
Second, the close proximity of the New World allowed Europeans to
avoid the difficulties attendant to increasingly scarce land.
Pomeranz’s study satisfies historians’ desire for a narrative account
of the past. At the same time it engages economists with arguments
they recognize (e.g. European economic successes depended on
labor-saving possibilities from abundant land and energy). His
explanation is temporally focused on critical changes that take place
in the late-eighteenth and early-nineteenth centuries.
We share with Pomeranz a desire to anchor our explanation of economic
divergence in time and place even as we search for more explanatory
mechanisms that go beyond the circumstantial (location of coal or
proximity of new and unexploited resources). Yet, while we find
European national units too limiting a spatial focus, we are not
arguing for a “world” perspective on early modern economic history. To
our minds, the most persuasive explanation for Europe’s
late-eighteenth and early-nineteenth-century transformations is best
provided by comparing the politics of economic change within China and
Europe in the centuries’ preceding their visible economic divergence.
When one extends the analysis back in time, political reasons for
different conditions in China and Europe come into sharp relief. These
differences initially favored China because the empire could and did
grow through Smithian principles of specialization and exchange. Yet
the same forces latter favored Europe as political fragmentation
increased the likelihood of capital using technological and
organizational innovations. It stresses comparisons between world
regions and the differences between them creating increasing
likelihoods of dramatic economic change taking place in Europe rather
than in China.
Like Pomeranz, we also avoid invoking cultural traits as putative
reasons for different economic outcomes in China and Europe. Unlike
him, we do observe important differences between China and Europe that
have been considered by some to reflect cultural traits peculiar to
one or the other world region. For instance, China’s
eighteenth-century construction of a vast granary system can be
considered as the emperor’s Confucian and paternalistic commitment to
the people’s subsistence security. While eighteenth-century China’s
language of paternalism was no doubt difficult to translate into
European languages, such difficulty did not prevent Enlightenment
thinkers from identifying it as enlightened despotism (Montesquieu).
European rulers also expressed paternalistic aspirations. What they
lacked were the capacities to implement their political desires while
at the same time pursuing war. European rulers faced political
challenges radically different from those confronting an agrarian
empire like China. What others have ended up thinking of as cultural
differences are in fact better understood, we argue, as products of
choices made in response to very different kinds of circumstances.
These different circumstances are not simply natural and geographical,
like Pomeranz’s access to coal or proximity to the New World, but
rather are produced socially and politically under diverse ecological
and environmental conditions across both China and Europe.
We are interested in explaining economic change in China and Europe
according to a common set of economic principles and in observing how
the political contexts influence outcomes. We seek to use economic
theory to explain variations within China and variations within Europe
as well as variations between them. Price theory figures prominently
in our explanation of economic change. Long before the visible
divergence of China and Europe after 1750, differences in the relative
factor prices in China and Europe set in motion the incentives to save
on labor and invest in capital that figure prominently in Pomeranz’s
account as well as in other scholarship on European economic growth.
To explain these differences in factor prices we will stress
conditions that are the outcomes, we will argue, of more basic
differences in the spatial scale of polities in China and Europe. In
this analysis we parallel Robert Allen’s recent work on the progress
of industrialization in England (2009). Indeed, Allen puts special
emphasis on relatively high wages and low fuel costs in explaining why
the technologies we associate with industrialization were developed
and deployed in England. Yet his analysis cannot explain why Europe
(and not just England) developed a cadre of skilled workers and
techniques that would blossom most fully in Britain. Nor can it
explain why wages were high in Britain after 1650 relative to the
continent without recourse to politics. Moreover an analysis of
politics cannot be restricted to comparing England to France or China.
It must start by examining differences between two units of similar
scale that evolved separately: Europe and China.
Like Pomeranz we seek to get at the roots of the divergence, but we
believe our approach integrates more social science and history. We do
not, however, aim to offer a close historical account of the many
particular changes made possible by those differences between China
and Europe Pomeranz emphasizes. What we lose by presenting a less full
history, we gain in temporal reach. For it is our claim that in 2050
when China will look so much more like Europe economically than it
does today, the factors we stress—institutions and political
scale—should continue to help guide our exploration of the way
polities and economies evolve. Whereas the importance of endowments
has faded as transport costs have collapsed. After all, China’s
coastal provinces have been able to boom despite their distance from
coal fields and the New World’s natural bounties—the interactions
between politics and economics remain fundamental to explaining
economic changes in the future.
We believe that China and Europe were set upon their separate paths
long before 1750 when energy or distant land resources became more
important. These factors no doubt contributed to the path of economic
changes in Europe and may well have exacerbated the relative
performance of economies like that of England in the nineteenth
century. Yet, they were neither sufficient nor necessary for the China
and Europe’s economies to diverge. Similarly, economic change in China
and Europe was not driven mainly by differences in people’s
intentions, abilities, or personal circumstances (however much these
factors can matter at the individual or local level). From the
perspective of what individuals choose, we think some of the most
important factors influencing different likelihoods for economic
change in the early modern era were unintended consequences of actions
taken for reasons largely unrelated to improving the economy. Finally,
we reject the idea that some narrow institutional differences between
China and Europe were sufficient to change likelihoods of economic
success, because, as we show, different institutions can work as near
substitutes in different circumstances. To observe that institutions
are different doesn’t necessarily mean that one set is always better
than another. We will argue that political factors made it
increasingly likely that parts of Europe rather than any parts of
China would make the transition to modern economic growth by the late
eighteenth century, irrespective of their relations to the New World
or the location of their coal deposits.
As historians, we reject the myth of a contrast between European
growth and Chinese stagnation in the centuries preceding the very
visible nineteenth-century divergence. The evidence is clear: China
did not stagnate economically until the nineteenth century, and even
then not all parts of the empire were unable to grow. Analytically,
slower growth is fundamentally different from stagnation with all its
Malthusian ghosts. Before 1700, similar dynamics of Smithian growth
worked through different political and economic institutions in China
and Europe. The key reasons for economic divergence were political and
these increased the likelihood that modern economic development would
take place in Europe before China. European advantages were unintended
consequences of political differences with China.
To make our case we proceed first with some history in Chapter One to
highlight both the striking differences in political scale in Europe
and China and, some disturbing facts for anyone who wants to take
these as given. Indeed, we are well aware that China experienced long
periods of fragmentation, and that the entity known as the Roman
Empire endured for centuries—even if we exclude its Byzantine temporal
extension. What then are the consequences of political scale?
To answer this question we develop a sequence of frameworks. Chapter
Two considers the old Malthusian workhorse of household structure and
demography as the possible source of significant institutional
differences that could help us account for economic divergence.
Kinship relations and population dynamics are implausible sources for
divergence. Chapter Three looks at the institutions enabling economic
transactions in China and Europe between the mid-fourteenth and
mid-eighteenth centuries. We find that although the two regions were
clearly not alike, their dissimilarities stem from political scale and
seem unlikely to have caused economic divergence. Chapter Four takes
us to the realm of manufacturing or craft production where we find
that the urban location of considerable European manufacturing and its
more frequent rural location in China is significant, but not exactly
in the ways conventionally argued and for reasons that others have not
clearly explained. We move in Chapter Five to consider how production
and trade are financed and in ways similar to Chapters Two and Three
discover institutional differences, but not ones we consider causally
crucial to have set of China and Europe of separate paths. In Chapter
Six we move to public finance and here we find differences that cannot
be accounted for by conventional contrasts of Chinese and European
states in the early modern and modern eras. The differences we
discover affect economic change in ways contrary to what previous
scholarship has suggested, though the impact on overall likelihoods of
economic growth are limited. The variation in public finance
institutions, so clearly tied to the agendas for rule in an empire
versus those prevalent among a set of smaller competing polities,
complete our analytical revolution. Europe succeeded despite rather
than because of political competition.
In Chapter Seven we return to the history introduced in Chapter One
and offer our interpretation of why the equilibrium size of polities
was so different for so long in China and Europe. Having taken
political scale as given and shown its importance in Chapters Two
through Six, we consider in Chapter Seven some reasons for the
differences of spatial scale for polities in China and Europe. We show
that the politics of economic change in China and Europe are quite
different from each other and that, as early as the end of the first
millennium BCE, enter into self-reinforcing patterns.
We put forward general arguments and exemplify them with Chinese and
European data. Not only are our economic arguments intended to account
for variations within as well as between China and Europe, they should
prove confirmable or falsifiable by data from other places outside
both China and Europe. Similarly, our arguments about the significance
of political scale are largely applied to explain one set of outcomes
in China and Europe (economic divergence), but in the conclusion we
also suggest they remain relevant to more recent times. Our method of
comparison is thus intended to be exemplary in the simple sense of
being an illustration of a way to approach explanation of economic
similarities and differences in the world of the past that can also
apply to predicting future changes. Our goal is to identify causal
mechanisms that we know work across varied particular conditions, like
those suggested by price theory, and apply them to conditions that we
think are best explained by considerations of politics. The exercise
is not intrinsically historical or limited to explaining what we
already know to have taken place. Indeed effective explanations of
what has happened in the past can help us anticipate future
possibilities because so many of the social processes at work today
have historical roots and antecedents.
Our method of analysis identifies what we find to be persistent myths
about China and Europe. It also allows us to reject accounts of what
made Europe so special or its growth so miraculous. We pursue
explanations of economic change that can account for observable
behaviors in China and Europe and we invite readers to assess the
persuasiveness of our analysis and to extend our approach to other
times and places. We will count ourselves fortunate if we engage you
seriously enough to evaluate our approach and compare its advantages
and limitations with those of other studies of economic change. We
will measure our success by subsequent efforts made to amass more
evidence and formulate research that confirms, qualifies or undermine
our explanations.
Chapter 1
Space and Politics
Introduction
A thousand years ago China was a vast empire. So it was a hundred
years ago. A thousand years ago Europe was politically fragmented. It
was still so a hundred years ago. These contrasts might suggest that
massive differences in the scale of polities are constants in the
histories of these regions. From such a perspective, it would be easy
for us to simply take the divergent political structures as a given
and put our energies into tracing out their consequences for economic
change. We could then appeal to either geographic determinants or
cultural constants for the early and persistent Chinese success at
creating a large integrated political space (and for Europeans, the
failure to do the same). We have decided, however, to avoid such an
approach because we are aware that this basic political contrast
between China and Europe was neither constant nor necessary. A bit
more than two thousand years ago both China and Europe were both
large-scale empires, but five centuries later both were fragmented.
Over the next five centuries there were several long episodes of
fragmentation in China and repeated attempts to put the Roman Empire
back together in Europe. Given this more complicated history, we seek
in this chapter to explain how it came to be that by 1300 the European
political equilibrium involved spatial fragmentation while the Chinese
political equilibrium featured spatial integration. Understanding this
basic contrast is critical for our subsequent analysis of the
political economies of growth at both ends of Eurasia.
An analysis of the history of China and Europe between 1300 and 1850
is necessary to begin exploring the evolution each region’s political
economy and how it mattered for economic growth. Let us briefly
consider a well-known example of the economic consequences of
political scale that we will come back to in Chapter 3. Empires afford
greater opportunities for large markets and the kind of growth
processes highlighted by Adam Smith than politically fragmented
regions. There, at the very least, war and customs barriers will
surely impede trade. The opportunities for Smithian growth in empires
can of course be undermined when rulers substitute themselves for the
market (in particular using taxes to secure grain for politically
sensitive locations like Rome, Beijing or Istanbul) or interfere with
the labor market (as Russia’s serfdom or Spain’s American encomiendas),
or that of land (as with the Ottomans’ timar system for funding the
army). But such intrusions into the market are not the peculiar
proclivity of empires; other polities did much the same. Instead, we
must accept the importance of the interactions between political
structures and historical circumstances in shaping economic
institutions.
For our purposes of comparing China and Europe, we set aside debates
about the definition of “empire.” We will call empires those polities
in Europe or China where a central ruler exercised effective authority
over a large fraction of a contiguous region. Clearly this definition
is not intended to be general or prescriptive. Any reader familiar
with the Ottoman, Hapsburg or colonial European empires will know that
these realms were neither geographically compact nor blessed with
great political capacities. Furthermore unlike the Roman Empire where,
after Caracalla, all free men were citizens or the Chinese empire
where the Han were, by far, the dominant ethnic group, most empires
have been ruled by a minority population that severely restricted the
political rights of other groups. We choose this definition of empire
simply because it encapsulates the key contrasts between Europe and
China analyzed in this book. Our empire, therefore, is neither an
ideal type nor a general phenomenon, it is just a practical
appellation for spatially large polities in contrast to far smaller
ones.
For most of this book we will trace out the direct and indirect impact
of differences in the spatial scale of polities on economic
development. We will argue that differences in political scale are
critical to understanding the economic divergence between China and
Europe but we will not base these conclusions on the specific details
of imperial political structure. Instead we take the contrasting
spatial scales of Chinese and European polities as key factors that
both let and led rulers in these regions to develop different
political priorities and policies. These policies in turn shaped the
paths of economic change. Policies in other empires did not closely
parallel those in China any more than policies in other fragmented
states, like those of Africa or Southeast Asia, reproduced the
practices forged in Europe. Political histories embody much that is
historically particular and spatially specific. For our purposes of
comparing Chinese and European economies, we need attend only to the
political histories of these two regions of the world.
In this chapter we seek to understand why, despite the existence of
successful empires in both regions two thousand years ago, the two
political systems diverged to such an extent, that five hundred years
ago, fragmentation was as stable in Europe as consolidation was in
China. In this largely narrative chapter we focus on early processes
of empire formation. We also consider the structures of the empires
and the challenges they faced. We then examine the processes that
after 200 AD led to the permanent breakup of the Roman Empire, and the
repeated reconstruction of the Chinese empire. By the early centuries
of the second millennium (Yuan and Ming dynasties) a political economy
had emerged in China that favored the maintenance of a large and
integrated political space. Henceforth dynastic transitions were
painful but brief. In Europe a political economy had emerged that both
overcame the extreme political fragmentation and instability of the
early Middle Ages but made the reunification of Europe extremely
unlikely.
Rather than seeing the repeated successes of Chinese rulers at
maintaining or reconstituting their empire as a direct and simple
consequence of the spread of Han culture, we show how Chinese imperial
rulers and their official elites learned from their mistakes in order
to become more successful at promising and delivering internal order
and welfare enhancing projects. We also argue that in many ways,
Europe’s political elites, from the rise of the Roman republic to the
fall of Constantinople, were striving to establish the kind of
prosperity that was achieved in China but they failed. Continued
political strife in Europe ensured that rulers who focused simply on
public order, access to markets, and infrastructure development would
not have survived. Instead, European rulers focused on gathering the
resources they needed for war. In the process of raising taxes to pay
for warfare, Europeans rulers enshrined many specific concessions to
local groups in a plethora of charters. These, in turn, proved to be
an enduring brake on the spatial scale of political consolidation in
Europe.
As noted above, one could attribute the existence of empire in China
to a variety of extremely long standing cultural attitudes or even to
endowments. One could do the same for fragmentation in Europe. But
such convenient explanations are belied by the fact that empires arise
in a variety of settings across world history and well beyond China.
Similarly political fragmentation is not Europe’s exclusive attribute.
For economists, and many other social scientists as well, the
evolution of the size of polities results from a competition between
the heterogeneity of demand for public services and economies of scale
in the delivery of these services. In our own contemporary world,
scholars focus heavily on domestic services such as welfare,
education, and infrastructure (Alesina and Spolaore 2003); in
historical settings one must also include the military. The breakdown
of the Roman Empire can thus be seen as the result of a collapse in
the returns to maintaining a large scale military, as well as an
increase in heterogeneity of demand for public services due to the
influx of populations that had hitherto lived outside the empire.
Similarly, the effectiveness of the Great Wall in containing nomadic
raiders, as well as the overwhelming demographic size of Han
populations in the empire, can be seen as making a large-scale polity
easier to maintain in China. Yet this simple contrast, like many
striking historical differences, requires closer analysis. The Great
Wall had its parallels under the Roman Empire; such walls were just as
much of a European innovation as the military techniques that allowed
Germanic tribes to defeat the Roman legions or the rise of siege
artillery that eliminated petty lords in much of Western Europe. A
parallel contrast between China and Europe regarding the diversity of
cultural identities in Europe and the dominance of Han identity in
China requires us to consider more closely the hows and whys of social
processes leading Chinese individuals to adhere to a common Han
identity rather than privilege some more local identities.
1.1 From early empire to the Mongol invasions: China’s memory
China’s empire has very old roots, but sustaining and increasing the
scale of China as a political entity required overcoming a number of
challenges. Central to the establishment of the empire was the ability
of the emperor to deploy overwhelming force over long periods of time
because, as we all know, empires are generally created through
military success. Yet, sustaining the empire was a far more subtle
endeavor. Consider the short lived Qin dynasty. It achieved an
imperial unification in 221 BCE, yet Qin rulers fell only fourteen
years after proclaiming their dynasty. As the standard accounts
suggest, the Qin had a strategy of conquest but no imperial strategy
of rule and they fell to popular revolts prompted by their harsh
demands for resources and labor from the common people (Bodde 1999).
Like the Qin, the next dynasty (Han) prevailed in warfare among rivals
to assume imperial control, but unlike their predecessors they
developed policies of less harsh rule that allowed them to survive for
some four centuries. The government opened new agricultural lands and
maintained irrigation works that made these lands more productive. The
typical rural settlement that the government sought to promote was an
unfortified village with some one hundred households each owning its
own small amount of land that allowed them to meet their material
needs and pay taxes to the government. A society of prosperous small
holders formed a persistent ideal for Chinese rulers. (Nishijima 1999)
Benevolent rule was not a panacea, since not all Chinese were content
to settle with the Han dynasty’s rule. The emperors faced such serious
challenges from elites that they lost power for fifteen years between
A.D. 9 and A.D. 23 to a competitor whose failure to sustain his rule
leaves him with the label of “usurper” in history books. Although the
Han ruling family regained power, the dynasty was unable to extend its
authority to the local level. On the contrary, powerful landed elites
controlled small areas and over time increased exactions from their
peasants, leading to social conflicts which erupted into rebellion.
Neither these magnates nor the Han dynasty could suppress the peasant
uprisings. The social unrest proved fatal for the empire. Indeed, Cao
Cao, the general who put down the rebels, took the opportunity to
establish his own authority over a third of the empire. This potential
founder of a new dynasty could go no further, however, in
reconstituting the empire. After he died in 220, his son did force the
last Han dynasty emperor to abdicate but Cao Cao’s son could not
extend his territorial control beyond the third of the empire he had
inherited from his father. Three centuries of political division
followed (Bielenstein 1999).
Once again the politics of violence are central to understanding this
long period of fragmentation. Its causes include pressure from the
Xiongnu—a steppe people who began to organize themselves as powerful
foes long before the collapse of the Han dynasty. The Xiongnu had
earlier responded to Han military efforts to push them further north
and west with their own counteroffensive. Later, and to the Han
dynasty’s dismay, the balance of military strength shifted and the
steppe people gained more territory. Han leaders built walls in an
effort to protect their initial territorial gains and then to protect
themselves from Xiongnu advances. But nothing the Han dynasty did
prevented the Xiongnu and other steppe peoples from becoming important
military actors in the competition for control of north China after
the fall of the Han dynasty in 220 (Di Cosmo 2002).
For more than three centuries after 220, no set of leaders emerged
either from within the Han’s former territories or from the steppe who
could build a successor empire. Political fragmentation in north China
proceeded to a very local level in some places where militias were
formed for self defense and strongmen created their own small realms;
other parts of the north were under the control of larger and stronger
military rulers. In south China military rulers ruled small kingdoms
and faced two kinds of challenges. Domestically, they confronted
powerful families who controlled large amounts of property; at the
same time they competed not only with each other but also with the
much stronger northern regimes, which had richer resources bases and
strong military traditions. Where northern regimes had steppe
warriors, southern regimes recruited tribal minorities, convicts and
vagrants (Graff 2002). Southern rulers also aspired to civilian
political ideals that had been created under the Han dynasty; these
principles helped rule an empire but offered little guidance to those
seeking to build a new empire. By circumstance, this is precisely the
period when the Roman Empire began to break apart. There, as in China,
the same goals of achieving local security and reconstituting empire
competed with each other.
Between the early third century and late sixth century there was no
unified empire on the Chinese mainland. In 400 AD, had people been
able to look across Eurasia they would likely have doubted an empire
would form again either in Europe or in China. Although ethnic Han
Chinese dominated demographically, in North China they intermingled
with a diverse array of steppe peoples, some of whom spoke ancient
forms of Tibetan or Mongolian, while others spoke Turkic languages.
These populations were organized in a series of culturally mixed small
kingdoms. Their rulers were equally influenced by the steppe people’s
martial traditions and Confucian visions of imperial order. In
particular, the military practices of steppe armies became the model
for rulers of the northern portions of the former Han Empire. Yet
whatever the differences initially separating them, they were all
exposed to Chinese political philosophy that affirmed the norm of
empire and made recovery of empire the common goal. Although the
rulers of the northern kingdoms were of diverse origins, they adopted
the history of the Han Empire as a model for the world they had joined
and wished to sustain (Tonami and Takeda 1997:41-160).
To reform the empire required combining these imperial visions with
sufficient military might and it was by no means a foregone conclusion
that any set of leaders would emerge who could conquer all their
competitors. Yet such a leader did emerge in the North. Yang Jian
built the army able to embark successfully on a march of conquest,
first against other leaders in the north and then against southern
regimes unable to mount successful defenses; his successes culminated
in the formation of the Sui dynasty in 589. Yet just as in the case of
the Qin, a strategy for conquest was not a strategy for rule. Sui
leaders demanded too much of their subjects who funded the building of
the Grand Canal connecting the rice-rich south to the governing north
and also paid for military adventures that took Sui armies
unsuccessfully into the Korean peninsula (Graff 2002: ch 7).
The Sui rulers were replaced by a new set of leaders who established
their Tang dynasty in 618 and proceeded to rule for nearly three
centuries. The Tang focused on strategies for keeping the empire
together. In doing so they emulated and enlarged upon the policies of
the Han dynasty. For roughly half their period of rule, Tang emperors
were successful in expanding the empire’s borders to the west at the
same time as they put in place measures to stabilize the living
conditions of their peasant subjects. At the same time, they extracted
their revenues directly from the peasantry. The Tang worked to balance
conformity with Confucian principles, and thus uniformity within the
empire, with more practical consideration of the ethnic and cultural
diversity of their subjects. The Tang capital of Changan became home
to a diverse population of Han Chinese who intermarried with Di,
Qiang, Tuoba and other peoples, each of whom had distinctive
languages, foods and clothing. Cultures and blood lines mixed to
create a range of lifestyles that together represented a cosmopolitan
empire. In this setting, lifestyles and cultural practices were chosen
by individuals rather than imposed by genealogy. Imperial
reconstruction was aided by the absence of any European-like period of
the Dark Ages that followed their “barbarian” invasions. The arts,
especially inspired by Buddhist influences from India and Central
Asia, flourished. Scholars sustained, without major interruptions,
their classical traditions, at the heart of which were the Confucian
ideas and institutions that could inspire and form a bureaucratic
state. Many of the principal organs of the Tang central government,
such as the six boards, provided basic and important models for later
Chinese dynasties. (Adshead 2004)
The Tang dynasty shared the territorial ambitions of previous
dynasties. Its rulers extended and opened the country’s borders into
Central Asia. Like the Han dynasty, the Tang forged a presence to the
west in several of the oasis communities along the Silk Road. These
connections ensured that a diverse range of cultural influences
originating in distant places would continue to enrich what later
became considered typical of Chinese culture, including the poetry of
Li Bo and the tri-colored glazes of Tang porcelains. An open and
expanding empire not only welcomed new cultural influences but also
became ever more vulnerable to military threats, including those posed
by some of the very troops expected to keep the empire safe from
outsiders as the spatial scale of their responsibilities grew larger.
Some of the military forces employed by the Tang state to maintain
peace were themselves recruited from Central Asia and were descendants
of the ethnically and culturally mixed groups that formed beyond the
northern frontiers of the empire. Anxieties among some court officials
regarding the power of these military commanders led the general An
Lushan to strike at the capital because he feared that the imperial
court might attempt to limit or even undermine his power.
The Tang state was forced to seek strategic alliances with other
steppe region semi-nomadic military forces to defeat An Lushan’s
rebellion of 755. As a result the Tang military had to withdraw from
Central Asia and accept a far diminished empire. Then in the late
ninth century domestic unrest among impoverished peasants and powerful
local lords led to further troubles for the dynasty. Finally, in the
early tenth century, the capital was captured by an enemy general. The
fall of Chang’an marked the beginning of another period of political
fragmentation in which rival forces established smaller kingdoms. For
a period of nearly a millennium, from the Han to the Tang, Chinese
dynasties found the balance between external expansion and internal
cohesion difficult to maintain (Graff 2002: ch 10, ch 11). As a result
the empire was repeatedly overrun and fragmented. Nevertheless, the
empire was recreated later because subsequent rulers and their
officials could draw upon a growing repertoire of earlier ideas and
institutions to which their own innovations offered their successors
even more possibilities.
The Song dynasty, established by the general who came to power in the
old Tang capital in 960, was no different. Its founder not only took
over the remnants of his predecessors but also reconquered other small
kingdoms that had emerged in what had previously been the Tang Empire.
Nevertheless, the Song dynasty ruled a much smaller realm. It was in
this smaller realm that a set of key administrative innovations
occurred. These political processes may well have been spurred by the
more rapid pace of social and economic change that was occurring at
the same time (expansion of urban centers, small-scale tenant farmers
producing for the market, improved transportation technologies, and
new commercial institutions and merchant networks). Whatever their
cause, Song political innovations should be seen as fundamentally new
techniques of rule that reduced the transaction costs of internal
administration. In particular, the dynasty created a civil service
bureaucracy for which many officials were recruited on the basis of
passing examinations; bureaucratic sophistication and specialization
enlarged the government’s capacities to mobilize resources and order
local societies. However, domestic successes were undermined by a
renewed vulnerability to states formed along the northern border by
groups who combined military prowess with some of the bureaucratic
institutions of rule developed within the empire. This military
weakness ended up forcing Song rulers south where they maintained a
state that became one of several states on the Chinese mainland.
Although the Song rulers’ reach was not as extensive as that of some
the earlier empire, their retreat south was fortuitous as it also
reinforced the dynasty’s close connections to the emerging centers of
economic and social change (Ihara and Umemura 1997).
Sitting in the Southern Song capital of Hangzhou in 1200 a
well-informed observer of the dynasty’s domestic conditions and
foreign situation may well have been struck by the growing wealth of
the country’s cities and its increasingly precarious military
situation along its northern borders where a number of states,
especially if they joined forces, could threaten the Song government.
Without such a coalition or consensus among northern states, our
observer, if he could think beyond the framework of Chinese dynastic
history, might have imagined the possibilities of a multi-state system
emerging with militarily strong, but commercially poor, states in the
north and a wealthy, but militarily limited, state in the south. In
other words, the persistence of empire across the scale of space that
had formed during the Han and Tang empires need not have been
replicated thereafter. From the mid-eighth century when An Lushan’s
rebellion ended effective central rule under the Tang until the Mongol
conquest of the Chinese mainland in the thirteenth century, there was
no unified empire. Our imaginary observer, if particularly astute,
might also note that even with a multi-state system consisting of a
few large realms (Rossabi 1983), the political equilibrium on the
Chinese mainland need not have reached the level of spatial
fragmentation found in Europe. After the Mongol conquest neither an
imaginary observer nor subsequent historians were likely to recall the
possibilities for a multi-state equilibrium becoming more permanent on
the Chinese mainland.
Once again, the consolidation of the empire required a superior
military force that could drive out its competitors, destroy them, or
incorporate them. The Mongols did all three as part of an even larger
enterprise that, in the thirteenth century, absorbed not only China
but also much of central Asia reaching westward towards the Ottoman
Empire. The Mongols’ conquest of large parts of Asia created the
world’s largest empire. Their territories were so vast that it was
impossible for a single leader to rule them effectively. In 1251 the
empire was divided into four separate realms centered in southern
Russia, Persia, the Mongol homeland, and China. The last of these was
by far the wealthiest and most populous. From the vantage point of
Chinese history, it is difficult to exaggerate the importance of the
Mongol conquest; without the Mongols North and South China (like the
eastern and western Roman Empires) might well have gone their own way.
The Mongols simply destroyed all other would-be military competitors.
When their far-flung empire fell apart, and the Mongols in China
retreated to the pastures of Mongolia in the face of tremendous
domestic unrest, a native Han Chinese dynasty could take over and
establish its rule over the sedentary portion of their empire without
facing serious territorial threats from strong “barbarian” forces in
the north. (CHC vol. 6 chs 4-9)
Clearly, the history of empire on the Chinese mainland over the first
fifteen hundred years of imperial rule has a distinctive military
rhythm. Empires formed and fell because of military offensives often
coming from poorer, external foes. But we can also see a pattern of
internal processes that made the reorganization and persistence of the
empire more likely. These involve the successful spread of Han culture
among populations that were initially radically different in terms of
ethnicity, language, and social practices. These processes also
involved the progressive creation of an effective structure of
imperial administration. Thus, some core elements of the mature
Chinese empire have very old roots. But until the tenth century the
empire withered away several times.
The persistence and growth of the Chinese empire, and its equally
recurring collapse leads to some reconsiderations of the political
economy of scale. Rulers were regularly tempted to expand their
dominions in ways that were unsustainable. Furthermore they were not
always able to adjust their political organization to respond to new
challenges (domestic unrest or foreign threats). Time and again we
observe changes in internal governance or in the size of the realm
that led to serious problems of governance and even the collapse of
the dynasty. Over the long run, however, Chinese dynasties proved
quite capable of learning elements of rule that made the empire more
successful. The history of China before 1350 (from the Qin through the
Mongols) can in fact be seen as a long apprenticeship in the
strategies of internal rule.
Later dynasties, the Ming and even more so the Qing, capitalized on
their predecessors’ experiences. Successful rule involved finding
balance with respect to the internal governance of the polity and its
external relations. Domestically, emperors recognized the value of
uniformity within the realm (which eased the flow of information) and
the return to letting localities choose more specific institutions
(which allowed innovation and specialization and reduced
administrative expense). They also had to choose a level of requests
for tax revenues from their subjects that was compatible with the
services their officials provided. In each of these cases failure to
maintain balance led to revolts and lower tax collection. Early on,
rulers and their officials seem to have miscalculated repeatedly. In
terms of international relations, imperial failures shine a bright
light on the importance of balance. The Sui collapsed because their
excessive appetite for territory brought about a reaction they could
not control. Mongol rule of the Chinese mainland lasted less than a
century; they viewed the people of the northern and southern halves of
the empire differently and ruled them in institutionally different
ways. What they had conquered they could neither transform nor rule
for an extended period of time. Like their predecessors, they did
promote conditions conducive to gains from trade and supported local
governments that provided social goods inspired by Confucian ideas
about good governance.
1.2 From Rome to Charles V: Europe skirts anarchy
From China’s history one might well be tempted to build a theory of
successful empire based on military innovation and an ideology of rule
that equated a ruler’s success with his benevolent treatment of his
subjects. The history of Europe suggests that these are far from
sufficient if empires are to endure over the long run. Indeed, the
Roman Empire was built on a military technology that vanquished foe
after foe for half a millennium. Its cultural practices spread
throughout the Mediterranean world; and its rulers espoused views of
administration that are not dramatically different from those inspired
by Confucianism. But after 200 AD the empire entered into a slow,
violent, and inexorable decline. As we discuss below, and others
before us have noted (Scheidel 2009; Potter 2004: 530), the Roman
Empire shared many similarities in its rise, expansion and fall with
the contemporaneous Qin-Han empire. In a wider comparison of empires
world-wide, the fall of the European Empire may be or may not be
exceptional. What strikes us is that every attempt to put the Roman
Empire back together was an utter failure. If we have only the Qin-Han
and Roman cases to consider, it is difficult to take one as the norm
against which to judge the other unusual. To expand upon our two
cases, with attention to the kinds of concerns we raise in subsequent
chapters, will remain a task for future work by other scholars. For
our purposes, as we have already suggested, we wish simply to explain
how and why the equilibrium spatial scale of Chinese and European
polities ended up being so different.
Had a subject of the Song dynasty found himself visiting Europe in the
tenth century, he would likely have been shocked by the parochial
nature of polities and statecraft. While some princes could claim to
rule over an area as vast as a Chinese province, few could exercise
the same authority as a Chinese ruler over more than a fraction of
their territories. Their powers were hemmed in by what Stephan Epstein
has aptly called “freedoms,” a host of particularistic privileges that
limited the prince’s capacity to tax, to regulate the economy and to
provide public goods (Epstein 2000). The recipients of these freedoms,
be they elites or commoners, peasants or urban dwellers, stood ready
to revolt should the prince attempt to gain more power. Hence, not
only did monarchs face the natural consequences of fragmented
polities, namely the threat of conquest, they also had to meet very
serious internal limits to their authority. By Chinese standards,
European monarchs were henpecked by their subjects. By almost any
standard, the rise of nation states in Europe is nothing short of a
miracle.
Europe had not always been so fractious, and had Chinese travelers
managed to make to it Rome around 100 AD, they would have found a much
more familiar polity. Like the Chinese empire, the Roman Empire was
born from the fire of war: from Hannibal’s invasion of Italy in 218
BCE to Varus’s defeat in 9 AD, the Roman army was only dealt minor
setbacks. While Varus’ loss of three legions in Germany was shameful,
it had limited consequences for the Empire, and the westward movement
of tribes in northwestern Europe was contained for another 400 years.
Expansion continued in the East, until Trajan’s army found itself on
the banks of the Tigris. 116 AD marked the end of conquest, not
because the army had found too strong a foe, but because Persia was
simply too distant from Rome to keep.
Like China, Rome took its imperial responsibilities seriously. The
first was keeping the peace. While until 116 AD the empire grew,
effectively pushing its foes further from Rome, after that time an
effective policy of containment prevailed (Goodman 1997 ch 7). As in
China, walls were built. By the standard of the Great Wall, Hadrian’s
stone barrier across Scotland was short. A much longer wooden palisade
was built across Germany and parts of central Europe. Legions were
stationed along the border and, by 150 AD, one of the Emperor’s most
important responsibilities was to secure the revenues to pay the
troops—failure at that task easily marked him for death. The other key
responsibility was the provision of public goods. While historians
have emphasized the Emperor’s lavish spending on ‘bread and circuses’
in Rome and later in Constantinople, one should not forget that the
cost of these activities is likely to have been quite small relative
to the investments in useful infrastructure. Indeed, the political
structure produced massive private and public expenses for
infrastructure that included roads as well urban amenities like paved
streets, arenas, theaters, temples, and water works. Such investments
were particularly noticeable in the western half of the empire because
these provinces had been relatively less urbanized prior to conquest
(Goodman 1997 part IV). Although many of these costs were borne for by
elites, rather than paid for with tax revenues, these ‘gifts’ as Paul
Veyne tells us were a key element of elite political control (Veyne
1978).
Like the Han dynasty, Roman emperors promoted bureaucratic integration
and a common set of cultural beliefs for elites. All around the
Mediterranean and throughout Western Europe, provinces saw cities
mushroom with their triumphal arches, arenas, waterworks, and similar
administrative structures. Not only were the men who lived in these
cities considered citizens of their home towns, soon enough they were
citizens of Rome. In fact, by the time the western Empire collapsed,
all free men in the empire were citizens as were many of the leaders
of the ‘barbarian’ invaders. One need but tour the remarkable
archaeological remains that survive from England to North Africa and
from Spain to Turkey to get a sense of the scale of expenditures that
went into forging a common identity for the elites of the Roman
Empire. At the time of Trajan and Hadrian, the empire was prosperous,
powerful relative to its neighbors, and culturally successful since
its diverse populations were adopting Roman ways. In short one could
easily surmise that the Roman Empire was following a course parallel
to that of China around the shores of the Mediterranean. By the second
century AD, the elites of the empire were drawn from all over the
Mediterranean basin and the emperor could and did dispatch them to any
of the provinces making up his domains (Potter 2004 ch 2).
Success was not very long lasting. By the reign of Marcus Aurelius the
empire was on the defensive. Over the next century, the demands of
military operations pushed emperors to divide the empire into western
and eastern halves. Although Constantine would reunite them to some
degree in 324, he would also move the capital of the empire away from
Rome which made reunification limited in its impact. By the time of
his reign, war demanded individuals with considerable authority be in
command of large armies both in Asia and in Europe. Because the
Emperor could personally attend to only one of these two areas, he had
to find someone else to leader wherever he was absent. A successful
general in the other part of the empire was a natural rival. Less than
a century later Rome would be sacked (410AD). There is no obvious date
for the end of the Roman Empire; its western half ended in 480 but its
eastern half would endure for another millennium. Over that time the
territories of the Eastern Empire were slowly but surely incorporated
into the Ottoman Empire, but this new political entity proved unable
to push into Europe north of the Danube or west of the Alps
The Roman Empire, like its Chinese counterpart, faced many
interrelated problems. That it endured is a sign that it could
overcome them at least for a time. From a modern perspective two sets
of difficulties stand out: first was the political instability of a
regime with no set system of succession; second were the continued
problems with the peoples beyond the borders of the empire. To begin
with, a Roman emperor was foremost a military leader. This was
particularly so because Cesar and Augustus’ claim to the throne came
from their military prowess. Not surprisingly, the legions and the
Praetorian Guard had much to do with the selection of emperors. Few
emperors died peacefully. Most seem to have met their fate at the
hands of angry soldiers or as a result of an internal challenge from a
relative or a general; some later emperors would die in battle.
Succession contests were further heightened because there was no rule
that required a single emperor, nor was there a rule that allowed the
army as a whole to make a decision as to whom should be its supreme
leader. Instead, as early as Galba (69 AD), troops in one region could
proclaim an emperor and if successful, either in intimidating the
sitting emperor or in battle, see their choice rise to top of the
hierarchy (Potter 2004 ch 3). The convulsions that marked the death of
Nero and Commodus, as well as the longer crisis of the third century,
were all internal struggles over who should lead the empire. Yet
despite this apparent fatal flaw the empire managed to survive many
successions.
The second problem the empire faced came from military conflicts with
neighboring peoples. These varied in intensity and structure. Along
the eastern border the Romans faced organized polities. In the first
two centuries following Augustus, Rome’s eastern neighbors were of
limited importance; it was Rome that chose where to mark the borders.
The legions encountered only limited opposition. Yet, by the reign of
Caracalla, the Parthian kingdom could muster a powerful army. It would
best the Romans in battle in 217; then, the rise of a new dynasty in
Persia would lead to further battles resulting in the capture of the
Emperor Valerian (259) (Potter 2004: 254-6). Although the conflict in
the east was expensive and protracted, like the conflicts over
succession, it was a threat that could be contained. In fact, the east
remained the more valuable and safer part of the Empire even after
Valerian’s defeat.
It was another external threat—from the North—that eventually proved
fatal. Semi-nomadic populations, living on the northern edges of the
empire, from the Black Sea to the North Sea grew in military strength
over the course of several centuries. Despite the defeat of Varus (AD
9), Rome was able to maintain the advantage over these peoples until a
major invasion of Italy in 259; the final blow came a century later
after the defeat of Valens at Adrianople in 378. Yet the empire did
not collapse. Like their Chinese colleagues, Roman emperors tried to
co-opt some of the nomadic populations. By 270 Aurelian and his
successors regularly negotiated with Germanic tribes so as to turn
enough of them into allies as a means to pacify the frontier. Yet that
proved insufficient, and there were attempts to alter the empire’s
political structure to both meet different armies’ needs for imperial
leaders and avoid the civil strife of contested leadership. Commanders
were needed both in the West and the East, and under Diocletian a
remarkable political experiment was attempted: the Tetrarchy. It
involved two senior emperors and two junior emperors. The members of
this ruling collegium could provide enough commanders for the troops.
At the same it offered the potential for co-opting new members in ways
that should have discouraged revolts. By the reign of Constantine the
experiment had failed, but it left open the idea that there would be
eastern and western emperors.
Meanwhile in the western reaches of the Empire the Roman army enjoyed
great advantages that allowed massive territorial expansion into
regions that were sparsely populated by Mediterranean standards. As
long as the Romans maintained their military advantage (which they did
up to Marcus Aurelius) the western legions could police the frontier
at low cost and the ability of the emperor as military leader was of
little consequence. However, the relentless migrations of populations
westward did not allow such an equilibrium to persist. Indeed these
thinly populated territories could barely feed the legions stationed
there, and these provinces could not provide enough soldiers to defend
themselves. As a result, the emperor recruited auxiliaries from
Germanic tribes and, if they served faithfully, settled them in the
Empire permanently. Because Rome’s frontier blocked migration, the
populations nearest the border were under pressure from populations
migrating from further to the east. Under these conditions instability
was rife. The frontier populations, like those on the borders of the
Chinese empire, were in close contact with Rome, at time serving as
allies, and at times raiding into the empire. The Goths, who defeated
Valens in 378, were in fact refugees from Hun expansion. The Goths
turned against Rome when local administrators failed to uphold their
settlement treaty. From there to the sack of Rome, the western
empire’s decline was extremely rapid. Neither efforts coming from
Constantinople, nor those of Germanic tribal leaders could reunify the
empire.
The collapse of the empire, seen in light of Chinese history, is not
surprising. It was based on an idea of overwhelming military force
that could not endure forever. What is more surprising, however, is
the failure to reconstitute the empire. While a large and integrated
polity survived in the East as the Byzantine and later as the Ottoman
Empire, in the West the process of political fragmentation proceeded
well into the Middle Ages. And, even once the process of
nation-building characteristic of the early-modern period was under
way, it was territorially un-ambitious by Roman or Chinese standards.
In fact, by the Middle Ages, within Europe, inheritance or marriage
was a more likely way to create larger domains than conquest.
Why all this territorial fragmentation in Europe? It is clear that the
Great Invasions—the massive population movements that followed
259—bear a great deal of responsibility. The invasions involved waves
of population whose demographic importance was locally quite large,
and they had dramatic political implications (Bury 1928: 37). Whether
the western half of the Empire was always thinly settled or whether
plagues or political disruptions drove population down is the matter
of some debate. What is clear is that the Great Invasions were a
process quite different from that of a military elite taking over an
agrarian empire. The secular nature of the migration, as well as the
serial nature of political change insured that dislocation was far
more extensive than was the case with invasions of the Chinese
mainland prior to the Mongols. It has often been argued for Europe
that these ethnically divided populations had cultural practices and
political structures incompatible with the Roman Empire (Bury 1928).
The evidence on this last point is far from compelling. Indeed, there
is ever-increasing evidence that these populations were not so
fundamentally heterogeneous and that they were far more attuned to
Roman culture than had been thought before. For instance many
‘barbarian’ leaders were in fact also Roman citizens. What is also
clear from the new scholarship is that notions of identity on both
sides of the frontier were very fluid (Geary 2002). Having breached
Roman defenses (or simply taken over some piece of territory),
invaders then faced the need to create the political conditions for
lasting control. For example, the leader of a group like the Burgondes
in fact faced multiple challenges. First, he had to hold together his
‘invading’ army—for without troops his capacity to hold his territory
would evaporate. Second, he had to find ways of ruling over the local
population that had been ceded to him. Most often this local
population was larger than that of his ‘invading’ group but failing to
establish his power would mean that his revenues would vanish. To
achieve these two goals invading leaders initially often chose to
integrate themselves with what ‘imperial’ authority existed. Yet, the
trend between 400 and 800 AD was unmistakable--the value of allegiance
to some higher authority declined simply because no authority could
guarantee protection. The Burgondes for instance were absorbed by the
Franks. It became clear that to ward off the threat of a new invader
or a neighbor, a ruler could only rely on the populations he
controlled. The value of political and cultural practices that would
have helped rebuild the empire collapsed, while those that promoted
local identity and local solidarity rose.
In Western Europe, the Roman Empire ended but it endured in the
Eastern Mediterranean. Indeed the polity centered on Constantinople
(and later Istanbul) proved to be a durable empire. The direct
successor of Rome managed to maintain a spatially large polity for
several centuries, including some outposts as far flung as Spain,
Italy and North Africa. For at least half a millennium after 378, the
Byzantine Empire was the wealthiest and most powerful remnant of its
Roman predecessor (Ostrogorsky 2002). It was also one of the locations
where the knowledge and culture of the Empire endured. Like its Song
counterpart, however, it proved militarily incapable of reassembling
the empire. In time Constantine’s heirs were replaced by Moslem rulers
who would take over all the Roman Empire’s eastern dominions, and make
its capital their own; but despite significant advances, (at one time
including Spain and into France and more durably in the Balkans)
neither Arab nor Turkish Caliphs were able to put this humpty dumpty
back together. Time and time again their advances were stopped either
in the Iberian Peninsula or in the Balkans. Thus by 800 AD the former
Roman Empire included a large polity in the East and many less stable
ones in the West.
Since Marc Bloch, medieval historians of Western Europe have had an
uneasy relationship with the Byzantine Empire and prefer to leave it
aside as a territory where feudalism did not take root (Platagean
2007). Modern research has both lessened and sharpened the political
differences between the Eastern Empire and Western European kingdoms
but the debate does go on. Ignoring the Byzantine Empire in European
history has two consequences that we prefer to avoid. First, it makes
the Roman Empire an epiphenomenon, political fragmentation one might
then argue is the norm at the western end of the Eurasian landmass.
The Iberian Peninsula for instance was fragmented before Rome took
over from Carthage, and it remains fragmented to this day. European
polities thus, for political or cultural reasons, can be assumed to be
small. Including Byzantium makes this proposition untenable. Second,
it reminds us that the political institution of Rome did no vanish
like some Atlantis but remained quite visible the Byzantine state.
Thus even in Western Europe, the idea of empire endured.
The political institutions of Rome did not fade from memory because
they evoked levels of security and prosperity that most Europeans
found wanting in their own times. Consider Charlemagne’s empire, the
last of the large empires before Charles V, as a sort of turning
point. Charlemagne succeeded in controlling a swath of territory from
France to Germany and from the Netherlands to Italy, though he did not
attempt to conquer either England or North Africa and his Spanish
campaign was a failure. Having achieved conquest he thought to
stabilize his polity by having himself crowned by the pope. He also
began the process of creating more enduring means for stability,
developing a centralized administration intent on providing some
public services. But the empire did not outlast him. Upon his death,
his three sons divided his territories amongst themselves and soon
were at war with each other. The upshot was that the eastern part of
his domains down to Italy would be known as the Holy Roman Empire,
while the western part would become France. By this time, most
political entities throughout Europe (kingdoms, principalities,
duchies, bishoprics and so forth) had no formal allegiance to the
emperor.
By the end of the first millennium, one lesson that the rulers of the
smaller polities did learn from Charlemagne’s heirs is that they
should not themselves contribute to fragmentation. Hence the
traditional practices of competitive succession or egalitarian claims
were replaced by rules of primogeniture. Primogeniture insured that
one kingdom would not be divided into separate parts if a ruler had
multiple male heirs, but it did not preclude a ruler of multiple
kingdoms from dividing them among his children. Had rulers merged
their territories into a single kingdom whenever they acquired new
ones, Europe might have taken a very different political path.
Rulers in fact did the exact opposite of consolidating their disparate
domains by formally recognizing a variety of localized practices in
territories they acquired. These practices or ‘customs’ covered
subjects as varied as the nature of real property, relations between
landlords and farmers (or lords and peasants), inheritance rules,
units of weights and measures, mechanisms for deciding levels of
taxation and the means to collect taxes, trade privileges, and more.
In fact, late medieval and early modern societies were most often
constituted of many clusters of such rights for specific groups based
on their social status, professional occupation, or place of
residence. Until the seventeenth century at least, the trend was for
the continued creation of such specific rights and hence for the
continuing fragmentation of political space. We can use the ruler
whose territorial sway could next rival that of Charlemagne as a case
in point. Charles V of Spain was separately the ruler of more than two
dozen territories, notably he was King of Castile, King of Aragon,
King of Naples, King of Sicily, Archduke of Austria, Duke of the
Netherlands, and Holy Roman Emperor. While the crowns of Castile and
Aragon had been united under Ferdinand and Isabella, this did not
imply that the territories were administered in a unified way, just
that the heir to one throne would also inherit the other. Lordship of
even the puny kingdom of Aragon involved the separate administration
of many territories: most importantly Aragon proper, Valencia, the
county of Barcelona, and separately the city of the same name.
Why did rulers in Europe accept such formal limits on their powers? To
a large extent they were motivated by expediency. European rulers were
well aware of the nefarious consequences of recognizing or granting
economic and political rights to specific groups. Nevertheless they
ceded these rights both to reduce the likelihood of revolt and because
it was often the only way to secure prompt tax revenue for the crown.
A local population might have conceded much greater authority to its
ruler had he offered them the kind of economic and social environment
that prevailed either in the heyday of Rome or around 1000 in China.
But everyone was well aware that rulers could promise little more than
Churchill’s blood, toil, tears, and sweat. Indeed the competition for
territory remained keen for centuries and rulers were eager to
participate in this contest. Thus, promises of using tax revenues for
local prosperity would surely ring hollow. Instead, local populations
wisely insisted on preserving their local privileges so as to limit
their rulers’ military ambitions. Certainly, a ruler who desired to
extend his domain further was unlikely to remove tolls or tariffs
between two of his territories. Doing so would have reduced his
revenues at the very time that he needed them most. The political
economies of empire and fragmented polities, as exemplified by China
and Europe, will prove, as we demonstrate in subsequent chapters, to
be significantly different.
Conclusion
Previous studies have taken the difference in the size of polities in
China and Europe to be a significant factor in explaining economic
change. For them political competition among European states has been
said to have positive economic consequences and China’s empire
delivered stagnation. We will suggest that the costs of such
competition were, in fact, heavy. Moreover, what advantages obtained
from political competition and warmaking in Europe were indirect and
unintended. Up to the eighteenth century perhaps, the direct and
deliberate positive consequences of empire far exceeded the indirect
and unintended benefits of political competition in other regions.
Many of the economic contrasts between China and Europe we develop in
succeeding chapters depend on the different spatial scales of states
in these two regions of the world. We will also discover that not all
economic and institutional differences are equally important; some
putatively economically significant differences between China and
Europe historically did not in fact have clear consequences, while yet
other assumed differences turn out to be not as stark as previously
portrayed.
Our first chapter has addressed the historical reasons for the
emergence of durable empire in China and the contrasting political
equilibrium of small competitive states in Europe. Military factors
and domestic political change clearly shaped the spatial scales of
polities across China and Europe. Up to the reign of Charles V of
Spain, Europe had made little impact on the world and China had little
interest in the western end of the Eurasian land mass. To the extent
that Chinese and Europeans had a common experience, it involved their
difficulties in dealing with the steppe people. The thirteenth-century
Mongols were the most formidable of these pan Eurasia-invaders; Their
leader, Tamerlane (1336-1405), was the last great challenger to
sedentary rulers in both the East and West. Once it became clear that
no military forces from the steppe would be able to take over both
China and Europe, their political histories became largely unconnected
for the next four centuries and their economic histories powered by
often similar but usually separate dynamics. In the next five chapters
we explore the consequences of differences in political space for
economic change in the era following Tamerlane. These differences help
us provide an account of what the Chinese and European economies
shared, how they diverged in the modern era, and why differences in
the spatial scale of polities in China and Europe still matter to
their economies today.
Chapter 2
Population, Resources, and Economic Growth
Introduction
The variation in family and household structure across Eurasia is
astounding. Be it age of marriage or the role of kin, nothing seems
alike in pre-industrial China and Europe. Because demography matters
to important economic phenomena, including the rate of savings, the
structure of markets, and ultimately for economic growth; scholars
have leaned heavily on variation in household structure to explain de
different pace of economic change in China and Europe (Smith 1976,
76-77; Jones 1981, 17-21). To do so they have relied on the notion
that the European nuclear household (with one generation of adults)
was demographically more prudent and more willing to participate in
factor markets than the Asian extended household. Much of this logic
was derived by considering mid-twentieth century data where the
connection between nuclear households, low fertility, high per capita
income, and market interaction is particularly strong. More recently,
rapid growth has taken place in many different societies as has the
fertility transition, weakening these connections. Hence we need to
take a closer look at the historial evidence and the logic of the
frameworks that invoke demography to explain economic change.
At heart, the arguments that seek to provide a demographic explanation
for the fact that industrialization first began in Europe draw on a
contrast between nuclear and extended households. The differences
between household types can be purely demographic or they can lie with
the incentive to participate in markets. In our view these differences
have been overstated. Chinese households, whether extended or not,
like European households, whether nuclear or not, practiced fertility
restraint. While extended household societies may have been less
involved with factor markets than nuclear household societies, both
extended and nuclear households were involved with factor markets to
some extent. When technological progress provided incentives for a
larger fraction of wage workers we expect households in both types of
societies to respond by entering the labor market at greater rates.
Demography does not explain why China was poor.
To develop our argument we rely partly on simple economic models. We
forsake a simple quantitative approach because, as we will show, one
cannot understand the evidence without a framework for understanding
its institutional and social context. In this and following chapters
our models are spartan and thus leave out nearly everything. What
these models lose in precision they gain in transparency. They should
be judged on a simple metric: The models are valuable: do they allow
us to highlight fundamental relationships and their implications,
including some important ones the literature has so far ignored? If so
they are valuable tools. In framing our arguments we are particularly
sensitive to the appropriateness of generalizations: while nuclear
households were dominant in Europe and extended households the norm in
China, in both regions there was a good deal of variation. To reduce
Europe to nuclear households and China to extended ones is to maximize
the differences between the two regions, and thus bias the argument
from the start in favor of the thesis that demography mattered for
divergence.
We will argue that differences in the extent to which activities in
early modern China and Europe were structured around families or
markets created only differences in degree rather than kind of
economic change. Indeed, we think that lineage relationships offered
Chinese households some economic advantages not available to European
households and that it is, therefore, difficult to establish
persuasively that the institutional differences in household form
actually favored certain European households over all Chinese ones in
the era before urban industrialization. This assessment on its own
still allows households’ labor practices to limit the development of
inter-regional and inter-sectoral labor markets. The evidence,
however, suggests that households’ or firms’ choices did not determine
where and when labor markets emerged and where and when workers moved
into urban factories.
I this chapter we will begin our journey into a comparison of Chinese
and European economic history. We will also begin to discard many of
the routes chosen by other scholars because they either lead to dead
ends or take us in circles.
Prudence and Poverty
The study of economic development in agrarian economies owes much to
the early nineteenth-century theories elaborated by Malthus in his
effort to explain the co-evolution of population and well-being in
England and elsewhere (Malthus 1992). Malthus recognized and brought
to the fore of social science the long run-interplay between
population and economy. But his work was hampered by a lack of solid
evidence beyond England, a cultural predisposition for finding the
good in all things English, and a serious methodological mistake.
Indeed, he projected the short-run correlation between social
structure and economic outcomes (such the English nuclear family and
early industrialization) into a general truth (that nuclear families
are everywhere essential for economic growth). The appeal of his
conclusions did not lead to any significant questioning of the logic
of comparative model. In both logic and substance, he remains
remarkably relevant to much work on comparative economic development.
Malthus’ ideas have endured because they are both simple and general.
As abstractions they are incontrovertible, but when deployed in
comparisons across countries, his conclusions prove untenable. He
posited four iron laws. (1) The resource base expands slowly, and thus
in the long run there is a fixed negative relationship between
population size and individual income. (2) In most societies, nearly
all women marry early and thus produce large numbers of children. The
population is kept within the bounds dictated by natural resources by
a “positive check”: most people are poor and mortality is consequently
high. (3) A select few populations constrain their fertility rates
below what is biologically possible and they are better off. This
“preventive check” requires that most women marry late and that some
do not marry at all. (4) In such meritorious societies only
individuals who can form a viable economic unit can marry. Thus,
marriage depends on being parents or children accumulating the capital
necessary to run a farm or a shop. In periods of high income, such
capital accumulates faster and thus individuals can marry younger
leading to higher fertility, while in bad times they are forced to
wait and would thus either have fewer children or not marry at all.
The ideas are remarkably simple, and not surprisingly they are the
base of a large edifice of social science research (for an elegant
synthesis see Wrigley 1988). Scholars have extended Malthus’
conclusions to posit that high-income economies are more likely to
undergo industrialization for both supply and demand reasons. On the
supply side a high-income economy has the resources to invest in the
(physical and human) capital needed for sustained growth, while at the
same time, higher incomes are disproportionately spent on manufactures
rather than food. We do not wish to quibble with the logic of these
arguments; what is of concern to us here is their suitability for
comparative economic history. In Europe, long after Malthus’ demise,
British scholars have continued to extol the distinctive virtues of
the British family relative to that of France, for the lack of a
better horse to flog. In their monumental study of the English
population, Wrigley and Schofield (1981) tried to demonstrate that
only the British practiced fertility control effectively. Yet later
work has largely invalidated their claim that other European
populations, and in particular the French, were less zealous in their
preventive checks. Indeed, David Weir has shown that in the early
eighteenth century, French families were more sensitive to their
environment than British ones; marriage rates and birth rates declined
more, while death rates rose more in response to a increase in grain
prices on the continent than they did in England (Weir, 1984). The
truly distinctive characteristic of the British population experience
is it’s uniquely (and clearly non-Malthusian) rapid increase from the
mid-seventeenth century onwards. Moreover in the eighteenth century
only one country started practicing fertility restraint of the kind
that was supposed to accelerate industrialization: France. But its
leisurely pace of demographic change did not turn the country into the
workshop of the world. To the contrary it seems to have slowed the
pace of economic transformation.
Nevertheless, continuing work pioneered by Hajnal (1965), some social
and economic historians are still interested in assessing economic
performance on the basis of whether demographic structure is similar
or different from Malthus’ prudent society, which they take to have
been realized in Early Modern England (e.g. De Moor and Van Zanden).
Most scholars, however, are now coming to recognize that secular
economic progress was achieved in Europe in a wide variety of
different demographic settings (Ketzer and Barbagli, 2001, 2002).
Although the family of European populations is large and diverse, it
remains a family and one might therefore want to range more widely to
seek confirmation of Malthus’s ideas. Indeed, Malthus himself (and
Adam Smith before him) speculated on demographic differences between
Asia and Europe (Malthus 1803 (1992), 41, 183-4). After Malthus the
speculation endured, carrying with it the unexamined inductive premise
that Asia’s development failure had demographic roots. One impetus for
this comparison came from the extraordinarily high population density
that Europeans visitors encountered in certain parts of Asia. When
Marco Polo returned from China with tales of riches and splendors, the
dense populations he described were prosperous. Over time, however,
this connection disappeared and, by the eighteenth century, travelers
were emphasizing the deep poverty of Asia’s large populations. These
conjoined density-poverty observations were not lost on social
commentators. And China’s population size (the largest in the world
for nearly all of recorded history) was often invoked as a constraint
on efforts to spur economic development. The serious Chinese famines
of the 1870s, early 1920s and late 1950s were taken as further
evidence that the country labored under a severe Malthusian
constraint: its population was too large for its economy to support.
Even during, the past quarter century critics of China have continued
to indict China’s demography—nowadays for its tilted sex rations. In
the past, however, the theme was constant. For cultural reasons,
Chinese families were unwilling to limit their fertility whatever the
social consequences.
Recent work in Chinese historical demography forces us to revise our
thinking. To begin with, as Lavely and Wong (1998) point out the
growth rate of China’s population was slower than Europe’s over the
long run (1400-1900). For the pre-industrial period the two
populations grew at two-thirds of one percent a year. Given that China
was considered prosperous at the close of the Middle Ages, one can
hardly impugn demography for its poor performance. Similarly, slow
population growth cannot explain either Europe’s or any particular
European country’s prosperity. Indeed those areas where economic
growth was more rapid had the fastest population growth. For China,
the literature has moved beyond casting doubt on the importance of the
Malthusian positive check to documenting how some Chinese populations
deployed their own preventive checks on fertility. In Europe the
Malthusian preventive check involved women marrying late and or
remaining unmarried. In contrast, Chinese women married earlier and
more universally than their European counterparts. Although such
marriage patterns prevented the operation of European preventive
checks, the Chinese did control fertility, among other ways through
post-natal selection of which children would be reared. Some Chinese
populations, at least, had much greater spacing between siblings and
ended childbearing at an earlier age than did Europeans (Lee and
Campbell 1997, Lee and Fang 2001). In China, as in Europe, demographic
growth rates were far below those biologically possible (Lavely and
Wong 1998). Although the areas in which rigorous population
reconstructions can be made have been limited to parts of northeastern
China, this evidence requires us, at the very least, to invalidate the
prejudiced view that all Chinese families sought or had high
fertility.
BIN HERE I THINK it would be simplest to reproduce the table on life
expectancy in LAVELY AND WONG. OK I’LL CHECK
Over the past millennium, Jiangnan, the region near present-day
Shanghai, has been the country’s focal point for the most advanced
forms of production, densest networks of markets and merchants, and
most sophisticated types of consumption. If the Jiangnan population
were regulated by the Malthusian positive check, we would expect it to
grow rapidly, either after a positive economic innovation or after
some political or demographic catastrophe had reduced the numbers of
residents. Yet, some scholars have suggested that population growth
rates in this region were lower than those elsewhere in the eighteenth
century (Citation). Similarly, others have shown that the repopulation
of this area, after the mid-nineteenth century rebellions almost
brought down the ruling dynasty, was very slow (Li 2003; Bernhardt
1992). This latter observation might initially be thought to support
the notion that the resource base of Jiangnan was near its carrying
capacity limits. But the region’s high standards of living, both
before and after the rebellions, suggest that slow population growth
was not likely the result of extreme poverty and the consequent
Malthusian positive check. Instead, the population’s slow expansion is
consistent with the existence of preventative checks. Unfortunately,
we have no direct evidence of fertility rates in this region. We
cannot be sure as to how much of the population moderate rate of
change was owed to family strategies and how much to emigration. One
thing is sure though: the incidence of famines there was limited
(Citation).
Famines play an important role in comparative economic history because
qualitative evidence of their occurrence and of mortality crises is
abundant; even in the many places where demographic rates cannot be
measured far back in time. For preindustrial societies famines are
taken as a prime indicator of the operation of the Malthusian positive
check (cf Fogel 2004). Few would dispute that Jiangnan, China’s
wealthiest region, was productive enough to support its population
well above subsistence, and that demographic constraints alone did not
prevent capital accumulation. Other parts of nineteenth-century China
did suffer serious natural disasters and famine, and many have assumed
that crop failures and the lack of food led to hunger and death
(Citation). It is no longer obvious that these tragic circumstances
represent a Malthusian indicator of an agrarian economy overburdened
by a large population. Instead, scholars have come to emphasize that
in times of crisis access to resources is economically, socially and,
most importantly, politically determined. Famines typically result
from the poorest members of society’s lack of entitlements to food.
People die even though there is food physically available, but those
who need it the most fail to gain access (Sen 1981, Drèze and Sen
1989, Fogel 2004). In this light a Malthusian population explanation
of the limited likelihood of economic development outside of
northwestern Europe appears less powerful than one that considers
political institutions and crises. Indeed, politics as well as other
forces outside the demographic regime, can cause failures both in
production and in distribution and in turn lead death rates to jump
skyward (Citation for china re famines as politics).
Overall, the evidence increasingly suggests that China between 1600
and 1800 was not a society laboring under the positive check. Instead,
much like Europeans, Chinese families controlled their demography.
That their mechanisms for doing so were different from those practices
elsewhere does not mean that they were less effective. Moreover other
than in the traumatic political period 1850 between 1978, mortality
does not appear to have been very responsive to income. Although
volcanic explosions like Tambora, foreign invasions, or large scale
civil wars could engender significant spikes in mortality those events
are irrelevant to understanding any particular demographic regime (Li
Bozhong?). In China, as elsewhere, famine and mortality crises were
only rarely resources crises, far more often they were social crises.
During the first century and a half of the Qing dynasty, in
particular, there were with very few episodes of severe food
shortages, let alone major famines (Citation).
Our discussion of Europe and China’s demographic regimes has hinted at
the fact that there were remarkable differences in demographic regimes
within Europe and within China. For Europe this has led scholars to
emphasize the importance of the prudent nuclear family in both
demography and in economics. Because areas where the nuclear family
dominated were at the center of the burst of economic change that
preceded the Industrial Revolution, such arguments have had force both
for comparisons within Europe and between Europe and the rest of the
world. As we have seen, China has often been taken to be populated by
imprudent extended households. Yet we know that extended households
were never the only family structure in China, and such households
were more prevalent in the south and southeast than in the north and
northwest. To the extent that nuclear households are prudent and
extended households imprudent, glossing over the variation internal to
China and Europe, builds in a bias that is favorable to Europe over
China in ways that artificially inflate the relevance of demographic
structure. If the logic of prudent nuclear family leading to more
prosperity were to hold, then those areas of China where families were
smaller should have been more prosperous than those where families
were more extended. There is little evidence that northern China was
substantially better off than southern China however (Allen et all
2007). Just as in pre-1600 Europe, within China the relationship
between family structure and economic success is hard to discern.
Family structure and demography are important, even if they do not
allow us to split the world into positive versus preventative check
zones. They are important because in the pre-industrial era when most
enterprises were tiny, there were critical interactions between family
structure and labor markets. As we explore these issues in the next
section we will once again find that generalizations about labor
markets are based on a series of assumption that bias findings in
favor of Europe and against China. Equally important, if we abandon
the idea that in most areas of Europe or Asia populations were at
subsistence, then it become worthwhile to chart the course of incomes
and to understand how such resources were allocated to consumption,
savings and demography.
Real wages
Measuring and comparing levels of incomes or well-being over long
periods of time in any given place is difficult. Our task is made more
complex since we seek to understand the changes in standards of living
for the two ends of Eurasia. In doing so we cannot settle for evidence
from some time in the twentieth century and project such evidence back
into earlier periods. Doing so would inevitably put China in a bad
light given the enormous differences in income levels that prevailed
between China and Europe by 1900. A better solution is to rely upon
wage evidence from earlier centuries. To be sure, such data are
imperfect for a variety of reasons but there are excellent grounds to
believe that the evolution of income roughly followed the evolution of
wages. We know that when economies are growing rapidly, wages will
rise and when economies run into trouble, wages fall. This, after all,
is nothing more than the principle that in the long run wages are
equal to the value of the marginal product of labor: when economies
are growing that marginal product is rising. Over a decade or so, some
growing economies may experience little change in wages because
technological change may substitute physical capital for either
unskilled or skilled labor. In the long run, however, capital
accumulation makes labor more valuable rather than less.
Even if we accept that wages are in the long run related to the
performance of an economy, there remains a second conundrum. It is
likely that the relationship between wages and growth was different in
Europe and China because the relationship between households and labor
markets was different. Rather than either assume the problem away as
recent work on relative wages has done, or assume that the problem
makes the comparisons impossible, in this section we provide a
framework for understanding how differences in labor markets might
drive the wages we observe. This problem is not merely technical; it
harks back to the long-held myth of self-sufficient, market-averse
agricultural households. Suppose for now, as historians of both ends
of Eurasia would often have it, that the myth is true (e.g Reddy 1984,
Huang 1985) and that many or most households do not participate in the
labor market. By definition, autarkic households’ income have nothing
to do with wages and it could be that wages rise while at the same
time the incomes of autarkic households are falling or vice versa.
Finding that wages were higher in, say, China than in Europe would
tell us as little about the relative incomes of these two economies as
looking at garment workers in the U.S. relative to civil servants in
India. More generally, any comparison between Europe and China would
be hostage to the fact that the fraction of households in the wage
economy varied over space and time. Fortunately for us, self
sufficiency was at most an ideal. One that may have been more imputed
to agricultural households by later analysts than espoused or attained
by the peasants themselves. The key is to recognize that most
pre-industrial households were both families and farms: units of
consumption and savings as well as businesses.
Farm households found it expedient to participate in some factor
markets, like those for land or labor for many reasons. But the root
of such transactions lies in differences between the household’s
endowments of factors and its desired size as an enterprise. At any
point in time a farm household has some land, capital, labor and
skills. Here land and capital refer to the real assets a household
owns or to which it has long-term rights. Labor refers to the work
capacity of the household’s members. Skills include talent and
experience with farming or other endeavors. A household’s endowments
clearly reflect the history of its fortunes. Real assets will be large
in a household that is productive and that saves rather than consumes,
while its labor and skills depend on demographic strategies and age
distribution. For very large numbers of households to have avoided
factor markets would require far more predictability in demography and
enterprise than is conceivable. Indeed, self sufficiency requires that
the family’s land-labor ratio remain roughly constant. Among other
things, that would require that no family ended up with either more or
fewer offspring than expected and that these children began
contributing to household labor precisely at the time their parents
were becoming older and less able to work. It also requires that
ability be fully transmitted from parent to child along the
generations, so that better farmers (who could farm more land) never
end up growing up in a land poor household. Clearly, no one believes
such assumptions hold anywhere in the world.1 And the reader can rest
assured that equally daunting assumptions must be made about savings
behavior, the predictability of crops and so forth. Whenever these
assumptions fail, there will be large differences in the marginal
product of labor across farms. Peasants have long participated in
factor markets to equalize these differences.
Yet we do not need to rely on mere theoretical arguments. Both in
China and Europe land, even among farm households, was unevenly
distributed so that almost no family ended up with just the right
amount of land to farm efficiently on its own. For Europe evidence of
such inequality abounds (see Baehrel 1962, Kaplish and Herlihy 1985,
Soltow and Van Zanden, for an interaction with demography see Emigh
2003). In China, as in Europe, most families had too little and some
had too much land. The imbalance would lead to them to hire in and
hire out labor, or alternatively to buy, sell, rent in or rent out
land. Of course whether households entered the land or labor market
surely depended on a variety of factors, and one might imagine that
the burden of adjustment was laid squarely on the land market in many
societies effectively shutting down the labor market. Yet a very
powerful force against such a one-sided solution lies with variation
in entrepreneurial talent: the ability of the household head to run
the farm efficiently. A capable rich farmer would want to hire
workers, while one that was less talented would be better off renting
out at least part of his land. When making these decisions, a farmer
could compare the cost of the labor they might hire to what they could
earn from it, and if thinking of selling their service they would also
contemplate the return to more labor applied on their farm to what
they could get as wage earners. To be sure the institutions behind
these exchanges could be quite complex and we will not debate the
issue of when and where they would qualify as markets (Hoffman 1996 ch
3). For our purposes, it suffices to recognize that few agrarian
households could be self-sufficient and, thus, that wages can tell us
a lot about general economic conditions. From here we proceed in two
steps, for the rest of the section we review the wage evidence,
leaving its interpretation to the next one.
For Europe, economic historians have been able to measure wages in
many locations and over long periods of time (the most famous of the
early work being Beveridge 1965). They have been particularly
fortunate that government agencies, municipalities, and charitable
organizations like hospitals and monasteries all kept detailed records
of their expenditures, and in particular of the wages they paid. What
is even more remarkable is that a very large number of such account
books survive and detail both wages and the prices of many of the
commodities that one would want to include in a consumer price index.
This effort has led to three relevant findings (Brown and Hopkins
1981, Allen 2001). First, over the half millennium that preceded 1800,
wages fluctuated roughly inversely with the level of population. While
real wages were low prior to the Black Death (1347-48), they rose
steadily for the next half century and achieved a peak in the early
fifteenth century that was not clearly surpassed until the eighteenth
century. Second, at any point in time the variation across places in
the real wages of comparable occupations was as large as the variation
of wages in a given place from 1600 to 1800 (Allen 2008). This kind of
variation is surely not consistent with a Malthusian equilibrium
everywhere. If the lowest wages are those of subsistence, then most
places at most times were not at subsistence. Finally, the highest
wages were found in the most densely populated areas (Ditmar 2009).
Overall, European wages accord well with more narrative sources of
economic success or failure.
Population densities were positively associated with levels of
urbanization but not closely tied to levels of agricultural
productivity. Urban settlements certainly required agriculture to
produce a surplus in order to be viable, but as George Grantham has
noted, lower levels of agricultural productivity do not seem to have
been a significant constraint on economic progress (Grantham, 1993).
Economic growth depended directly on the capacity of localities to
structure their markets in ways that encouraged the division of labor
and specialization. This reorganization was largely an urban
phenomenon, which placed demographic demands on rural areas. Where and
when cities boomed, agricultural productivity growth followed. (But as
was the case in the Low Countries and the Veneto, imports of food were
also often an integral part of the urban expansion). The European
demographic regime, on the other hand was a more serious constraint.
Fertility was simply not high enough to sustain large populations in
what were biologically hostile environments. To put it simply, cities
killed people at such a rate that the countryside had to produce a
large surplus of births to sustain cities’ demographic needs (Wrigley
1967). In turn cities had to have high wages to induce the immigration
necessary for their expansion. What is striking overall for the
pre-industrial period relative to the nineteenth century is the
failure of advances in wages in one area (say the Low Countries) to
spread to the whole of the region. This localized success, far more
than Malthusian subsistence, seems to characterize the pre-industrial
economy.
By the mid seventeenth century, Robert Allen argues, a clear division
had emerged. In the northwest of Europe, and in particular in Britain,
wages were high and had a long term tendency to rise. In contrast, as
one moved South or East workers earned less and their pay tended to
stagnate or fall (Allen 2008). It was not until late in the nineteenth
century that wages began to climb more generally. It is also likely
that returns to land rose faster than wages (because land was a scarce
and fixed factor and there was productivity growth in agriculture) and
that while the price of capital had been falling since the Black
Death, the capital stock had grown faster than the economy (Clark
2007, Van Zanden 2007). Indeed, economic growth in Europe was at least
partly a process of capital deepening. Hence, the path of wages
probably understates the aggregate level of increases in income.
Nevertheless, because in most places the return to land and to capital
accrued to a small number of people, wages remain the most
representative form of income.
Until recently, income levels in eighteenth-century China were not a
topic of concern, because nearly everyone agreed that it was a
subsistence economy. Debate centered on the degree to which the early
twentieth-century Chinese economy was growing or not, with some
scholars using wage data to buttress their arguments (Rawski 1989;
Brandt 1989). For scholars who saw economic growth coming in the
late-nineteenth and early-twentieth centuries, as well as those who
remained skeptical, the eighteenth-century situation was assumed to be
one of poverty. Either the eighteenth century provided a base from
which growth subsequently emerged, or it was little different from the
conditions in which the economy found itself in the early-twentieth
century (Huang 1985; 1990). Together, scholars all assumed or asserted
that the Chinese were poor. A first revision to this view came from
scholars who doubted that Qing China was such an economic failure, and
the recent work of Kenneth Pomeranz (2000) provided tantalizing
suggestions that in the Yangtze delta at least consumption might well
have been quite high. In the last few years the debate has taken on
new life as scholars are slowly but surely bringing wage evidence to
bear on the matter (Yan 2008, Allen et al 2004).
The effort to measure individual incomes in China prior to 1850
remains in its infancy, but already some important findings have
emerged and they are fully relevant to our endeavor. To begin with,
from the eighteenth to the twentieth century there have been
considerable differences in real wages across China and the range of
variation seems to be on the same order of magnitude as in Europe
(Allen et al 2004, Yan 2008). Thus China was not a simply an ocean of
poverty, there were regions with relatively high incomes. It was also
not simply a spatially static empire. As it expanded, Han population
migrated to what was a very large frontier (Pomeranz 2000:84). Yet,
this migration presents a puzzle since individuals appear to have been
leaving high-wage eastern and southern regions to settle in poorer
western and northern ones. One can resolve this puzzle, by once again
breaking free from the idea of homogeneous self-sufficient households,
and consider that it is likely that wealth was unequally distributed
within the richer regions. While successful lineages would have had
little reason to migrate, poorer people might well have been tempted
to venture out toward the frontier. Indeed, poor people in prosperous
areas could well have expected that combining the lower wages in poor
areas with the income from opening land might be more desirable than
simply the high wage in their home area. Such calculations were
certainly important elements of the motivation for westward migration
in the U.S. (Galman and Pope1989, Ferrie 1999). Chinese peasants could
have made the same calculus in terms of the income they might have
gotten as tenants in the Lower Yangtze versus the income possible as
owner/occupiers on the frontier. Potential migrants could not make
such decisions without some knowledge of factor prices—whether the
wages in the first scenario or rental price of land in the second.
Once again households and markets are conjoined.
The evidence on wages also suggests that there was relatively little
growth, if any, in Chinese real incomes between the mid-eighteenth and
the mid-nineteenth centuries. Taking into account demographic change,
the findings can be reframed to note that despite an increase in
population from some 200 million to well over 300 million there does
not seem to have been much, if any, decline in wages in many parts of
the empire. The implications of this stasis are, on the one hand, that
the divergence in economic performance between Europe and China is
likely to have started before 1700 and was probably quantitatively
significant before 1800. On the other hand, those scholars who have
argued for a Malthusian involution in Qing China will find little
comfort in the new evidence. That wages remained relatively stable
despite a large population increase does not suggest that the marginal
product of labor in agriculture declined at all.
Recently, attempts to compare levels of income in China and Europe or
to be more precise between port cities at opposite ends of Eurasia,
have been stymied by a methodological conundrum. As Robert Allen has
shown this comparison suffers heavily from an index number problem
(2004). Rice was cheap relative to wheat in South China while the
reverse was true in England. Wages were such that an English dock
worker around 1700 was simply unable to afford a Chinese consumption
basket in London. Similarly, a Chinese dockworker could not afford an
English consumption basket in Canton. If one takes a composite index
(making both buy a mixed basket) then wages in Canton and London were
similar.
The index number provides arguments for scholars who favor a
relatively high Chinese income in the heyday of the Qing dynasty. The
larger issue, however, is how to interpret these findings. We begin by
gleaning those lessons that do not depend upon whether or not the
labor markets in China and Europe had similar institutional
structures. First, other data suggest the wages of unskilled Chinese
along the coast were higher than those in the poorest areas of Europe,
but lower than those in the richest ones in the eighteenth century (CITATION).
Since the coastal areas were those with the highest income, one should
conclude that the range of wages in regions of China and Europe
overlapped at least through the early eighteenth century. Second, the
lack of growth in Chinese wages between 1650 and 1850, when wages in
Western Europe began to surge upwards, leads us to conclude that the
country was falling behind. Third, no matter what trends or stasis
obtained before the mid-nineteenth century, the subsequent hundred
years witnessed profound political and social turmoil. Some economic
gains were no doubt made in the twentieth century, especially in the
region centered on Shanghai. Nevertheless it is also likely that there
were few sustained gains, so that in many provinces Chinese incomes in
1950 were similar to those of 1700. Political dislocations starting in
1850 with the Taiping Rebellion to Communist victory in 1949 force us
to consider the possibility that in some places incomes might have
been lower in 1950 than in 1700.
Clearly then, the wage evidence strongly supports the thesis of a
“Great Divergence,” and on such data alone one would date it rather
late, perhaps later than 1750. Yet, as we will argue in the next
section such conclusions are unwarranted. Indeed once placed in their
institutional context, wage data become difficult to interpret and
would place the divergence much later, perhaps as late as the 1820s.
However, we believe the structural and institutional divergence is
actually far more ancient than 1820 dating back at least two
centuries. While the disconnect between wages and economic structure
is partly explained by differences in how labor markets operated, most
of the differences in wages come from the slow process at which
industrial technologies spread(Mokyr 1985:5).
The Household and Labor Markets
Our interpretation of the available demographic and wage evidence also
has a more proximate implication. It forces us to reexamine the
conceptual frameworks that scholars use to interpret the economic
histories of China and Europe. Indeed, the lessons we learns about the
connections between vital rates and say food prices depend on what
kind of an economy and society we are studying. Economic historians
have, by and large, come to see most European economies and in
particular households as imbedded in markets. Hence prices tell us
about relative demands and relative productivities. For China, most
scholars would concede that the exchange of commodities was largely a
market phenomenon and one in which the state did not play its usual
role of hindering change or growth. Factor markets, however, find
themselves squeezed between two institutions that are given far more
importance—the state and the household.
Here we must focus on the interaction between markets and households,
leaving aside politics; a factor that we will consider extensively in
a subsequent chapter. The key question we want to tackle is might
China’s poor long run performance be traceable to family structure?
The traditional answer is yes and the reason was that extended
households substitute for markets while nuclear households are
imbedded in them. As was the case for fertility, scholars have assumed
that (European) nuclear households were more favorable to markets.
Indeed a smaller household depends on markets precisely because of its
varying labor supply over its life cycle. Moreover, the breakup of the
household when progeny reach adulthood creates a demand for land and
capital markets. Indeed, even in an initially egalitarian society some
parents will find themselves short of children, while others are
abundantly endowed. Hence there will have to be flows of land and
capital from ‘rich’ household to ‘poor’ ones, or there will have to be
flows of labor from ‘poor’ households to rich ones (here rich only
means that the household’s land is large relative to its labor). In
addition, the children must be set up with a livelihood prior to the
death of their parents. Hence one generation must save (in land or
finance) in expectation of the marriage of the next and unless these
plans are fully realized there will be some borrowing of capital and
land around the time of marriage.
Unlike Europe where the nuclear household (one generation of adults)
prevailed, China has been seen as the domain of the extended household
and the kin group. In the classic Chinese extended household all the
male descendents of a household head live under one roof. Further, the
household head occasionally had more than one wife; finally, a
potentially large number of permanently celibate males may also find
themselves living with and under the authority of a kin head of
household. Hence the household is large. Making the demographic unit
even more complex are the important relationships among non-residing
kin groups and even more broadly, lineages (groups of individuals who
share a distant patrilineal ancestor). What then of the relations
between Chinese families and factor markets? To begin, Chinese
families may have faced less demographic uncertainty because the shock
to individual couples was already partially averaged out in the
extended household and even more so in the lineage group. Hence, there
was less demand for land and labor reallocation at every level of the
socio-demographic structure. Furthermore, these large demographic
structures reduced the demand for markets because they reallocated
resources internally (Chaianov 1966). The connections between the
despotic state and extended households as twin impediments to factor
market development can be drawn quite easily. For instance, an
extended kin system may be efficacious when law fails to secure the
property rights of individuals against the despot’s greed. Moreover,
the authority of the lineage head may be more useful in securing such
transactions than the justice meted out by corrupt public officials.
Resources reallocated within the kin group are, in a sense, less
visible, than those reallocated through the markets. In a more
optimistic vein, the opportunities for long-distance trade in an
integrated empire were more easily realized by extended households.
They could secure their dealings with familial norms rather than
having to rely on formal contracts left hanging on the whim of the
judicial system as nuclear households would have to do.
One can summarize the literature’s argument in the following way:
Extended households will have fewer interactions with the market than
nuclear households. The larger the share of nuclear families in the
population, the more market interactions there will be in the economy.
More market interaction implies a higher efficiency in the allocation
of resources. Hence economies with more nuclear families will have
higher incomes, higher wages and higher growth. A positive correlation
between the fraction of the population that is in nuclear households
and wages is taken as proof of the argument. To a large extent the
argument was developed inductively by scholars who knew that England
was the exclusive domain of nuclear households and that it had been
the cradle of the industrial revolution.
A more careful look at the argument will show first that societies
with more nuclear households did indeed have larger labor markets.
Second, the positive association between wages and the share of
households that are nuclear obtains mechanically even if the market
does not raise aggregate efficiency; thus a positive correlation
between wage and nuclear households is not sufficient to make
conclusions about efficiency. Third, the impact of household structure
is largest when average firm size is smallest. In fact, once factories
employ hundreds of workers differences between extended and nuclear
household societies are negligible (because almost everyone is a wage
worker).
Let us start with the first point. We develop the mathematical
analysis in Box 2.1, but the mathematically disinclined reader can
just focus on the text. Firms (and farms) in pre-industrial economies
were small, so lets imagine that all the firms in the economy employed
two individuals, an entrepreneur and a worker. Let us assume that each
adult individual is equally likely to be good at management (then he
is an entrepreneur) or not (then he is a worker). Now we can complete
the model by laying out family structure. Suppose a nuclear household
has only one member who might become part of a firm (women are fully
employed in domestic activities). In this setting half the households
will have an entrepreneur and they will hire workers from the other
half of the households. Now examine a society with extended households
that have two members who might become part a firm. Some will have two
entrepreneurs and will start two firms, others will have two workers
who will both participate in the labor market, but some will have one
worker and one entrepreneur and they will not be in the labor market.
In fact, the share of the worker-entrepreneur population that is in
the labor force in the extended household society is half that of the
nuclear household society. An alternative would take the nuclear
household as having two working members (husband and wife) and the
extended household three or more, and again, a society with smaller
households will have a larger share of its workers in the wage labor
force.
{BOX 2.1 about here}
More realistic assumptions that would consider gender would dampen the
difference, making extended households larger would increase the
differences between the two societies (in the extreme, where the
economy is a single household, no-one is in the labor market because
everyone works for a relative). As we shall see increasing firm size
dampens these differences. In any case the basic intuition that larger
families have less need for the market is extremely robust. It is
important to emphasize that there is a radical difference between less
and none. Even if we allow extended households to be very large—to
have, say, ten members who can be workers or entrepreneurs and make
firms very small, the labor force is still 10% of the
worker-entrepreneur population. Labor markets remain active because a
large fraction of households are either hiring at least one worker or
have at least one member worker in the labor market.2 Thus factor
markets are important everywhere at the margin and available to
respond to economic change. To rescue the idea that China’s smaller
factor markets were responsible for poor economic performance, one
would have to believe that larger markets are massively more
efficient. In fact, as we show below, China’s smaller factor markets
in the early Qing had probably no long-term impact, because factor
reallocation proceeded in other ways.
Beyond the simple scale of markets, we can make sense of the
connections between socio-demographic factors and markets by
considering the household as a firm. In that sense this section
borrows heavily from the work of Gary Becker (Becker 1981) though our
claims are less universal. To be sure considering the household as a
firm ignores many dimensions of its activities and its internal
structure. Yet for the secular problem of interest here, treating
families as firms proves both parsimonious and valuable. In particular
it allows us to ask: when will being a member of a large family,
extended kin group, or lineage organization be economically valuable
relative to being on one’s own? Membership in an entity larger than
the nuclear household is desirable because it gives access to
resources without recourse to the market (and thus saves on the
transactions costs of market interaction). On the other hand, the
leaders of such entities make demands on individuals and must devote
resources to insuring that these demands are met. Thus membership in a
larger demographic entity implies bearing the cost attendant to
maintaining this organization. The question is not so much which
household structure is best, but what combinations of family structure
and markets are best?
We can begin to answer this question by considering Coase’s work on
firms. In his celebrated ‘The Nature of the Firm’ (1937), Coase argued
that markets stand between organizations. Obviously a market stands
between a producer and a consumer, but Coase saw the point as more
general. A business will purchase some of its inputs and sell some of
its outputs. Some inputs (say the land on which a plant is located)
may be owned and some outputs (say machinery made in the firm’s own
shop) may not be sold. That much is true of all enterprises. At the
limit, one could, however, imagine a firm that only carries out one
step in a single production process and owns nothing: it buys all its
inputs (including renting its equipment and plant) and sells all its
output. Most often firms are somewhat vertically integrated (carry out
several steps in a production process), or somewhat horizontally
integrated (make different kind of products), or both. In another
extreme we can imagine a full command economy where a single firm
organizes all of production. When a firm extends its reach up or down
one step in the production process it is eliminating a market and
replacing it with a structure of authority. For instance, if a miller
buys a bakery to turn his flour into bread, he is no long selling his
flour hence the market for flour has ‘disappeared’ but the miller must
now supervise the bakery. If he buys the bakery the miller is
effectively deciding that he prefers supervising the bakery than
dealing with the market for flour. Coase argued that the extent of
integration would reflect transaction costs, and that, a priori, it
was not possible to decide whether firms should or should not be
integrated.
This logic has important implications for our households. The extended
household is simply a more integrated family than a nuclear household.
By analogy with Coase’s firms, in some situations the extended
household will function better than the nuclear household but not in
all. Unlike Coase’s firms, individual families do not chose the extent
of integration, rather the prevalence of nuclear households is an
historical pattern persistent both across space and over time. Because
industrialization proceeded in Europe before the rest of the world,
and in England before elsewhere in Europe, scholars have succumbed to
the temptation of associating the nuclear family (and whatever other
characteristic of England strikes their fancy for that matter) with
efficiency (see De Moor and Van Zanden 2008). It now seems most
prudent to abandon that inductive reasoning. In fact, nothing
obviously links industrialization with the nuclear household.
Industrialization, after all, does not occur in small firms but in
relatively big ones. Having jettisoned a doubtful relationship between
household size and economic change, the question of whether the
nuclear or extended household is to be preferred is no longer easily
answered.
One could extend this line to demography. It would lead us to ask: are
large or small family units going to do a better job at controlling
the rate of population growth? In particular, following Malthus, we
are concerned with the intensity of the operation of the positive
check. Here we can see the full force of the firm metaphor. The small
family will limit its fertility because, as Becker has suggested,
parents care for the welfare of their children and thus have only as
many children as they can afford. In particular their calculus will
involve their wealth and the prevailing wage rate (since someone who
is not wealthy enough to set up a farm or an artisanal enterprise will
have to work for wages). Yet, to the extent that families want to have
descendents and that mortality is both severe and random, they will
tend to have more children than they would like on average. The small
family is oblivious to the effect that individual fertility has on the
pay rates, the rental price of land, or capital because it takes all
these prices as given. Consider now what a benevolent despot that
seeks to maximize the individual income of the next generation might
do. She will use the same reasoning as the altruistic parent but at a
social level. And that would eliminate both the prudential motivation
for high fertility and the externality caused by the aggregate effect
of fertility on wages. One might well argue that extended households,
being larger than nuclear families will approximate the benevolent
planner better. When one observes demographic behavior among large
extended households in North China the exercise of Malthus’ preventive
check was quite intense and the patterns of fertility unequivocally
show that the head of the household chose its size deliberately (Lee
et al. 1992, Lee and Campbell 1997). To take one example, the head of
a large extended household or kin group would want to limit the
fertility of the couples in the lower part of the hierarchy when wages
were low. Indeed, he could then purchase any labor that he might need
on the market without having to accept the cost of additional kin. Kin
might seem cheap but in an economy with abundant labor, they are in
fact expensive. Familial responsibility implies that co-resident kin
(in particular for men) must receive a welfare no lower than what they
could get on the market. Moreover unlike outside workers they cannot
be fired in adverse economic circumstances. Thus unless the leader of
a large kin group wanted to keep many retainers about for military
purposes, he had every reason to be responsive to the Coasian
trade-off between making its own labor force or buying it.
It might seem that we have exaggerated the efficacy of larger
households in dealing in the market or controlling fertility. While
the extended household offers a mechanism for regulating fertility, it
requires a structure of authority. That authority is not necessarily
benign, just as management in a firm cannot be assumed to be
maximizing profits. One might consider two kinds of inefficiencies:
the first is the household head has a personal desire for a large
household (because although that may make him poorer it may make him
more powerful). The second is that the household head may repress
everyone else’s fertility while maximizing his own to the extent that
the household is larger than the equivalent set of nuclear units. The
Campbell and Lee evidence does not suggest that either bias was large.
It may be that in other political and cultural contexts, extended
families have massive demographic impacts; that does not seem to have
been the case for China.
The argument above suggests that differences in household structure
and demography, while striking on their own, were not likely to have
had much economic impact. Although extended households may interact
less with markets, they do not suppress them. And while there may well
be some transactions costs in dealing with family members, some (that
would be born in market interaction) are avoided. While factor markets
were likely to have been less active in China than in Europe, it would
be preposterous to think that household heads would not have paid
attention to wage rates or the rental price of land. In fact, our
analysis suggests that even abstracting from life cycle issues,
households in China and Europe would both be engaged in some markets
at high frequencies. The model’s starkness has further implications.
Since whether one is or is not in the market for workers depends
solely on whether the household has an excess supply of entrepreneurs
or workers. It follows that wages of workers in the labor market are
an accurate measure of wages for all workers. Hence differences in
wages would be a good statistic for the marginal product of labor in
the whole economy. But that would make short shrift of the transaction
costs that prevail in labor markets. We must elaborate the framework.
Households and Wages
In our model, individuals grow up e as ither entrepreneurs or workers,
but now let workers be of two types: high or low ability. The better
type could be more diligent, clever, or in other ways abler. We
analyze the effect of this changed assumption on our model in box 2.2
to show that wages are lower in extended household economies than in
nuclear household economies even though aggregate output is identical.
In this setting half the individuals are entrepreneurs, a quarter high
ability workers and a quarter low ability worker and each household
received the luck of a draw. Assume that the marginal product of the
better type is twice that of the less able type so average ability is
1.5. Ability is not observable on someone’s face or from a diploma but
it can be learned by employers over time. The family head knows the
ability of the worker, an outside employer does not. Hence the initial
wage in the labor market is simply the wage one would pay a worker
with average ability. The question is thus what is average ability in
the labor market? For societies with nuclear households all workers
are in the labor markets so average ability is 1.5. But as households
get larger only net surplus workers are sent to the market and they
will to the extent possible be the low ability workers. The reason is
that the pater familias who knows ability can promise his able worker
more than the labor market can offer. As the analysis in box 2.2
shows, this selection effect leads to market wages that fall as
households get larger. The reason is that as households get larger
they send a smaller fraction of their workers into the labor market
and that makes it more and more likely that these will be low ability
types. Our model reproduces the findings of the literature that
extended households interact less with the market than nuclear
households and that wages are lower in extended household economies
than in nuclear household economies without having any productivity
differences between the two economies. In other words productivity
differences are not necessary to produce the result that extended
household economies have lower market wages than nuclear household
economies. Thus differences in wages are probably the wrong diagnostic
with which to date divergence
{BOX 2.2 about here}
It is also worthwhile to recall that we built our model to maximize
the differences in labor market participation between the two
economies, so as to reproduce the conventional wisdom that extended
households are inimical to markets. It is time now to reconsider this
premise. Pre-industrial firms were overwhelmingly small farm and craft
enterprises, thus the assumption of firms with one entrepreneur and
one worker is reasonable. But the process of industrialization is one
in which larger and larger firms are created, thereby increasing the
size of the paid labor force and reducing differences among societies
with different types of households.
The argument that household structure and demographic regime were
responsible for either the divergence in economic fortunes between
China and Europe, or England’s early lead in industrialization is the
result of convenient induction. Because the household structure facts
fit the case, other elements like the role of markets were added
without significant examination. More than anything else there was
something attractive about a framework in which economic development
was produced by the meritorious and culturally, induced behavior of
northwest European households. Culture in these theses set the stage
in organizing households and then economic logic took over. Yet, as we
have shown, there are serious flaws in the chain of logic that runs
from culture to extended households to market participation and then
to growth. Neither Chinese culture nor stunted labor market prevented
the creation of a large industrial labor force around Shanghai in the
first third of the twentieth century, or all around China in the last
third. What is left is a pure cultural and historically circumscribed
thesis: Culture, one could argue, limited China’s labor markets and
growth in the pre-industrial period. That is a far cry from the
abstract generalizations of Malthus and many other scholars.
A less biased perspective might have led scholars to enlarge their
temporal view back in time, and ask why China was richer than Europe
for such a long time if people were so imprudent and why Europe was
poorer than China if its population was always so virtuous? Similarly,
why have societies with extended households throughout Asia been able
to have economies that perform so well today? Finally, recognizing
that not all households in China were large or extended, one would
have to wonder why heterogeneity in family structures persisted
despite the supposed intrinsic superiority of the nuclear household.
Any model that accounts for these facts will have to be more intricate
so as to provide some advantages to extended households. Indeed,
absent some countervailing advantages, extended households should
break apart. In such a model the differences between China and Europe
would be smaller and one would then be more likely to wonder whether
demography was all that important after all.
Our argument does not require us to bear the burden of accounting for
heterogeneity within China and Europe. Indeed we are only interested
in tracing the interaction of families and markets and in sustaining
the observed greater reliance of factor markets in Europe. Since we do
not seek to attribute either greater efficacy to one form of family
structure, or a permanent advantage to a society with greater markets,
we need not worry about our simplifications. Indeed, consistent with
the cultural and environmental variation that we know existed within
the two regions we would expect there to be variation in household
structure and in the prevalence of markets.
We can distinguish different kinds of relationships if we accept that
neither institutions nor culture determined a single outcome for China
or Europe. If we also accept to examine this diversity in a Coasian
light then we can expect to see the relationships between households
and markets to evolve over time as technology, relative prices and
transaction costs change. In fact we may well be faced with a perfect
wheel that turns from households to institutions and back to
households, without any ability to assert some clear causal importance
to household structures and demographic regimes on economic growth
possibilities. Clearly demography matters to labor markets, and
clearly labor market institutions will affect the decisions of
children as to whether to remain in a family enterprise or not, but it
seems there is considerable flexibility in these relationships both in
China and Europe.
What is clear is that neither region was locked into a particular
mode. Although in China (and especially in South China) the extended
household was popular and in Europe (and especially in northwest
Europe) kin groups were small and rarely co-resided, when
opportunities changed these social structures evolved. Consider some
European examples. Le Roy Ladurie famously examined frereches--kin
groups joined formally in a common enterprise--the prevalence of which
increased during a particularly difficult time in Languedoc
(1966:160-8). But extended households were not only found in difficult
times. Indeed, starting in the Middle Ages and through much of the
pre-industrial period, large families were key actors in Italian
politics where urban politics were family politics (Greif, 2006). Even
if households were nuclear in residence patterns, the larger kin group
was of great political relevance. Extended kin groups were also an
economically important unit; recall for instance, that the Medici were
bankers before they became princes. Their economic and political
successes were kin-based stories rather than that of any single
individual.
Nor do extended households simply mark a southern European
predisposition for informal institutions over formal ones. David
Sabean (1998) has documented the progressive rise of assortative
marriage and an increase in marriage among close kin in
eighteenth-century Germany as economic change made well-to-do heads of
households more concerned with keeping their assets within the family.
More generally, the persistently successful commercial banks of
Europe, like Barings, Mallet, or Rothschild prospered and endured
because they deployed the talent of more than a nuclear household.
Thus co-residence is not a requirement for economic behavior to
resemble that of an extended household. To be sure, there was entropy,
at each generation some individuals would move and abandon the
family’s traditional business. But entropy also arose in China,
because assets were often divided when the household head died (Lavely
and Wong 1992). In other words, in Europe even if the household was
nuclear, the size of the economic unit could be a much larger
kin-based group. This variation in kin group sizes makes the wage
information we have greatly relevant. A young man who joined a trading
firm—rather than strike out on his own—had to accept its discipline,
but he would realize higher earnings by combining his labor with the
experience of the other members of his kin group. His alternative was
selling his skills in the labor market. If his parents were poor or
inept the labor market was likely to be the more attractive option.
Conversely a parent could decide whether to keep his sons in the
family firm and face the risks that that entailed or to set them up in
other professions and hire employees instead. Wages thus mattered in
Europe because nothing required the firm to endure beyond its founder.
As we shall see in Chapter 3, Europe’s political fragmentation may
have been partly responsible for maintaining the abundance of nuclear
households working alone either in agriculture or in trade and crafts.
In Europe, Christianity eliminated Roman culture’s focus on ancestors
as an object of worship, but dynastic concerns continued to remain
important elements of European societies. In China, the persistent
importance of the cult of ancestors thus appears as a stark contrast.
However, to imagine households enduring for generations and to suppose
that the culture of kin was static is to ignore both considerable
historical evidence about the culture of kin and the remarkable rate
of internal migration within China (Lee and Wong 1991). Much of the
flexibility lay in the fact, that unlike in Europe, Chinese kin groups
generally, and lineages specifically, were typically much larger than
economic enterprises were. Hence what set of kin or lineage resources
were invested in what enterprise was a question just as relevant to
the Chinese case as the European one.
For Guangdong and Fujian provinces in south and southeast China,
lineage leaders often owned land, the rental income from which went to
maintain an ancestral hall and pay for expenses of lineage rituals
(Faure, 2007). Lineage leaders in Jiangnan sometimes set up charitable
estates composed of agricultural properties that were rented out, the
income from which went to support widows and at times other indigent
members of the lineage (Rankin 1986: 87-88). But more importantly for
our purposes were the instances in which kinship relations provided
individuals with a network from which they could choose people to join
them in economic activities. They could form firms based on the
intimate knowledge and trust embedded in their kinship relations.
These kinship networks could provide many people from whom business
partners could be selected. The kin network could also then become the
context within which problems in a business partnership could be
raised and resolved. For Taiwan, which was both administratively and
culturally part of Fujian in the eighteenth century, Joanna Meskill
has reconstructed the multiple trusts and estates associated with a
wealthy lineage, noting “Joint holdings and individual holdings, while
discrete were also interconnected. At times, estates or trusts
collaborated with wealthy individuals in income-producing ventures; at
other times, estates and trusts borrowed from each other or from
individuals.” (Meskill 1979: 245). The significance of kinship
principles to the formation of firms in southeast China is confirmed
by the use of fictive kin relations in the formation of firms engaged
in maritime trade (Ng 1983). As Teemu Ruskola has perceptively
observed, where the Euro-American legal tradition takes the legal
“person” as its key unit, the Chinese legal tradition used as its
units those based on kinship relationships (Ruskola 2000). We thus
expect that kinship relations will prove an important resource for
creating mechanisms required for economic growth.
Kinship relations played two related roles facilitating economic
growth in late imperial times. First, the kin group gave entrepreneurs
a pool of likely partners. Second, kin relations provided a context
for resolving economic disputes; a particularly important matter in
the context of long-distance trade. We suggest that kinship practices
in late imperial China offered opportunities to form firms and to
adjudicate economic disputes without recourse to state-operated courts
and laws. These kinship practices were further complemented by the
activities of native place and occupational associations. All these
institutions were accepted by the state which developed laws and
courts to a lesser degree than in Europe. The difference between China
and Europe is not, however, as stark as a focus on the household as
firm might suggest, as we will see in chapter 3.
When thinking about the deployment of lineage resources, one might
well imagine that the relative price of land to labor is irrelevant.
This is because the cultural predisposition of lineages to deploy
their resources internally rather than through the market would imply
that there is a relevant relative productivity of labor to land for
each lineage rather than one at the aggregate of the economy. Evidence
in favor of this would be the infrequent purchase or sale of land
outside the lineage. Precisely because lineages were large, the sale
of assets would be less frequent (than in the case of nuclear
households). That does not imply, however, that lineage heads did not
have to evaluate the opportunity cost of keeping a piece of land in
the lineage rather than using the same resources in some other way
(increasing investments in other land for example). If we consider
labor, married males may have been rather unlikely to leave the kin
group and its resources. Yet the unbalanced sex ratios that prevailed
in China created a group of men whose opportunities within the lineage
or kin group were limited because they would never marry. For such men
the larger kin group of extended family or lineage offered few
advantages. The growing amount of wage data over time implies that
poor unmarried men were unable to benefit economically from extended
kin relationships. Thus, while it is entirely possible that, on
average, individuals engaged in little wage work there existed a very
real group of Chinese men for whom the calculus was similar to that
employed by Europeans.
Conclusion
This chapter has introduced our comparative approach. It has taken a
set of apparently radical differences between Europe and China and
argued that these differences are not very important. To do so we have
mixed evidence and economic theory. In this case theory has been
largely negative, in that the frameworks seek to argue against the
notion that differences wages between China and Europe were
necessarily related to both household structure and to the
productivity of the economy. In developing the theory we have been
forced to confront a second key element: variation within Europe and
China overtime. We determined that such variation was problematic for
any theory that bases Europe’s advantage in the nuclear family. Rather
than reject any theoretical argument we decided to follow the
implications of the simplest model one could build. This leads us to
two important conclusions.
First, although differences in household structure may matter to the
size of factor markets, they do not necessarily have any effect on how
many households rely on these markets. Nor do they necessarily imply
that differences in markets have implications for long-run growth. In
fact, just as we now reject the stereotype of an Asia ruled by a
Maltusian positive check, we must also be careful to imbed the
information that we receive about prices and wages into the
institutional context that produces them.
The second conclusion is that rather than considering the family group
and its attendant economic unit as coincident and closed, we should
consider that both are in steady interaction with markets. How much
they interact with markets does depend on transactions costs. We then
examined the consequences of the fact that in many economic activities
in Europe the family group was smaller than the economic unit engaged
in production, while in China it was typically larger. Again relying
on Coase’s insight we argue that it follows that the volume of trade
in factor markets was likely to be larger in Europe than in China, but
it does not follow that there would be efficiency consequences to this
difference. Indeed all heads of households had profound interest in
paying attention to factor prices in making decision about how to
deploy their resources. To suggest that the Chinese equilibrium is
less efficient has about as much empirical content as to suggest that
the modern integrated corporation is less efficient than tiny single
activity firms.
The next chapter turns away from families and relative prices to
consider differences in commercial institutions. Given the importance
we have given to markets, it is pertinent to examine the contracting
environment. This will give us a first opportunity to consider the
consequences of different spatial scales on the economy, and to
evaluate an argument about China’s relative failure that is nearly as
prevalent as the demographic one: institutional lock in. In doing so
we will discover again that persistent differences in institutions are
necessary but not sufficient conditions for divergence.
Box 2.1
Assume: each firm/farm needs one entrepreneur and one worker.
Individuals can be either workers (W) or entrepreneurs (E). A firm is
profitable if and only if it is run by an entrepreneur and hires one
worker. Capital markets are perfect so we can ignore the other inputs
into the firm. When families are nuclear each family has one adult and
he/she must decide whether to be an entrepreneur or a worker. In a
lineage system the leader decides whether to form firms (and which
relatives to hire as entrepreneurs, including potentially himself)
whom to hire as workers, and whom to send out to earn wages.
Throughout we assume that the probability of being type E is a half.
Nuclear households (1 adult). E individuals become entrepreneurs and
they each hire one W adult. Half the population earns wages in someone
else’s firm.
Extended Households are simply larger families. Consider a society
where each household has two adults (the smallest possible case). The
household can be of 4 types: (E, E), (E,W), (W,E) or (W,W). As in the
case of nuclear households it pays to start a firm for each E adult.
Thus some households (E,E) have two firms, some will only have one
(E,W, and W,E), and some (e,e) will have none. As in the nuclear
household case, half the population will be managers and half will be
workers. Yet because (E,W) and (W,E) households satisfy their labor
demand internally their workers are not in the labor market and
‘receive’ no wages. In fact wage earning workers (adults from (W,W)
households) only amount to one quarter of the population.
The extension of the model to households with more than two members is
straightforward and firm with more than two members.3 It is easy to
show (1) if society 1 has households of size n1 and society 2 has
households of size n2 then if n1>n2 the share of the workers who are
in the paid labor force is less in society two. (2) The fraction of
households who either hire or send out workers is larger in society 2
if n1>1. (3) Let f be the number of workers needed in a firm, while
the qualitative difference between society n1 and society n2 will hold
for any f, differences among societies shrink. Figure 2.1 reproduces
graphically the results for households size 1 to 10.
Box 2.2
The model is an extension of the one in box 2.1. An adult can be
either E or W—that is he has either high (W) or low (w) ability as a
worker. As a result individuals can be one of three types {E, W, w}.
Half the population is E, one quarter W and one quarter w.
One adult (Nuclear) households: including worker skills does not
change the analysis: indeed all workers work for wages. Note that the
average skills of wage workers are the same as that of the population.
Two adult households: Wage workers are a quarter of the population and
their average skills are the same as that of the population (because
they all come from households with two workers (WW, wW, Ww, ww)
Three or more adult households: Consider first three adult households.
As in the previous analysis, half the population will be entrepreneurs
and half workers, and half the workers will be employed in their
family firms. Because the number of adults is odd all households have
either an excess of entrepreneurs or workers. For those that have an
excess of workers the question is whom do they send out to earn wages.
Households that send workers into the labor market have either one
entrepreneur or none. If they have none they will send all their
members into the labor market. Thus the average skills of the workers
of these lineages will be the same as that of the general population.
This is also true in lineages with one entrepreneur and two workers
where the workers have the same skills. But now consider a household
in which the two workers have different skills (e.g. (E, W, w)). It is
reasonable to assume that household members have better knowledge of
each other’s ability than the ability of individual hired in the labor
market. For simplicity assume that the E member gets to decide which
of his two relatives to hire and can pay whatever wage he wants, while
the labor market cannot differentiate between high and low ability (W
and w) for new workers and thus their wage will reflect average
ability. Clearly then the household can pay its high ability worker
more than the market while the market will pay the low ability more
than the household would. Hence whenever households have choices they
systematically send out low skill workers. Thus although average
skills in the three adult household society are identical to those of
the two adult household averages skills in the labor market are lower.
Indeed, one can pursue this analysis and show that the selection
effect increases as households get larger (for any household size n,
the selection effect is smaller in economies with larger firms than
where firms are smaller). If households are on average larger in China
we would expect labor market wages to be lower than Europe independent
of productivity.
Chapter 3
Formal and Informal Mechanisms for Market Development
Introduction
The study of market institutions is a central endeavor of economics.
Understanding how people structure relationships in which one party
extends credit (henceforth time contracts) to the other has sparked a
particularly abundant literature. One key lesson from that research is
that not all exchanges can be supported by formal contracts. Some
market transactions are too trivial to contract upon or to sue after
non-performance. Others involve dimensions of performance that third
parties cannot observe. In this case, informal means and, in
particular, reputation and repeated interaction serve to sustain
markets and their implicit credit relationships. Because the size,
frequency and complexity of deals vary from transaction to
transaction, a division of labor between formal and informal
contracting arises within economies. Some transactions are supported
by informal means and others by formal ones. When it comes to
comparative economics, many have found that such a division of labor
occurs at the social level: in some places most, if not all, exchanges
are informal; in others contracts and courts play a central role. It
is also commonly argued that differences in early history can have
large and persistent effects on the types of transactions that
ultimately prevail (e.g. Greif 2006, Hoff and Stiglitz 2004, Tabelini
2008). Scholars thus seek to classify societies as either collectivist
(dependent on informal institutions) or individualist (dependent on
formal institutions).
Yet the agreement to classify societies hides some serious tension as
to what institutions promote growth. Western Europe’s success is often
attributed to the capacity of commercial elites to wean themselves
from reliance upon networks, while Islamic and other Middle Eastern
societies failed to do so (Greif 2006 pp 269-301, Kuran 2003, 2004).
In contrast students of Asia have often ascribed the success of these
economies in more recent times to the wondrous flexibility and
ubiquity of informal networks (CITATION). Formal institutions are
claimed crucial in one set of cases and informal ones in a second set.
Most scholars would be willing to concede that individuals in China
and Europe have been deeply involved in market transactions for
centuries. Yet borrowing from the comparative economics literature
many have seen Qing China as failing to develop the legal
infrastructure to sustain formal contracting. In contrast, European
states (and in particular the Dutch and the English) developed a law
of property and contracts that facilitated exchanged. China’ did not
do so because the state failed to supply these institutions and the
extended households and lineages had little demand for them. Such
conclusions, however, face problems similar to those uncovered in
Chapter 2 regarding the argument that differences in household
structure were responsible for differences in economic performance.
Scholars who extol the value of networks tend to focus on
long-distance trade, while those who favor formal enforcement tend to
examine real estate transactions, or local credit. As we shall see the
selection of evidence is largely responsible for the arguments that
there were large structural difference in contract enforcement between
Europe and China.
This chapter argues that these differences are real but not
structural. We will show that the extent to which individuals in China
and Europe used either formal or informal means of enforcing contracts
depended on the nature of the transaction. The observed differences in
the types of enforcement deployed in China and Europe are thus the
product of differences in the economic environment, in particular the
scale of long-distance trade. The variation in institutions will be
largest when the two economies are the most different in economic
structure and spatial scale but it will shrink as the structure of the
economies become more similar.
Our revision of the prevailing contrast between East and West is
possible because of significant recent contributions by scholars of
China and Europe. For Europe, Avner Greif’s work on informal
institutions has given a new perspective on the sources of European
growth (Greif 2006). It has spawned renewed interest in private order
mechanisms as alternatives to state-based enforcement. Starting with
Shiga Shuzo (2002) and Kishimoto Mio (2007) in Japan, scholars of
China are discovering a rich formal contracting sphere. More recently,
American scholars such as Madeleine Zelin (Zelin et al 2004), and
Melissa Macauley (1998) and Chinese scholars such as Liang Zhiping
(1996), have shown us that written contracts underlay the exchange of
assets as diverse as land and equity in businesses and that
magistrates intervened to resolve disputes. In both the European and
Chinese cases, recent work reacts in part to a very large
historiography whose primary effort has been to document the existence
of a formal bias in Europe and an informal bias in China. We can now
see that this dichotomy is far too simple to describe the interplay of
formal and informal mechanisms actually at work historically in China
and Europe.
In order to re-examine Chinese and European contracting institutions,
it is useful to take a moment to clarify what we mean by “formal” and
“informal” enforcement mechanisms and highlight the distinction
between the two. Put simply, "formal" ways of enforcing agreements
rely on government officials (e.g. judges) to decide over disputed
points and impose coercive or financial penalties when contracts are
broken. "Informal" mechanisms, in contrast, require that private
parties decide when contracts have been broken and what penalties to
exact, whether that means shunning offending parties or other
sanctions. It is common to suppose that choosing between these formal
and informal mechanisms requires assessing the tradeoff between the
cost of enforcement and the losses associated with limiting the set of
potential partners. Formal enforcement offers the broadest set of
potential partners, but it is costly, especially when transactions
occur at a distance, since one has to be willing to go to court to
settle disputes. Meanwhile formal enforcement limits the set of
potential partners to members of a group, but enforcement costs are
potentially trivial, as long as the duration of the transaction is
limited.
This chapter proceeds in a manner slightly different from what we
followed in Chapter 2. We begin by examining the literature on long
distance trade, and verifying that once we chose to compare similar
activities there are fewer differences between China and Europe than
one might have supposed. Based on this conclusion, we propose a
framework for analyzing contract enforcement across types of
transactions. This framework allows us to recast a much broader set of
evidence and argue that individuals relied on formal and informal
enforcement both in China and in Europe. We then show that part of the
differences between China and Europe came from differences in the
scale of long distance trade. The last section of the chapter argues
that the extent of reliance on formal and informal enforcement varies
over time. While there may have been and there may be societies that
are locked into informal enforcement this was certainly not the case
for most European polities or for the Chinese empire.
Lessons from Long Distance Trade
Consider long distance trade, which for the preindustrial era we
define as exchanges of goods where buyers lived 200 kilometers or more
from sellers. Such commerce would have included trades between a
foreign merchant and local consumers, those among merchants at trade
fairs, or the early days of inter-regional exchanges. Initially,
through the first millennium AD in Northern Europe and many other
parts of the world, these exchanges were infrequent and time consuming
as they involved someone travelling several days, if not weeks, in
each direction. Suppose, for instance, a merchant arrived in a town
after a long journey with a load of sugar. He faced two choices: he
could sell his sugar for cash or he could give credit to the buyer. As
many have noted, a commercial system based on cash is going to be much
smaller than one based on credit because it requires a coincidence of
wants on all sides. In our case the sugar merchant must arrive when
there is an accumulation of export goods of equivalent value that he
is interested in purchasing. Extending credit would allow the merchant
to unload his sugar and then seek out the return cargo that might
offer him a better prospect for profit in nearby towns. Yet credit
would be extended only if the lender (in some cases the traveling
merchant and in others the local producer) could expect to be repaid.
To the extent that the lender had to appear in court, distance made
formal enforcement expensive, and if the distance was great enough,
downright unprofitable. Even if one were able to hire a local agent,
distance still raised the cost of relying upon courts, to the extent
that few long distance merchants were prepared to use them. Meanwhile
if the lender wanted to employ informal means to enforce repayment,
the best he could do would be to refuse to have further relationship
with any recalcitrant borrower. As long as interactions were
infrequent and small scale, such threats would have been as hollow as
going to court. It is no surprise then that the early commercial
system was based on cash or barter. Hence we know that itinerant
peddlers were typically paid immediately by their clients, merchants
who met at European trade fairs early on could not carry balances from
one fair to the next, and Europeans traveling to the coast of China
brought silver to pay for their purchases. In general, when
transactions were both distant and infrequent, buyers were not
extended credit. This did not mean that such credit was not desired,
it was simply that there were no mechanisms to support this kind of
lending. As long as this state of affairs persisted trade remained
limited.
Societies, however, have devised a variety of methods to turn distant
and infrequent transactions into either local or frequent ones. The
most pervasive was to organize long-distance trade within networks. In
that situation, credit was extended within groups who shared ties of
either family or geographical origin and who interacted frequently.
This was true both in China and Europe. These networks did not rely
heavily on courts. Instead, their members respected their obligations
because this was required for continued membership in their networks.
Dating back to the sixteenth century, Chinese merchant networks were
large and widespread. The most famous was comprised of merchants from
Huizhou in Anhui province wo were major actors in the Jiangnan textile
trade. These merchants bought cotton cloth produced by rural
households at local markets and took the cloth to be dyed and finished
in nearby market towns by businesses also run by Huizhou merchants.
Other Huizhou merchants controlled the wharf from which many of the
textiles were shipped to other parts of the empire. While Huizhou
merchants were engaged in a variety of trades and were located in many
parts of the empire, other merchant groups had more limited
inter-regional routes. Merchants from the southeast coastal province
of Fujian, for example, established businesses in the Jiangnan region
either to export textiles and other products from Jiangnan back to
Fujian or to import Fujian goods into Jiangnan. Regardless of the
differences in spatial scale or range of goods traded, the same basic
principles of trade among merchants sharing some combination of native
place and kinship ties applied (Fan Jinmin 1998: 185-206).
The European research tends to highlight the diversity of informal
institutions and to focus on a comparative analysis of their relative
efficiency. Yet, from our perspective, the striking fact is that
informal institutions and merchant networks were at the core of most
long-distance trade in pre industrial Europe. Whether one considers
seventh-century Maghrebi traders in the Eastern Mediterranean (Greif
1989) or the family trading firms established in the following
centuries throughout Italy (De Roover 1963, Hunt, 1994, Braudel 1966,
Drelichman and Voth, 2009, Muller 1997, part 3) and subsequently
throughout Europe (Ehrenberg 1922, De Roover 1948, Neal and Quinn
2005, Trivelato 2009, Gelderblom 2010), one finds informal enforcement
mechanisms. The same is true for the Protestant and Jewish commercial
and banking houses of the early modern period (Lüthy 1959-61,
Trivelato 2009, Moulinas 1981). Finally, informal mechanisms were
critical to the success of the family banks that linked cities in
Europe in the eighteenth and nineteenth centuries, the most famous of
these being the Rothschild banking family (Ferguson DATE).
The European literature has tended to make much of the importance of
political boundaries. In the Middle Ages at least, European political
fragmentation meant that long-distance trade was always, in effect,
international trade. For many subjects the next polity was no more
than a couple days’ walk. Rulers and urban elites were suspected of
discriminating against foreign merchants, yet discerning where
jurisdiction lay for a contract between two parties, both of whom were
foreign to the place in which they made an agreement, was not always
clear. For instance, jurisdiction for a debt contracted in Antwerp by
a Parisian with a Lisbon merchant was very much uncertain. Here China
holds a lesson for Europe: political boundaries may be less important
than sheer distance. Most Chinese merchants carried out the entirety
of their business within the confines of their empire, and they could
nominally have relied on imperial administrators to settle disputes.
But they, like their European and ‘overseas’ Chinese counterparts,
preferred to remain in the informal realm. The reason is not that the
empire failed to provide an appropriate institutional structure;
rather courts are just not very efficient at enforcement contracts
over long distances.
Clearly long distance trade had an abiding affinity for informal
networks, and financial capital (that most modern of enterprises)
continued this tradition into the modern age. What is important for us
is that this affinity seems to have little to do with any particular
culture since we observe it nearly everywhere and, in particular, both
in China and Europe. This observation alone raises serious questions
about the usefulness of recent analyses that emphasize differences
across societies in the level of formality. If Europeans were so
formal why did they rely so extensively on networks and reputation in
trade?
A Model of competition between enforcement mechanisms
The next step in our analysis is to develop a framework for examining
the issue of contract enforcement that is sensitive to the type of
transaction at stake. We build upon the simple but powerful insight
from long distance trade: individuals choose their enforcement
mechanisms depending on what is available. To be fair to the
literature, this model must encapsulate both the purported advantages
of formal and informal mechanisms. We start with contracts and trade.
Trade allows an individual to sell a good or service that he has in
relative abundance for something he desires more. This process leads
to an increase in aggregate welfare. At the same time each party is
well aware that his counterpart will seek to discharge his obligations
at the least possible cost. Many individuals will be tempted to
deliver goods of low quality, slouch in the delivery of services,
delay payment, or better yet refuse it completely. In each such
instance the value of the transaction to the individual’s counterpart
falls: exchange is beset by transaction losses. If these losses are
not brought under control, trade may cease entirely. Henceforth, and
for simplicity, when a party misbehaves we will say that he cheats, if
not that he performs. Although issues of performance are far more
extensive than just in credit transactions, those will serve as our
guiding example. There, default, and in particular default that occurs
as a result of actions or inactions of the borrower, is what the
lender wishes to minimize. Once the loan has been made, the lender’s
profits fall directly as the default rate increases.
To reduce losses from default, individuals invest in information and
expertise to determine the quality of the items being exchanged. Time
transactions also require further institutions because although the
buyer/borrower can observe what he receives today, the lender/seller
does not know exactly what he will get in return. For such contracts,
institutions that provide punishments for individuals who fail to
perform are critical. Why should a lender bother to determine why a
borrower fails to repay, unless he has the means to punish the debtor
who has engaged in fraud? Information and enforcement are thus
complementary for time transactions because investment in information
only makes sense if that information can be acted upon. Effective
action requires both institutions to detect miscreants and to punish
their misbehavior.
How do formal and informal institutions deter cheating? Informal
institutions rely on reputation and other private sanctions. The
parties to exchange have incentives to perform, because good past
performance is a pre-condition to being able to engage in future
transactions with members of a reputational coalition. The coalition
contains all the individuals who restrict interaction (exchange) to
members in good standing of the coalition. The coalition could involve
an ethnic minority as in Greif (1989) or a lineage group or
individuals of a given place of origin as was common in China (CITATION).
As game theorists have shown, cheating is most easily deterred when
information flows are good, when alternative occupations are
unrewarding and when individuals are patient.
These conditions are highly intuitive: if information is poor then the
one party cannot decide whether the other has cheated or not; thus
dampening the effectiveness of exclusion. If individuals who misbehave
can find other ways to secure income, then the threat of exclusion has
little bite. This last condition suggests that coalitions that govern
exchange not just of one but of a large number of commodities and
include many people are more powerful than those that involve just one
type of transaction across a small group. Enlarging the coalition,
however, increases information costs and nearly all examples that we
have of groups that engage in reputational behavior are subsets of the
general population. Thus coalitions have costs because trade must be
restricted and hence a member may not obtain the best possible price
for his item because the individual with the greatest willingness to
pay may not be a member of his or her coalition. These costs will be
highest when groups are small and goods highly heterogeneous. Finally,
if individuals are very impatient they will want to enjoy the ill
gotten gains from cheating rather than wait for the more virtuous
return of future transactions that follow good performance. Impatience
is not simply a characteristic of the individual it is also a
characteristic of transactions. If one engages in a particular kind of
transactions (say real estate purchases) sufficiently infrequently the
cost of exclusion from a future transaction will be outweighed by the
immediate gain from cheating.
In the case of formal enforcement, the incentive not to cheat comes
from avoiding a punishment that would be meted out coercively by an
agent of the state. Again, would be cheaters are deterred by the fear
of prison, fines or the damages the courts will force them to pay if
they misbehave. There are costs and benefits to formal enforcement as
well. Perhaps the biggest of these involve the setting up of courts
and legal system. There are also costs associated with the
adjudication of specific disputes. These costs will depend on a
variety of factors, notably where the case will be tried. If the
amount lost is sufficiently small, a cheater may not fear a suit
simply because the costs of litigation outweigh whatever can be
recovered. Moreover, pursuing redress in a court will depend on where
a litigant has standing to sue someone who has cheated him. This could
be where the contract was signed, where the absconder lives, or in
some third location specified in the contract. The further away the
cheater and his assets are from the plaintiff, the more expensive the
case is likely to be. The benefits of using a formal mechanism are
that it does not depend on the identities of the parties to a
transaction. This does not imply that transactions are anonymous: the
parties to a given trade must still know a lot about each other—but
their capacity to sanction a defaulter does not depend on the
particulars of an ongoing relationship.
To build our simple model we reduce the set of factors that affect the
relative efficacy of formal and informal mechanisms to two: frequency
and distance. As in Chapter 2 the interested reader can follow the
mathematical analysis in Box 3.1. Given that the incentives to remain
honest decline as transactions are geographically more spread out,
while the incentive to cheat is immediate, the argument comes down to
one simple rule: if the interval between transactions is too long,
then these types of transactions cannot be sustained by an informal
mechanism. Denote by T* the largest time interval between transactions
such that reputation sustains performance. When the expected interval
between transactions rises beyond T*, informal enforcement will fail.
This stark result echoes much of the diaspora and social capital
literature that argues that social networks play a critical role in
supporting trade, and that they are also dependent upon dense
interactions. Similarly the cost of punishing a defaulting borrower by
taking him to court increases with distance even though what the
lender can recover does not. Again this leads to simple argument: if
the borrower lives too far from the lender, then formal contracts
cannot sustain trade. Denote by D* the largest distance between two
parties such that it is worthwhile to sue in court if someone cheats.
We further assume that D and T are not systematically related—that is,
there are infrequent transactions among neighbors such as real estate
sales as well as frequent transactions among neighbors (as in the
market for occasional labor).
[Box 2.1 about here]
Supposing for now that individuals decide to use only one type of
mechanism to enforce one type of transaction (e.g. the market for
livestock) and that they could select whatever mechanism they wanted
for any particular type of transaction, which would they chose? Given
the foregoing formulation of issues, there are four possible cases, as
shown by the four different regions in Figure 1.
1. Distant and Rare Transactions (the upper right hand quadrant in
Table 1)
Because D>D* and T>T*, neither mechanism will work and trade in such
conditions will be in cash.
2. Distant but Frequent Transactions (the lower right hand quadrant in
Table 1)
Because D>D* formal enforcement is not feasible but because T informal enforcement is feasible.
3. Local but Rare Transactions (the upper left hand quadrant in Table
1)
Because T>T* informal enforcement is not feasible but because D formal enforcement is feasible.
4. Local and Frequent Transactions (the lower left hand quadrant in
Table 1)
Because D feasible; hence our theory does not decide the issue. For the moment
we will leave this case aside.
[Table 3.1 about here]
The framework has some immediate implications for thinking about the
historical record.
*
Societies that are wealthy should have both formal and informal
institutions. Indeed prosperous societies will enforce contracts
for infrequently traded assets like land and they will engage in
long distance trade. Since both China and Europe have been
economically prosperous at various points in their histories, we
would expect both formal and informal institutions to flourish in
both places.
*
Comparable transactions should be enforced with similar
mechanisms, both in China and in Europe. It is unlikely that one
will find a society where the market for land rests on reputation
while commercial finance relies on state enforcement and another
where the reverse is true.
*
If, as is commonly assumed, Chinese informal networks work better
than European ones but European courts work better than Chinese
ones then some transactions enforced informally in China will be
enforced formally in Europe but that range will be small. The
dominant effects would be that the geographical reach of courts in
Europe would be larger than in China (so more low frequency
transactions might occur with credit in Europe than in China). At
the same time European networks would fail to sustain some distant
transactions because the frequency of such interaction would be
too low. (see Table 3.2)
*
Finally, if one economy has more long-distance trade, and
consequently evolves more efficient informal mechanisms, then it
may well appear at a point in time that this economy is more
‘informal’ relative to another which has less long-distance trade
and uses formal mechanisms more. The relative prevalence of
different mechanisms in any given society should evolve along with
changes in economic structure. Thus, finding informal enforcement
in China and formal enforcement in Europe is not sufficient to
tell us how the regions evolved. Instead one should look for
similar activities that rely on different mechanisms in the two
regions.
The foregoing conclusions depend upon a critical assumption: rulers
supply an adequate level of formal enforcement and do not intervene to
make courts inefficient, or to disrupt trade networks. While we
contend that cultural and social differences are unlikely to alter the
arguments made above, we are well aware that political constraints
could play an important role. This consideration clearly requires
great care at two levels. First, we must ascertain that China’s
informality was not simply the result of imperial neglect or
oppression (as has been suggested (CITATION). Second, the foregoing
argument probably cannot be exported to any and all settings or to all
populations. The mere fact that there is a functional logic to develop
formal mechanisms doesn’t mean rulers have the requisite capacity or
proclivities to act accordingly. For example, as we saw in Chapter 1,
during the eleventh and twelfth centuries European rulers more often
than not failed to provide formal institutions (Bisson 2009).
Our interest here involves China and Europe starting after 1400. We
must now ask, what were the political constraints on institutional
choices? Political constraints take a variety of forms. At its
simplest the Chinese empire’s very size encouraged the formation of
networks of long-distance traders whose volume of activity was, for
centuries, far larger that what occurred in war-torn and fragmented
Europe. Looking not at the fifteenth but the eighteenth century, one
might believe that the empire failed to put in place a court system
capable of providing formal enforcement to all comers. Conversely
European countries’ tiny medieval size may have created segmented
markets whose transactions were comparatively easy for courts to
enforce. Moreover, war may well have made it difficult to sustain
reputational networks on a large scale. Thus, Europe may have been
politically pushed into a more formal equilibrium than was efficient.
The rest of this chapter is devoted to presenting evidence that both
formal and informal mechanisms mattered in China and Europe; that
their distribution in the economy can be explained by a common logic;
and finally that political factors were important in shaping the
boundaries between formal and informal enforcement. Moreover, it is
likely that the distribution of labor between formal and informal
institutions that prevailed in one place at one point in time (say
Medieval Europe) would not have been as practical in other places and
other time (say Qing China or nineteenth century Europe).
China and Europe: Similarities and Differences
Our earlier discussion of long distance trade emphasized that both in
Europe and China such commerce was carried out either, in cash, in the
case of distant and infrequent transactions, (the upper right cell in
Table 3.1), or with credit through informal networks as distant but
frequent interactions (the lower right hand corner). Beyond a certain
distance, that we put at 200 kilometers simply for illustrative
purposes, courts could not effectively enforce contracts. It is not
that courts did not perceive international trade as an attractive
venue to sell services. On the contrary, they made every effort to
attract business. By 1600 Low Country courts, for instance, promised
to judge disputes among foreigners according to the law of the place
where the contract had been signed. Yet this provision benefited two
merchants from Venice who happened to be in Amsterdam, not a merchant
in Venice who wanted to recover from a Dutch counterparty (Gelderblom
2010 Chs 7-8).
What then of interregional trade? In the case of very long-distance
trade, the need to settle accounts voyage by voyage explains the fact
that every ship that left Europe was laden with silver to settle its
accounts on foreign shores. It would have been no different were the
ships Chinese ones going to Europe with goods. In either case cash
relationships were required irrespective of the location at which two
parties of such different identities transacted their business.
Nevertheless, the evidence of institutional change is clear. On the
European side, both the Dutch and the English East India trade evolved
from clubs of investors that funded a given ship’s voyage to joint
stock companies that regularly sent out ships and established
permanent bases in the East (Gelderblom and Jonkers 2004, Harris
2005). Over the eighteenth century, as transactions became more
frequent between the same parties, credit arrangements became more
common. As in the other forms of long-distance trade, inter-regional
trade moved into the lower right-hand box of our table. Various forms
of credit were extended in Canton between the Chinese and foreign
merchants, as well as among foreign merchants themselves. Brokers and
agents in Canton and Macao rented space on ships, purchased goods and
arranged for their sale (van Dyke 2005: 150-59). The China trade does
tell us that we should develop flexible notions of distance and
frequency that accommodate changing institutional forms that were far
from fixed over time and space. Among new commercial ventures, those
that were profitable tended to become more frequent simply because it
paid for merchants to invest more in such voyages. As these markets
expanded in scale, they allowed the establishment of informal
mechanisms for time contracts when none had been feasible before.
We turn now to a third category of exchanges identified in the upper
left hand corner of Table 1. These are infrequent transactions for
which formal enforcement is possible. Contracts over land are the
prime example of such a situation.
In Europe, as is well known, enforcement of local long-term contracts
was a key element of local justice. This was true in the Roman law
prevalent in the south as well as in the common or customary law of
the northern areas of the region. It was true in the countryside where
at first local lords, and later royal officials provided judicial
services. It was also true in urban areas where political authorities
often delegated the tasks of resolving disputes to merchant groups
(guilds). Nevertheless the enforcement of the decision of guild
officials relied upon public officials (Epstein, Ogilvie). Over time
practices across the many discrete local jurisdictions were harmonized
through the processes of centralizing state formation, but the
principle of a judicial system that was close at hand persisted.
Moreover the bulk of the activities of local courts involved the
adjudication of economic disputes (Duby, 1979, 1973). While it is not
possible to estimate the value to the private economy of formal
contract enforcement, it is clear that recourse to the courts was
widespread from an early date in the medieval economy, in particular
when disputes related to land, long-term credit or labor arrangements.
In fact, when European economic historians want to trace out the rise
of formal institutions they often turn to contracts over land and
describe the evolution of tenure from a feudal system, in which
individual claims were only known locally and largely enforced by a
local lord’s thugs, to one in which a national justice system enforced
titles to real assets (North and Thomas 1971, Campbell 2006). In many
places titles and rental contracts were secured by registration in
public information systems (Hoffman et al 2001, Gelderblom 2010 Ch 8).
Moreover, parallel registrations systems allowed lenders to learn what
liens had been placed on a particular piece of property. Thus
formality and publicity were central to European conceptions of the
organization of the land market and its attendant credit market. They
are also central to the standard narrative of Europe’s success.
While most students of Chinese history already know that land was sold
and rented in much of early modern Europe, many students of European
history may not realize that agricultural land in late imperial China
was also typically held as private property. While there were regions
where the imperial government interfered with the market for land so
as to better control its supply of troops for duty on the frontier
(Lee and Campbell 1997), such cases seem more an exception than the
rule. In the last two decades historians have unearthed a vast
documentation trove of private contracts. They show that in general,
title to land was a matter of written record and transfers of land
involved written documents (Citations). In fact, by far the most
common forms of contract to survive from imperial China concern land
transactions. While writing down transactions is a first step in
creating some formality in the market, one might well do so even if
the ultimate enforcement mechanism involves reputation. Indeed, a
written agreement detailing the transaction and witnessed by multiple
parties might reduce the likelihood of disputes compared to purely
oral or private arrangements. Yet the documents are often also stamped
by the local magistrate, suggesting that individuals were doing more
than simply writing things down.
The crux of the matter involves the enforcement of these contracts. In
the eighteenth century, when abundant local archival materials begin
to survive in China, they include legal cases heard by county
magistrates. These show that one of the four most frequent categories
of dispute brought before these magistrates concerned land
transactions (the others involved debts, marriage disputes and
inheritance). These were typically between neighbors or kin. In some
instances these disputes could continue over generations. Both the
duration of some disputes and the more general indications of how long
a given piece of land had been in the possession of a household and
its ancestors suggest that public authorities were centrally involved
in securing land assets. Further, they show that private property
rights in land were well established in many parts of late imperial
China.
The enforcement of local property rights by formal mechanisms and
long-distance trade relying upon informal mechanisms is common to both
China and Europe. It is important for us to stress this baseline of
similarities because scholars have been all too eager to point out the
differences. In transactions where one kind of mechanism is clearly
better than the other (distant but frequent and close but infrequent),
China and Europe look alike. What about transactions where one could
rely on either formal or informal mechanisms to sustain trade?
High frequency local transactions
Transactions that can be sustained with either formal or informal
types of enforcement are high frequency local transactions that occupy
the lower left-hand cell of Table 1. These include transactions
between local producers and resident merchants, between resident
merchants and local consumers, and between local creditors and local
borrowers when credit is short-term. Our simple model is agnostic
about what type of mechanisms will be chosen. The conventional
approach is to phrase the question as if the two mechanisms were
mutually exclusive. Scholars have provide both logical and cultural
grounds for this either/or approach. On the logical side, some have
argued that when a contract is broken the injured party will try to
get redress at the lowest possible cost. Hence, the logic goes, if in
China reputation is ‘cheaper’ than courts, then individuals will cease
to use courts. The cultural argument has been taken by others as
equally powerful in supporting a strict separation between formal and
informal mechanisms: in some societies where reputation is important,
someone who tries to have recourse to courts will be tarred, while in
places where courts are broadly available reputation is of little
value. Certainly these approaches are appealing because they provide
the fuel for theories of divergence between a dynamic Europe creating
anonymous formal markets and a static China mired in its informality.
In fact, some have gone so far as to suggest that the introduction of
Western institutions for economic transactions in the late-nineteenth
and early-twentieth centuries was necessary for modern economic growth
in China. Seen this way the historical divergence between the Chinese
and European economies was the result of a cultural-institutional lock
in (Debin Ma, 2006, Mu Li 2002)
Yet, although these arguments are seductive, they do not stand up to
the evidence. Simply put, rather than being mutually exclusive,
reputation and formal enforcement were deployed in conjunction. In
Europe, where the research on local markets has been abundant, the
evidence is compelling. It shows that although the role of reputation
was more important in some transactions than others (say commercial
credit rather than mortgages), it was never irrelevant. Similarly,
while individuals were more likely to engage in litigation over land
than over a basket of fruit or some small debt, willingness to sue
over minor matters was remarkably high (e.g. for Burgundy, see Brennan
1997, Hayhoe 1988). That is why we have records of millions of court
cases and registered contracts. One could conceive of these issues in
many ways; the simplest is detailed in Box 3.2. Each actual
transaction is either an informal or formal transaction depending on
the relative value of interacting within a network or with strangers.
Within the network enforcement is informal and free but counterparties
are few in number. Outside the network, there are more counterparties,
thus one can expect a better match but enforcement is costly. Because
conditions vary by type of transaction, by types of individuals, and
by the circumstances when the transaction is contemplated, individuals
interact regularly with both types of partners and rely on both types
of enforcement. Although from the outset a transaction is either
informal or formal, transactions that look identical can be enforced
in different ways, depending on the identities of the buyer and
seller.
This kind of logic applied in the past. Going back to the medieval
period, a central feature of European society has been a judiciary
that from early on made its services available to nearly everyone
(though slaves did not have the right to sue, manorial courts settled
disputes involving serfs). Initially these very local courts were
under the authority of local lords who either dispensed justice
themselves or appointed whomever they pleased. In Western Europe, at
least, the combination of the decline of serfdom and the growth of the
power of regional lords, cities, or even kings, led to a
professionalization of the judicial system and its progressive
centralization. In economic matters, centralization was limited in
many places because rulers found it expedient to give people of
commerce (merchants and manufacturers) a fair deal of autonomy in
resolving their conflicts. Yet the royal justice system was always
available to enforce commercial courts’ verdicts.
Despite the deployment of a ubiquitous judicial infrastructure more
than five hundred years before the Industrial Revolution, merchants
and private persons were all wary of ending up in court. It was
universally agreed that procedures were long, expensive and rarely
rewarding. Instead, the general advice was to develop networks of
relations within which one could interact with trustworthy people. The
mass of qualitative information that extols these strategies would not
be understandable had informal mechanisms been unavailable to enforce
time contracts. Thus, Europeans relied upon both formal and informal
sanctions without hesitation.
For China, in contrast, we have so little evidence of formal
mechanisms, and such considerable material on informal ones, that we
have come to expect that a cultural preference for informal mechanisms
can explain the persistence of informal mechanisms amidst a general
reluctance to use formal mechanisms such as courts. From our point of
view, however, the case that has been made for China’s informality is
weak. It rests on two kinds of evidence, neither of which is
conclusive. First, the fact that informal institutions were important
to the Chinese economy from long before the Qing dynasty and
continuing to this day, has led scholars to assume that formal
institutions did not matter. While this long standing reliance on
informal institutions is useful to recognize, it is only necessary but
not sufficient to support the standard argument. The second set of
evidence comes from the difficult century (1850-1950) before China
fell under the sway of the Communists (CITATION). It should surprise
no one that in a century when the imperial administration was
stretched by very costly domestic and foreign challenges, its capacity
to provide formal enforcement of property rights declined. To be more
specific, can we really trust the detailed survey evidence collected
during this time, a period that includes the Japanese occupation, to
reveal the fundamentals of Chinese village society? To answer “yes”
assumes such institutional rigidities that one wonders how such a
society could have survived for millennia. A less distorted picture
emerges if we return to the eighteenth century. In this earlier
period, the state’s involvement in local society was both extensive
and valuable. We know, for example, that the imperial administration
kept detailed records of grain harvests and prices. It also deployed a
variety of institutions to limit the impact of ecological variation
(granaries and water control). And, as we shall see below, it had the
capacity to provide a legal infrastructure for commerce and industry.
Some careful research has cast doubt on the idea that local dealings
were only done on an informal basis. Among English-language scholars,
Kenneth Pomeranz (1995) has shown the durability of a pickle-making
firm in north China which did not rely on kinship ties or other
informal mechanisms for financing and management, but instead sold
shares and selected its management according to performance-based
criteria. In a larger and more recent work, Madeleine Zelin (2005) has
reconstructed the operations of both large and small firms producing
salt in one part of Sichuan province. She shows that firms were able
to obtain initial finance for their operations by selling shares, that
small and potentially bankrupt firms could gain much needed capital by
offering additional shares, and that large firms were capable of
vertically integrating production and distribution operations—all of
this with little or no reliance on the kinship and native place ties
that we conventionally assume to be the basis for Chinese
entrepreneurial activities
Because their convictions about cultural differences have been so
strong, scholars have easily adopted stark conclusions about the
institutional bases of economic change given the contrasting
visibility of informal institutions in China and formal institutions
in Europe. In their view, China and Europe embarked on two alternative
path-dependent patterns of change and, consequently, their economies
reached different institutional equilibria, with the European becoming
more efficient than the Chinese. Thinking this way, the absence of a
judicial system in late imperial China on the scale of Europe’s would
mean that the Chinese never end up expecting to use courts for most
economic matters. Conversely, the European penchant for setting up
courts to hear commercial disputes, begun in the late medieval period
would prepare for the eventual emergence of the contracts and courts
at the center of much of the new institutional economics story of
European economic growth. Such a contrast can even be seen to affirm
the idea expressed long ago by Adam Smith in his observations on China
that economies grow to the point that their institutions permit and
that these points will be different (Smith 1976:106). It further
echoes the belief of Karl Marx that Asian modes of production were
incapable of growth and could only be brought into the present by the
forceful transformation imposed by European power.
Although we take issue with the arguments about institutional lock in
or path dependence, it is not because there were no differences in the
salience of formal and informal institutions between China and Europe.
Our analysis of the use of formal and informal institutions depending
on the kinds of economic transactions in fact implies that there
should be substantial differences between Europe and China. But our
framework explains these differences without recourse to an assumption
about institutional lock in; relative costs are enough. This approach
is more attractive than one that assumes lock in, because as we
discuss below, the evidence supports the thesis that institutional
change occurred in both regions.
The next section examines the persistence and even expansion of
informal institutions for economic transactions in China. We argue
that ‘informality’ in China did not result from any cultural
preferences that prevented the emergence of formal institutions.
Rather, informal institutions were heavily used in the empire because
China’s spatial scale made long-distance trade both feasible and
profitable. Formal institutions, like courts, were not very useful for
commerce over hundreds of miles. In contrast, Europeans came to
formalize their contracts as a consequence of political authority
being exercised on very limited spatial scales. Such strategies worked
only because most exchange occurred on a local scale. When it came to
long distance trade, Europe’s political fragmentation and violent
conflicts surely reduced its volume. Thus, historical narratives
concentrate on how merchants overcame these barriers. It is less
straightforward, however, to conclude that the same factors had any
implication for how contracts were enforced in such long distance
interactions. Indeed, in China, such trade developed internal to the
empire and was unfettered by transit taxes, but as we saw,
relationships were overwhelmingly informal.
The combination of European evidence of informal mechanisms
complementing formal ones and Chinese evidence showing alternatives to
the conventional informal mechanisms playing important roles shows us
that the cases in the lower left corner of Table 1 are truly mixed,
not simply between Chinese and European cases, but among cases in each
region as well. Thus, the contrast in the use of formal versus
informal institutions is neither absolute nor fixed. Evidence on
formal contracting in Chinese firms is at the moment quite limited.
Nevertheless, existing evidence suggests that when presented with
opportunities for which formal institutions were clearly desirable,
Chinese entrepreneurs were able to develop such mechanisms. They did
so well before the examples of Western formal practices became known
in China in the closing decades of the nineteenth century. The
‘native’ development of enterprise forms warns us against the easy
assumption that the Chinese only learned about the superior virtues of
formal institutions from Europeans and after the latter had arrived on
the scene. This assumption is twice flawed. First, the Chinese
development of formal institutions before exposure to Western formal
institutions means they could adopt and adapt foreign models based in
part on their own previous practices. They could equally decide to
forego such adoption when they felt their own mix of formal and
informal institutions worked at least as well, if not better, than the
alternatives offered by Western institutions. Second, as we have
already seen in previous sections of this chapter, formal institutions
are not unconditionally superior to informal ones as the storyline of
the spread of European formal institutions typically implies. Indeed,
informal institutions may well be clearly superior under certain
circumstances. We turn now to explain why this was the case for late
imperial China.
Trade Institutions and the long shadow of empire
This chapter has so far argued that formal and informal institutions
were used both in China and Europe. This stands in stark contrast to
earlier scholarship that invokes cultural preferences as autonomous
forces to explain the institutional choices made in China and Europe
(Landes 1998). One reason for exploring alternative hypotheses is that
such explanations are difficult to evaluate: there are certainly
cultural differences as well as differences in commercial institutions
between China and Europe. But it is not clear how one can go from
correlation (culture and institutions exist together in each place) to
causation (specific cultural traits determine institutional forms)
when we have only two cases to compare. Here we develop a thesis and
offer a word of caution. The thesis is that divergent political
processes over the long run led to different roles for formal and
informal institutions in China and Europe. The caution is that these
contrasting historical processes did not lead to particular kinds of
institutional lock-in. Rather, the relative roles of formal and
informal institutions evolved over time and continue to do so to this
day. It is the evolution over time that offers possibilities for
confirming or disproving our argument.
Indeed, in our framework the relative importance of formal and
informal institutions will depend upon the distribution of
transactions across our four quadrants. Political, technological, and
environmental changes that lead to more long-distance trade will
increase the role of informal institutions in the economy. Conversely,
events that lead to an increased use of fixed capital assets will push
the economy back towards formal institutions. To be sure, the
distribution of transactions according to their frequency and the
distance between parties cannot be evaluated with even a semblance of
accuracy for any pre-modern economy (let alone over a millennium as we
would desire). Nevertheless we can be confident from evidence for the
tenth through fourteenth centuries that the fraction of transactions
occurring over long distances was larger in China than in Europe. We
can also begin to chart the changing role of different institutions.
At the end of the first millennium A.D., China’s lead in long-distance
trade was a direct consequence of its more settled politics. China was
more often than not an empire, while Europe was persistently
fragmented. During the decline of the Roman Empire, some of the most
important trade routes for bulk commodities were simply abandoned and
they would take centuries to reestablish. Europe’s ‘commercial
revolution’ was to some extent a recovery of trade patterns dating
back to the Roman Era. In China, political unification preceded
commercial expansion. Chinese rulers did not view trade over distances
of hundreds of kilometers as international trade but rather as
domestic trade within their empire. Emperors had little reason to
interfere with such commercial activities. That internal peace was the
norm for many centuries over much of the empire was a key element that
facilitated trade to flow across the empire. Moreover, the fiscal need
to intervene in trade within the empire was generally reduced after
the mid-fourteenth century when the state returned to an earlier
reliance on agricultural taxes as the main source of revenues with
only episodic use of commercial taxes before the mid-nineteenth
century. For these political reasons, there was little state
interference in the development of trade. Indeed, the central
government generally worked successfully to keep lower levels of
government from imposing their own levies or in other ways obstructing
commodity flows (Wong 1999: 222-25).
As Chinese and Japanese scholars have documented, high volumes of
trade in basic commodities (cotton, rice, sugar) covered vast
distances in China by1500 and especially in the eighteenth century.
(Xu and Wu 2000) Some of the organizational structures, including
distinct wholesale, retail and transport merchant functions, began to
emerge in the Song dynasty (Shiba 1970). During the Ming dynasty major
networks of merchants, each linked by multiple strands of native place
and kinship expanded the scale and scope of trade. Most famous were
the Shanxi merchants and Huizhou merchants. The former made their
initial fortunes taking grain to the northwest frontier to feed troops
in return for licenses to sell government-regulated salt, while the
latter also made fortunes controlling salt distribution in other parts
of the interior of the empire. Both groups also engaged in other
trades, such as timber and craft goods. (Fuji 1953-4, Fu 1956, Terada
1972). Additional groups of merchants developed trade with Southeast
Asia from ports in Southeast and South China. Some of their operations
began in the eleventh and twelfth centuries, while their expansion
continued up to the early seventeenth century.
China’s early forays into long-distance trade, we argue, privileged
the elaboration of a commercial regime based on networks and informal
mechanisms. Even the best local courts in the world would have proved
unable to adjudicate disputes between merchants who resided hundreds
of kilometers apart. Hence networks developed early and intensely.
Furthermore, these networks succeeded precisely because they became
efficient at acquiring information about the activities of their
members and about market conditions. In some cases, as noted above,
these merchants would go directly into local markets to purchase
commodities. In others there was an interface between merchants taking
goods over long distances and local producers or consumers; these
brokers (yahang) were the agents who either bought the merchants’
goods or purchased from local producers for subsequent sale to
long-distance traders. Merchant manuals counseled traders to be most
careful about choosing a broker and suggested means to gauge the
trustworthiness of a broker as well as the demeanor desired to
demonstrate one’s own credibility (Lufrano 1997). For their part,
officials worried about the asymmetry of knowledge regarding local
market conditions that might allow brokers to take advantage of
long-distance traders. Official involvement could go so far as to set
rules to establish permissible behavior. (Ch’iu Peng-sheng 2008) These
government efforts to influence the norms governing transactions
complemented more informal strategies to enhance trust in transactions
which lacked the advantages of large merchant networks. County
magistrates also adjudicated disputes between brokers and traveling
merchants as one of several kinds of commercial disputes that were
heard during the eighteenth and nineteenth centuries (Fan 2007).
Chinese merchants had recourse to formal dispute resolution, but the
costs in time and money were considerable. Suits grew in complexity
and cost over time, especially in cases involving merchants far from
their homes, as a county magistrate could seek information and
evidence from the merchant’s home county.{here at the very least we
need an additional citation to change over time, but} As a result,
informal methods prevailed in a complementary combination with formal
ones. Given the size of the empire and the many other duties that
local magistrates had, monitoring local markets was given over to
local elites and merchants for the most part(Mann 1987). In sum,
merchant networks using informal mechanisms to make long-distance
trade possible proved to be the most salient and significant
institutional context for commercial expansion. Trade beyond these
networks under the sometimes watchful and anxious eye of officials
also took place, but formal mechanisms were not the major instruments
for commercial exchange.
Much the same logic explains why European long-distance trade depended
largely on informal networks. The absence of empire in Europe did not
prevent long-distance trade. In fact, long-distance trade plays an
important role in the narrative of the rebirth of the European economy
after the great invasions, and in the narrative of all subsequent
growth spurts culminating with the Industrial Revolution. The rebirth
of the European economy was made possible by the demands of local
elites for distant goods, including spices, fine cloth, and other
manufactured products. Much of this commerce went through informal
networks, such as those in the Mediterranean. At first these networks
were centered in the East but by the eleventh century they had come
under the control of Italian cities like Genoa or Venice. The further
growth of Northern Italian cities was premised in part on the sale of
manufactured goods all over the region. In the Spanish Netherlands,
the growth of Antwerp centered around a Europe-wide textile market,
while the rise of Amsterdam was based on its capacity to control trade
in the Baltic, then the Mediterranean, and subsequently beyond Europe.
With the shifts in centers of trade and the growth of trade relations,
more formal institutions came to be employed for economic
transactions—in particular in credit with the formation of exchange
banks (Wisselbank, Bank of England) but informal links continued to be
the main stay of international trade and finance (Neal and Quinn
2003). This remained true all the way through 1912 when J-P Morgan
made his celebrated remark that the most important trait of a
capitalist was his reputation (Carosso 1967). As the Madoff scandal
reminds us, informal relationships remain very important in today’s
financial world.
Despite all that informal networks could accomplish, however, it was
still true that political fragmentation in Europe interfered with
long-distance trade throughout the Middle Ages. While scholars of
Byzantium, Genoa, or Venice have focused on the achievement of their
cities, what they have missed is that one key element in the
competition between commercial centers was the use of violence, which
seriously distorted trade. Had these cities been located in a real
empire they would have been forced to compete on purely economic
grounds. Even after Europe’s military capacities had propelled it to
become the world leader in commercial exchanges, trade was hindered by
politics. Even leaving out dramatic changes to European trade networks
associated with war (as in Genoa’s decline in the Eastern
Mediterranean after their loss in the war of Choggia or Antwerp’s
decline during the Dutch revolt), any graph of the volume of trade
shows dramatic -declines during periods of war (Daudin 2005:207-16, de
vries and Van der Woude 1997: Ch 9).
Beyond war, also trade suffered from the policies of European states.
All governments (absolutist and representative) tried to limit the
extent of long-distance trade. The British Navigation Acts were not
very different from Spanish commercial policies. The motives behind
trade policies were varied, including the protection of local
industries and stemming bullion outflows when their commodity trade
ran deficits. The reason for state interference in trade was, of
course, the need for revenue for war. Taxing long-distance trade was
easy: in each territory most long distance trade was international
trade and it went through a few ports. But European rulers did not
stop there, instead they also taxed trade between their different
domains. Trade between European polities, many of which correspond to
regions within national states today, was thus limited by tariffs and
prohibitions in the Middle Ages (Dincecco 2008). Under the Spanish
crown, for instance, there were significant trade barriers between
Catalonia and Castile, areas much smaller than the typical Chinese
province (Elliot, 1986, Lynch 1989). Despite a long-run process of
political unification within European countries, many barriers
persisted for half a millennium or more within what are now national
entities like France or Spain. Trade barriers between national
entities were first reduced in the middle of the nineteenth century,
and then raised again starting in the 1880s. The process of creating a
common market within Europe had to wait until the 1960s. In brief,
Europe has certainly had developed long-distance commerce, but its
political fragmentation by Chinese standards of political integration,
meant that the expansion of long-distance trade suffered more
challenges than it did in China.
Even with the development of ever more formal institutions for
economic transactions after the Industrial Revolution, much
long-distance European trade and international finance through the
nineteenth century depended heavily on informal networks. These
networks depended upon ties of kinship, geographical origin or
religious affiliation. Hence China and Europe are less different from
one another than some of the contrasts scholars have drawn would
suggest. Transactions in both areas of the world respond to the same
logic. In its starkest form, the model we propose suggests that
cultural variation has little to do with differences in the demand for
formal and informal institutions. Rather, the different kinds of
economic opportunities available at either end of Eurasia are what
counts in explaining the formal and informal mechanisms for contract
enforcement that became typical within each.
When late nineteenth-century conditions changed dramatically with the
introduction of Western technologies and enterprise forms, the Chinese
government began to create a new legal structure. Many of the ideas
and institutions developed in late Qing China were selected from
Western practices, much as when the Meiji-era Japanese elaborated
economic institutions from Western models. The European origins of
these formal institutions have conventionally been assigned a crucial
role in bringing economic change to China. Certainly it is
historically true that many of the concrete practices were first
articulated in European contexts. Yet there is no logical need for
such practices to have originated outside China. The very
effectiveness of the reforms depended on the Chinese contexts into
which they were placed. As we saw earlier China had indigenous formal
contracts of ship shares for overseas trade, as well as for
joint-stock firms (Pomeranz 1997, Zelin 2004). Had there been many
uses for other formal contracts, one should expect Chinese
entrepreneurs to have developed more such forms. At the same time
importing well-developed formal institutional models in the late
nineteenth century saved the Chinese much time and experimentation.
Transposing these new models was rapid in part because of the earlier
‘native’ experiences.
Still, foreign institutional models, including those promoted by the
Chinese government, did not always fit the Chinese context very well.
For example many Chinese firms declined opportunities to incorporate
when new company laws were first promulgated in 1905. (Kirby 1995) The
costs of incorporation may well have weighed heavier than the
anticipated benefits, either because the government could not actually
enforce the new property rights or because company owners feared that
incorporation would expose a firm’s financial situation and thus
increase its tax burden. As the Tianjin Chamber of Commerce opined,
the government promulgated a commercial code at odds with local
commercial practices and largely imitated foreign ones instead. (Fan
2007: 287) While Western formal institutions could not be imported
whole-cloth, Chinese institutions proved open and flexible enough to
accommodate some new practices from the West. A better way to
understand institutional change is as Chinese expansion of practices
that included selected Western practices, rather than a simple
substitution of Western for Chinese ones. These changes suggest that
we cannot take China’s onset of industrialization after Western
Europe, and at rates slower than those in Japan, as a sign of
institutional lock in. There were more obvious factors limiting
economic growth in the first half of the twentieth century, including
a very difficult political situation. Political turmoil overshadowed
any institutional change. The political disintegration started with
the mid-nineteenth century rebellions, it then accelerated with
international conflicts, resulting in the collapse of centralized
political order in the 1900s. On balance, therefore, it seems
reasonable to conclude that foreign models of formal institutions
accelerated changes that would have occurred in their absence and that
unstable domestic conditions slowed Chinese institutional change.
The existence of both formal and informal institutions for economic
transactions in late imperial China implies that institutional imports
were not necessary for growth to takeoff. The use of formal
institutions would have increased when economic change made formal
institutions superior to informal ones. The growing availability of
foreign models did, of course, change the choice set available to
Chinese policy makers and entrepreneurs. Not surprisingly, some
Western practices were adopted. Equally sensible, though less commonly
noted by scholars, new practices were developed that drew on older
domestic ones. As a result of both dynamics, a changing mix of formal
and informal institutions emerged wherein formal practices grew in
relative importance. Expanding our observations to look at the
variable use of formal and informal institutions in early twentieth
century China and Europe, we argue that decisions about the relative
attractiveness of formal versus informal institutions for similar
kinds of transactions no doubt were influenced by earlier choices to
use one set of practices rather than another. Thus, the optimal mix of
formal and informal institutions differed according to the
path-dependent contexts of both China and Europe in the early
twentieth century.
Conclusion
The rise of modern economic growth in Western Europe has led many
economic historians to posit the existence of a set of institutions
that are optimal for growth, namely those of eighteenth century
England. Because China’s contracting institutions are quite foreign to
the English common law tradition, one might be tempted to conclude
that, absent major legal reforms, China’s economy could not develop
successfully. The past three decades of explosive growth questions
such conclusions, especially since a significant part of the growth in
industrial output in the 1980s and early 1990s was achieved without
formal courts and contracts.
An alternative argument would have it that while English institutions
may work in Britain they cannot be expected to have similar benefits
in China or in other quite different cultural or geographic
environments. In particular different scholars have documented the
importance of the conditions Europeans encountered to the kinds of
institutions they set up in their colonies. (Acemoglu et al. 2001
Engerman and Sokolof 1997). Among their most startling findings was
that within an ecological zone there was little variation among the
different colonizing nations in the institutions they chose. This
approach, like ours, takes geography quite seriously, but for our
comparison it does not lead to strong conclusions. Indeed, the
environmental variation across Europe is extremely large and even more
so in China and yet the striking institutional differences are more
between the regions than within.
When trying to account for differences in contract enforcement between
China and Europe we have relied upon the observation that in some
relationships formal enforcement is not very useful, just as in others
it is the only way to make sure parties adhere to contracts. From
there we developed a framework of enforcement choice that showed that
successful economies deploy a mix of formal and informal enforcement
rather than rely solely upon one or the other mechanism. The
historical record is fully consistent with this view. While merchant
networks (organized around kinship or place of origin groups) seem to
have been very extensive in China, they did not replace formal
contracts. Further, when considering the relative importance of formal
or informal contract enforcement, we found that the structure of the
economy is likely to be key. Hence it is not surprising that Europe, a
region where politics interfered with long-distance trade, saw more
transactions that were local and formal than did China. Nor is it
surprising that the trade resurgence of the late-medieval period in
Europe brought about a flowering of informal arrangements among
merchants.
In many ways the argument we offer echoes the discussion of the theory
of the firm we presented in Chapter 2. That theory says that whether
an activity is structured by a firm’s hierarchy or by a market mostly
depends on the characteristics of that activity. In some cases it is
best for these activities to be coordinate with a set of other
activities in a firm, in other cases the market is more efficient. The
argument in this chapter emphasizes that the division of labor between
formal and informal enforcement depends on the nature of the
transactions. Yet time and again the historical record forces us to
concede that it is politics more than simple economics that matters.
That the Middle Kingdom had a large volume of domestic long-distance
trade and a small volume of international trade (while Europe had the
reverse) is a direct consequence of the imperial scale of the Chinese
state. European governments competed to provide formal enforcement of
contracts because they were after the revenues produced by contract
registration and dispute resolution. Those revenues in turn were spent
on military investments.
In this chapter we only examined the demand for formal contract
enforcement (how individuals would want to structure their
transactions), in effect we assumed that the supply of such
enforcement mechanisms would respond to changes in demand both in
China and Europe. In many places in the world, the supply of courts to
adjudicate cases is often wanting. Even in our two regions, rulers
sometimes failed to provide needed formal institutions. For example
tenth and eleventh-century Chinese policy makers declared government
monopolies for many commodities and foreign trade restrictions
variously imposed between the sixteenth and eighteenth centuries
curtailed growth possibilities (CITATION). Informal institutions
(smuggling) arose as a result. In Europe, political intervention that
wrecked the commercial networks of Antwerp in the late 16th century
and the failure of the French state to reform its judiciary prior to
the Revolution had even more serious political and economic
consequences (Gelderblom 2000, Rosenthal 1992). Yet in Europe and
China for most commercial transactions, governments were willing to
provide enforcement services or to delegate those tasks to local
authorities. In both regions formal enforcement was imperfect. There
has never been a time when business people have not complained of the
cost, uncertainties, and delay associated with courts but that has not
stopped them from litigating. Imperfect formal enforcement does not
make it valueless, however. In fact, both in China and Europe,
infrequent transactions like land were formalized.
Beyond these regions, however, there are numerous examples of states
failing to provide the rule of law and instead structuring economic
interactions to favor specific elites. The history of Latin America
and colonial Africa are often told in such a light (Nunn 2008, Haber
et al 2004). In such places, informal arrangements are the only ways
for the disenfranchised to structure interactions. Those with
political clout may have access to property rights and the courts but
such benefits are contingent on informal and unstable relationships.
Developmental lessons have been drawn from such tragic examples
(North, Wallis, and Weingast 2009). The history of China and Europe
alert us that these failures are political rather than economic or
cultural. Further, differences in the extent of formal enforcement
across societies may not be symptoms of pathology but rather the
consequences of differences in economic structure.
Much like North, Wallis, and Weingast we see politics as a critical
variable in explaining differences in performance. Starting with
Chapter 1 where we explored the spatial scale of Europe and Chinese
polities, the theme of the impact of political structures on economic
performance has loomed large in this volume. The burden of Chapters 2
and 3 has been to demonstrate that China and Europe were probably more
alike economically than their institutions or cultures would suggest.
Chapter 4 turns to the central issue of this book: the role of
politics in explaining the great divergence in economic performance
that could be observed by 1800.
Table 3.1: Contractual Arrangements in trade
Interval between interaction
Distance between parties
D D>D*
T>T*
Formal feasible
Informal not feasible
Prediction: Formal
Formal Not feasible
Informal Not feasible
Prediction: Cash only transactions
T Formal feasible
Informal feasible
Prediction: ?
Formal not feasible
Informal feasible
Prediction: informal
Table 3.2: Contractual Arrangements in trade with differences between
societies
Interval between interaction
Distance between parties
D Dc D>De
T>Tc
Europe: Formal
China: Formal
Europe: Formal China: Cash
Europe:Cash
China:Cash
Te> T>Tc
Europe: Formal
China: Either
Europe: Formal China: Informal
Europe: Cash
Europe: Informal
T< Te
Europe Either China: Either
Europe: Formal
China: Either
Europe: Informal
China: Informal
Note: in the darkest grey areas ( ) Chinese and European transactions
should be enforced in similar ways. In the light grey areas there are
differences and one society has an advantage, either because both
types of enforcement are available rather than only one ( ), or
because one type of enforcement is available rather than none ( ).
Only in the center box left unshaded can one find the traditional
opposition between informal China and formal Europe
Box 3.1 Formal and Informal Contract Enforcement
Consider a loan of value L: When the loan comes due if the borrower
default he earns π and the lender 0. If the borrower is honest and
repays he earns a return of h=π-(1+r)L and the lender gets (1+r)L. In
the absence of any enforcement mechanism the borrower never repays
(h<π), and consequently the lender makes no loans.
Formal enforcement of a loan of value L.
Here the borrower who defaults can be taken to court. If sued he will
lose, have to repay the debt with interest, and bear the court costs.
Hence the borrower always repays if he thinks he will be sued after
default. Will the lender sue? If he does not sue, he gets nothing
back. If he prosecutes he wins (1+r)L (gets his money back and some
interest foregone). He also bears the costs of litigation (payments to
the courts and to legal experts, and time spent, including travel to
the residence of the borrower/buyer). Those costs increase with the
distance between his residence and that of the borrower and we will
denote them C(D). He will sue if (1+r)L >C(D). Given a loan of value
L, the lender always sues borrowers who live less than D* away from
him. It is easy to show that if courts become more efficient at
enforcing judgments (C’(D) judgments get enforced (D*) increases.
Informal enforcement of a loan of value L.
When the loan comes due and the borrower defaults, he is always found
out and is, consequently, excluded by his network from a range of
transactions forever, so that his best future alternative will earn
him b per period; and b assume that the buyer interacts with the network once a period (the
period can be very short). Let d be the discount rate, the discounted
present value of receiving h forever is H. Similarly, the discounted
present value of receiving b forever is B. The borrower contemplates
the difference between h+dH (being honest today and forever) and π +dB
(cheating and being branded a bad partner ever after). Then the net
returns to being honest if interaction occur once per period are R(1)=
h- π +d(H-B). If the interval between interactions is T periods R(T)=
h- π +dT+1(H-B). R(T) is declining in T and there is a unique T* such
that R(T*+1)<0 valuable (B’ Box 3.2: Contract enforcement when both formal and informal
enforcement are feasible.
To resolve the issue of what occurs for frequent and local transaction
we must introduce some heterogeneity in the borrowers’
characteristics. Moreover, unlike prior models, we want both types of
arrangements to persist over time. In our initial model there was no
default in equilibrium (it was a dominant strategy for the borrower to
repay) hence both mechanisms were equivalent (the seller never goes to
court and never has to actually exclude anyone from trade). The
argument presented here is illustrative of the kind of conditions
under which both types of arrangements can persist in a locality.
Lenders are concerned about two issues: the interest rate and the
likelihood of non-performance. Each lender is a member of a
reputational coalition or network and can either make loans to his
fellows or to someone outside the coalition. If he transacts in the
general population he can expect the borrower to pay rF. But in this
case, the seller face some adverse selection (there are some borrowers
who do not repay) and he has to go to court. He then earns (1+rF)L-pc,
where p is the probability of non-payment and c the cost of formally
enforcing the contract. If he makes a loan to a member of his network,
there information is good and contracts are always performed because
the value of being honest is larger than the value of default for the
borrower. However, using the coalition has a cost—It is small (a
subset of the general population), thus the best interest rate offered
within the coalition is rI ≤rF. It follows that all transactions where
(1+rF)-pc ≤(1+rI,) occur inside the coalition while the others use
formal enforcement. It also follows that larger transactions are more
likely to be formal. Beyond loans, this analysis suggests that goods
where the range of willingness to pay is small are likely to be
transacted informally while those in which individual tastes really
matter are transacted among strangers and enforced by courts.
Chapter 4
Warfare, Location of Manufacturing, and Economic Growth in China and
Europe
Introduction
Our analysis of contracting arrangements put a significant emphasis on
the size of the Middle Kingdom and importance of long distance trade
in explaining why China relied on informal arrangements more than
Europe did. That the size of the Chinese empire encouraged the early
rise of long distance markets is well established (Pomeranz 2000).
Upon reflection it should cast doubts on a common thesis in economic
history that political competition is directly beneficial to economic
growth. Political competition historically has meant violent and
expensive domestic and international conflict rather than well ordered
and cheap elections or even armed peace. Empires, like China, have
little political competition and, for a long time in the past, they
were rich. In contrast, regions with multiple polities bear the cost
of war time and again, and even in peacetime bear distortions to trade
that reduce the volume of long distance trade. Economists would do
well to remember that most of the restraints to trade that Adam Smith
or David Ricardo identified as reducing economic efficiency simply did
not exist within Ming-Qing China.
In this chapter we explore the relationship between political
competition and economic change further. In particular we focus on the
role of war in the rise of mechanical technologies in Europe starting
at the end of the seventeenth century. Our diagnosis of the proximate
source of divergence accords with a large literature (most recently
Joel Mokyr 2009 and Robert Allen 2009): England and later Europe’s per
capita income began to rise rapidly after 1750 because that part of
the world was more successful in implementing mechanical techniques of
production. However, we differ as to the reasons why Europe sprang to
leadership. Some have argued for environmental (Diamond 1997),
cultural (Mokyr 1990, 2009), and political factors (Jones 1981). To
our mind each of these arguments suffers from problems of chronology.
While romantically attractive, Mokyr’s focus on European enlightenment
and openness to new ideas seem to put aside the extensive religious
and political conflicts that crippled many parts of the Europe both
before and after the Reformation. Enlightenment ideas may have
sustained growth but it certainly did not cause it. Much the same can
be said of the environmental bonanza that Europe reaped from the
colonization of the America (Pomeranz 2000, Jones 1981).
To our minds the causes of economic divergence between Europe and
China emerged earlier. By 1500, the European and Chinese economies
were on structurally different paths. Leonardo’s sketch books may
mostly contain drawings of machines that could not be built, but they
represent an early manifestation of Europe’s love of machines. The
passion for mechanical innovation that would blossom over the next
three hundred years was far scarcer if present at all in China. The
class of potential explanations is very large, but examining key facts
about manufacturing before the Industrial Revolution helps us focus
the analysis:
*
It is now well established that China had an early lead in
technology and that its technology continued to evolve long after
the famed peak of the 1350s. (Needham)
*
By 1000, although low-skill, low-capital handicrafts were rural in
both areas and high-skill high-capital industries were urban in
both areas, the range of manufacturing that was sustained in
cities was much larger in Europe than in China (Van de Wee)
*
By 1600, Europeans were developing and deploying machinery more
intensely than the Chinese. (Mokyr 1990)
*
By 1700, the technology was such that it paid to adopt the new
machinery only in a small area of Europe where particular relative
price ratios favored capital over labor. (Allen 2007)
Rather than build a theory to explain the fourth or even the third of
these points, as is common, we focus on the second and take the first
as given. We do so because we want an argument that allows technical
leadership to move from one location to another say from China to
Europe or vice versa. We must allow both societies to be technically
creative to avoid developing a trivial theory in which Europeans will
succeed from the outset. It therefore eliminates all possible
arguments that make European cultural or political arrangements
superior to those found in China (e.g. Landes 1998). Indeed if
coastlines or formal law favored Europe, then why Europe so poor a
dozen centuries ago. It also eliminates all the arguments that focus
on European institutions like the corporation, which diffused
throughout the world but only came to be important after 1700. These
include many arguments that focus on political and cultural
developments like the Enlightenment or representative democracy (Mokyr
2002, North and Weingast 1989); while they may have provided powerful
boost to the process, they occurred too late to matter. Instead we
need to a social process that first gives an advantage to China and
then at some point allows Europe to take over.
Our argument has two parts: first, war was responsible for Europe’s
urban manufacturing; second, Europe’s urban bias is precisely what
produced the high rate of capital investment and the adoption of
machinery in ever greater areas of Europe. In contrast, China’s
peaceful economy experienced neither the pressure to protect its
artisans behind city walls or the consequent inducement to use
machines to save on expensive labor. We highlight these long-term
tendencies rather than a moment of critical invention, like the
appearance of the steam engine in the 1690s because no one critical
event in the seventeenth or eighteenth century propelled England or
Europe towards mechanization. Moreover, in the fourteenth century,
Europe does not appear to have much of a mechanical advantage over
China, nor is there any evidence that wages were much higher there
than elsewhere in the world. The urban bent of European manufacturing
relative to its Chinese counterpart is, however, extremely old. What
produced this bias is the focus of this chapter.
Cities and Economic Growth
Rather than ask what pushed Chinese manufacturing to be overly rural,
we ask what pushed European manufacturing to be urban early on. While
the distinction appears academic it has important analytical
implications. In the first case, one sets up the European pattern as
efficient and then looks for a Chinese pathology. Such an approach
might be appropriate for the mid-nineteenth century when urban
industries had clearly become a critical element to growth, but that
was far from obvious in 1200. It is more historically relevant to seek
out what pushed Europeans to choose urban locations for activities
that could have been accomplished equally well in the countryside
where food was cheaper, raw material were easier to access, and
diseases were less prevalent. It has the additional advantage of
allowing us to recognize that China’s economic centers were likely
more efficient and prosperous than were Europe’s in the fifteenth and
sixteenth centuries. Such a reorientation will prove quite fruitful.
The question of why Europeans had so much more of their manufacturing
in cities than the Chinese did has three broad potential classes of
answers. First, there are demographic and economic factors that could
make cities more attractive in one place than another. Second, it
could be that differences in political economy led rulers to favor
cities at one end of Eurasia rather than another. Finally, there are
the consequences of regional differences in political structure, in
particular the spatial scales of polities. In our view it is this
third set of explanations that is correct, but unlike earlier
scholars, we do not view Europe’s surge to mechanical leadership as
the direct outcome of benevolent policies but rather as the unintended
consequence of a regional political system embroiled in costly
conflict.
Let us begin by dispensing with some simple questions that might
explain what Chinese entrepreneurs might have preferred the
countryside. The most obvious candidate is demography. In particular,
urban mortality might be responsible for the lack of manufacturing
cities in many parts of the world. Prior to 1800, cities everywhere
had such high mortality rates that they had to import people from near
and far to sustain their populations (Grantham 1993, Wrigley 1967).
Artisans might have been tempted to choose rural locations for their
shops simply to avoid the pernicious disease environment of cities.
Such mortality problems were perhaps more severe in warmer climates
(where waterborne diseases tend to proliferate) than in colder ones.
But then cities should have been larger and manufacturing more urban
in northern areas of China and Europe than in southern ones. In fact,
cities and towns grow more rapidly in southern parts of China than in
the north after 1100, while Europe’s larger cities with more urban
manufacturing were in the south rather than the north before 1500.
This demographic argument does not lead to a divergence between China
and Europe.
A second possibility for the lack of urban manufacturing in China, was
envisaged by Adam Smith; it focuses on the poverty of China’s larger
population. In other words, capital was more abundant in Europe than
in China, and because capital markets were more active in cities, the
cost of capital was both lower in Europe than in China and in cities
relative to the countryside. To a large extent this is the argument
developed by Robert Allen to explain the early adoption of machinery
in England (Allen 2007, 2009). While it is very appealing for England
in 1730, it is more difficult to sustain for Europe before the Black
Death when interest rates were considerably higher than they would
later become and wages lower, yet manufacturing was already very urban
in Europe (Epstein 2000). While there is little or no data on Chinese
wages and interest rates for this early era, the qualitative evidence
strongly supports the notion that the empire’s manufacturing was
already becoming increasingly rural after the founding of the Ming
dynasty in 1368.
Beyond simple factor prices one could seek an explanation for the
concentration of manufacturing in cities in Europe from economic
geography. Research on urban systems has long emphasized the
beneficial effects on costs and productivity growth of industrial
clusters. In economics parlance manufacturing derives increasing
returns from network externalities (see Fujita et al 2001). The idea
is that production processes are more efficient when they are
spatially concentrated. These externalities have been argued to come
from thicker and more specialized input markets, greater competition
among firms, and the willingness of workers to acquire job specific
skills, because if their firm treats them badly, they can find another
employer desirous of their skills next door. It is important to
stress, that the existence of such externalities alone is not
sufficient for a divergence between China and Europe. In fact, if the
returns from urban location are large and ancient enough they should
have been discovered at both ends of Eurasia and the location of
manufacturing should have been similar. Furthermore, given the
existence of large cities in Asia and in China in particular by 1000,
one would assume that China would have embarked on an urban path for
manufacturing prior to Europe. To obtain a divergence that favors
Europe we need to differentiate across economic sectors in order to
identify industry-specific externalities with those benefiting from
agglomeration economies accounting for a larger share of output in
Europe than in China. The one industry that long differentiated Europe
from China is probably weapons production (Hoffman 2009). But the
scale of those activities was another consequence of Europe troubled
politics which we will take up below. On the whole, economics alone is
unlikely to explain this urban bias.
The second class of explanations comes from domestic political
economy. The range of such theses is wide because either ‘bad’
policies in China or ‘good’ ones in Europe could be responsible for
the European bias towards urban manufacturing. If we found Chinese
emperors making it difficult for manufacturers to locate in cities
(and thus preventing their subjects from taking advantage of the
externalities associated with urban manufacturing), we could argue
that Chinese entrepreneurs preferred to locate in cities. If we
discovered that Chinese emperors suppressed capital markets, thus
negating the possibly cheaper costs of capital in cities, then we
could argue that bad policies hindered Chinese economic development.
Chinese emperors did valorize an ideal of men plowing the fields and
women weaving at home, but this political preference did not lead to
real constraints on geographic mobility. Nor were people prevented
from lending money; prohibitions on extremely high interest rates did
not affect the cost and hence availability of funds.
On the European side, medieval historians have long stressed the
explicit policies of rulers of Northwest Europe that aimed to attract
skilled workers to their territories and to their towns (e.g. Duby
1974, 1979). These policies surrendered some of the sovereign’s
authority to municipalities or more directly to groups of craftsmen or
merchants organized as guilds. There is also evidence that cities and
guilds actively attempted to limit the capacity of rural manufacturers
to compete with urban producers (Van Der Wee 1998, Vardi 1993, Epstein
2000 ch. 5). Although one might make guilds responsible for the urban
structure of production, one should bear in mind that each town had
not one but many guilds, and even in a given industry they favored
quite different policies. Towns were sufficiently small that no guild
controlled the production of any good over any geographically
significant market. As we shall see below, the boundary between rural
and urban manufacturing was never fixed in Europe. Moreover, European
guilds also served to protect their members from the rapacity of the
ruler. Indeed, kings and princes were often tempted to confiscate the
goods of merchants and craftsmen when they needed cash. As Greif
(2006) notes in his examination of the conditions under which
merchants might travel to distant markets, individuals have little
power to resist rulers’ temptation to tax or steal; in fact only
groups can stop such expropriation. A broader consideration of this
matter leads us to the observation that both the relative scarcity of
skilled workers and rulers’ rapacity had their root cause, not in some
flaw in the domestic political process, but in the ceaseless warfare
that Europe experienced. Domestic political economy, like economics,
drives us to consider international politics and, in particular, war.
The last class of explanation focuses on regional differences in
political structure. For simplicity we take China to be a region of
unified political control where war and civil disturbances were
infrequent (except on the frontiers), and Europe to be an area of
competitive politics where the likelihood of war and civil strife were
much higher and more local. Even in peacetime Europeans and their
rulers had to prepare for war; in China that problem was left to the
emperor and his generals. Unlike other arguments that emphasize the
benefits of political competition without measuring its costs, we
recognize that political conflict was not a mere threat in a
bargaining game, but something that happened and, when it happened,
was expensive. In our view the primary reason for European
manufacturing’s urban bent was war. Although everyone wanted to escape
war’s ravages, farming was necessarily tied to the land and peasants
to villages. Manufacturing, meanwhile, was both more mobile and more
prone to pillage, in particular in activities that produced objects of
high value per weight. European artisans therefore sought the
protection of city walls rather than the more modest defenses
available in villages. In China, by contrast, during the long
centuries of dynastic stability the low frequency of warfare led
manufacturers to choose their locations according to a different
calculus. The relative prices they perceived were less affected by the
anticipated ravages of war. The next section develops this argument
and begins to trace the long-term impact of differences in the
location of manufacturing.
Factor Costs and Manufacturing
We begin with the general observation that in most handicraft
activity, firm size is tiny relative to the market, thus competition
prevails. As a result, over the long run enterprises will locate where
the costs of production are lowest. While the long run may not be a
good way to analyze modern economies because factor costs and
technologies are constantly changing, it will work well for our case
because we are interested in secular tendencies in an era when
technologies and factor costs generally changed slowly but the latter
could be subject to shocks that changed the relative costs of capital
and labor.
Sixteenth and seventeenth-century cities had advantages and
disadvantages relative to the countryside. Urban dwellers faced
increased risk of death and illness because concentrated populations
are good loci for disease. People who lived in cities also faced
higher food prices because staples have to be brought in from rural
areas. Thus nominal wages must be higher in cities than the
countryside. In consequence an entrepreneur’s cost of labor will be
lower in rural areas than in urban ones. Evidence for such cost
differentials is particularly abundant for the nineteenth century, but
can be seen in earlier periods from the correlation between nominal
wages and city size (Ditmar 2009). If we consider capital, the reverse
relationship will hold: rural projects are more costly to monitor
because they are dispersed and (we may assume) individually smaller in
scale. The higher costs will bear on borrowers and interest rates in
the countryside will be higher than in cities. Evidence for such cost
differentials is harder to grasp because in most pre-industrial
economies interest rates were not specified in contracts. One can turn
to data about the geographic structure of credit market: The
systematic pattern of rural individuals going to towns and cities to
borrow rather than to lend (and of city dwellers making more loans in
the countryside than borrowing there) strongly argues that the cost of
capital was lower in cities (Hoffman et al. 2010).
To evaluate the impact of these relative prices we must define the
production technology. To keep things simple we begin with a
production function where the ratio of capital to labor is fixed (in
economist’s terms a Leontief technology). For example, assume there is
one kind of loom and one type of worker: a Leontief technology arises
if the only way to combine workers and loom is one worker per loom. In
the case of fixed factor proportions, the entrepreneur who is seeking
to minimize his costs simply picks the location where the input he
uses most is cheapest: capital intensive activities locate in cities,
while labor intensive activities are in the countryside. In fact, as
the analysis in box 4.1 shows, there is a unique level of capital per
worker k*such that all industries (or firms) that use more capital
than k* are in cities while those that use less are in the
countryside. This first step produces the classic differences in
capital intensities between urban and rural areas, but the proposition
on its own offers no help for understanding why China and Europe took
different paths.
Warfare creates the difference in factor costs that can cause a
divergence in the location of manufacturing between the two ends of
Eurasia. War matters because rural projects are more likely to suffer
from either civil disturbance or international warfare than urban
enterprises. This is particularly true for capital invested in movable
goods (equipment, tools, supplies and people) because they can
appropriated by bandits, warlords, or foreign armies during unsettled
times. Cities are not immune to warfare. Among other things, war
disturbs the trade networks that are essential to cities, and of
course their wealth makes them attractive places to pillage. Yet
cities can be fortified and resist redistribution through violence. To
be sure building walls and hiring guards was expensive, but many
manufacturers found it preferable to locate behind city walls than in
undefended rural areas. Our interpretation is that war increases the
cost to capital in both cities and rural areas but that the rural
increment is larger than the urban one. While cities can protect
capital they are not as successful with labor because a disturbed
peace will hinder economic exchanges between city and countryside,
further raising food prices.
In a region beset by threats of warfare, the entrepreneur decides
where to locate according to a different set of relative prices and
thus a new critical level of capital per worker kw* decides what firms
are urban or rural (see Box 4.1). Because war has made capital cheaper
in cities than in the countryside, that threshold is lower than in the
peaceful economy (kw*< k*). Simply put, some industries, those with
capital labor ratios between kw*and k*, are urban in the warring
economy but rural in a peaceable economy. The industries remaining in
the countryside during unsettled times are the least capital intensive
of all.
[Box 4.1 about here]
Because China had few civil and international disturbances between the
mid-fourteenth and mid-nineteenth centuries, it gives us our base
line. All industries with k Europe had lots of war, only European industries with k< kw* are
rural. Because kw*< k*, more industries in China locate in the
countryside. Thus, war produces the urban bias that characterizes
Europe from the fall of the Roman Empire forward. If warfare is
sufficiently severe, the bias will also be large. Although the
Leontief technology can be combined with war to explain differences in
the location of manufacturing, it has limited implications for
technical change. In such a technology, factor proportions are fixed,
(k the capital per worker describes the technology fully) and as a
result capital labor ratios in the same industry are identical in
urban and rural firms. If Europe’s primary characteristic is that it
is a warring economy, then relative to peaceful China it would be
poorer and have less manufacturing over all.
The fixed proportion production model is unfair to Europe because it
does not allow entrepreneurs to substitute cheap factors for expensive
ones, even though such substitutions are ubiquitous in reality. To
return to the example of the weaver, it is in fact possible to employ
more than one worker per loom (in particular women and children as
helpers) and it is also possible to have more or less capital per
worker (because looms vary in quality). In each case the combination
of labor and capital is different. The simplest class of production
functions that allow such substitution have constant factor shares5
(rather than proportions) and are known in economics as Cobb-Douglas
production functions (Q=KaL1-a, where Q is output, K is capital and L
is labor, and a is the factor share of capital). We will not carry out
the mathematical analysis here but the interested reader can find it
in Box 4.2. Just as in the Leontief case, entrepreneurs locate their
production based on relative prices and there is a unique factor share
of capital a* such that industries with larger capital factor shares
locate in cities and industries with lower capital intensity locate in
the countryside. Similarly the war economy has a smaller threshold
factor share of capital a w *than the peaceful economy. So far we have
reproduced the substance of the lessons of Leontief model. But
Cobb-Douglas technologies allow us to go further. Indeed, when we
allow for capital-labor substitution more industries locate in cities
in the warring economy. Box 4.2 provides the technical details, but
the intuition for this result is that Cobb Douglas technologies give
firms two ways of mitigating the impact of war: choosing their
location and adjusting their factor proportions.
[Box 4.2 about here]
The adjustment of factor proportions to urban locations is a general
phenomenon. Any firm that locates in a city will operate with a higher
capital labor ratio than if it had been in the countryside. Urban
firms face cities’ high labor costs and low capital costs, so they
will want to substitute capital for labor. Thus, when a firm locates
in a city rather than a village it uses more capital and less labor.
In our model, because all firms in the same industry choose the same
location this extends to the industry. Relative to the fixed
proportion model, the key difference is that industries pushed into
cities by war become more capital intensive. If Europe is the war torn
economy, it has a more capital intensive manufacturing sector relative
to China, because more of its manufacturing sector is imprisoned in
cities by warfare. As we argue below it was this capital bias that set
Europe off on a different path towards machine-based innovation: urban
manufacturers in Europe created more machines than their rural Chinese
counterparts because they had more use for them.
The chain of causation in our model has two parts. The first is static
and runs from war through relative prices, urban versus rural
location, and then to factor intensity. The second is more classic and
dynamic; it runs from factor intensity to technical change, falling
into the broad class of induced innovation theories of technical
change. The rest of the chapter defends the plausibility of this
causal chain, in particular the static elements.
Such a defense is required because although our model is plausible, it
is but one of many narratives of economic change one could construct.
Moreover, the model’s theoretical purpose is to produce the divergence
we highlight. To do so, we need an appropriate friction in relative
factor costs and war is just one of the processes that can potentially
produce such friction. Because there are many other differences
between China and Europe, candidates to act as friction are numerous.
Yet we can eliminate all those that were of such long standing that
they would have given a lead to Europe from the outset since Europe
was not always ahead of China economically. Similarly we can set aside
any friction that would have made it impossible for China to be ahead
early on or to grow extremely rapidly later. Finally, sharp changes in
Europe (like the Glorious or French Revolutions) are of limited
relevance because the process of technical divergence took centuries
not decades. We find that warfare has a singular advantage over other
long-term factors: its intensity waxes and wanes, and if we are
correct the location of manufacturing in each region should reflect
the ebb and flow of political disturbances—not just technology. The
ebb and flow of warfare, in fact, turn out be just what we need to put
our argument at risk of falsification.
Long-Term History before the Industrial Revolution.
Here we focus on the static or direct effects of war. Insecurity (to
put war and civil violence in more neutral or euphemistic terms) is
very costly. Indeed as war costs increase and manufacturing shifts
more and more to cities, the economy and the manufacturing sector also
shrink because of the toll that warfare imposes. By implication the
economies of societies in which warfare is prevalent are smaller and
have smaller manufacturing sectors. Thus, up to the sixteenth century
and perhaps beyond that, war should make Europe poor relative to
China. For similar reasons China’s manufacturing should be larger and
more rural than Europe’s after the Mongols reunified the Empire in
1279. Conversely in Europe, should the intensity of warfare decline,
some manufacturing should move back to the countryside. Finally,
should technology become more capital intensive Chinese (and European)
manufacturing should become more urban. In tracing out the urban rural
competition for manufacturing location we must explicitly deal with a
comparison between China and Europe and a comparison between England
(the cradle of industrialization) and the rest of the world. It would
be particularly desirable for our model to have implications not just
for the divergence between Europe and China but also for variation in
the location of manufacturing within each region.
The accounts of early European travelers, as well as the flow of
technology, also suggest that early on China was far more economically
advanced than Europe, and that Europeans went to the Far East in
search of manufactured goods, not raw materials or precious metals.
That China was technologically ahead of Europe at the end of the first
millennium CE is generally accepted in the literature and forms the
core of the China puzzle—namely why an economy that was so advanced
should fall progressively behind after 1300 (Elvin 1972). Could the
connection between warfare and urbanization help explain this? In the
mid-thirteenth century China’s cities may have amounted to between 6
and 7.5% of the total population. The empire certainly had a number of
very large cities. It seems that thereafter the urbanization rate
declined under the Ming and the Qing. To be sure by 1600, warring
Europe seems to have been more urbanized than China (CITATION), but
that is hardly a test of the argument. We must look more closely
within the two regions.
Let begin with a careful examination of urbanization and war in China.
The Middle Kingdom certainly had its share of military troubles, for
instance in the mid-seventeenth century with the collapse of the Ming
dynasty and the establishment of the Qing dynasty, and again around
the mid-nineteenth century when there were widespread peasant
rebellions. But, for most of the three centuries preceding the
Ming-Qing transition, and for the two centuries before the
mid-nineteenth century rebellions, Chinese society was generally quite
peaceful. Thus, Chinese entrepreneurs did not usually need to
anticipate warfare disrupting their production and distribution
operations. They were spared the costs of warfare not only in the
direct sense of having to pay taxes to support war-making initiatives,
but also in the less obvious way of not having to pay for protection
from the threats of confiscation and destruction.
If we take a broader sweep of history, China’s instances of political
fragmentation show patterns of urban manufacturing similar to those of
Europe. Recall that prior to its unification under the Qin dynasty in
221 BCE, China was the theater of severe political competition for
more than two centuries. During this time China was divided among
seven major warring states, each anchored around great cities that
hosted both commercial and manufacturing activities. Rulers minted
coins to facilitate trade, which they taxed in order to mobilize
resources to pay for warfare. They expanded agricultural output
through irrigation and improved iron tools in order to feed the cities
housing their governments and urban craftsmen. We lack adequate
information on urban and rural locations of craft industries for the
first twelve centuries of imperial rule that began with the Qin
dynasty. The long stretches of political division and military
competition between the periods of imperial integration and grandeur
account for more than forty percent of the entire period. It is
therefore not likely that rural manufacturing enjoyed the kinds of
advantages it would starting in the late fourteenth century. We do
know that the commercial expansion of the Song dynasty (960-1279) was
powered by a combination of improvements in agriculture,
transportation technologies and urban-centered craft production (Shiba
1968). This was also an era of great political insecurity for the
regime, forcing a move in the early twelfth century from the north to
Hangzhou, which became a great center of manufacturing and wealth
(Gernet 1962). Thus through the fourteenth century, it is likely that
competition between rural and urban manufacturing was intense in China
By the fifteenth century, with the advent of Mongol rule, rural
handicrafts began to play an increasingly important role in
manufacturing. A clear contrast of relative peace in China and
frequent warfare in Europe comes to characterize the early modern era
at the two ends of Eurasia.
Although, internal and international violence was less prevalent in
China than in Europe, even a casual glance at early modern renderings
of Chinese cities would convince skeptics that they were walled and
gated. Yet relative to Europe, the number of such cities was limited,
as was the size of their fortifications. Indeed, imperial officials
seem to have perceived investments in urban defense as having low
returns. For their part, most people appear to have felt little need
to locate within the confines of a walled city since some ninety-five
percent of the population lived in rural areas and some ninety seven
percent lived outside of walled cities as late as 1843 (Skinner 1977:
227, 287). Chinese with capital did not seek out cities to protect
their investments in the same way that Europeans did because of the
threat of warfare. Instead, for most Chinese dynasties the threat of
warfare came from the steppe; armies were routinely deployed along the
northern frontier. In both early and late imperial times the
fortifications collectively known as the Great Wall symbolized the
state's commitment to assuring peace from foreign marauders and
invaders for the whole of the empire--town and country alike. Before
the tenth-century shift of China's population toward the south, what
little industry that did exist seems to have had more urban locations
perhaps in part because the northern locations of industry made them
more vulnerable to foreign military threats.
Within the empire, especially after 1000, domestic social order did
not usually entail large investments in fortifications. Chinese
officials pursued a variety of normative, material and coercive
strategies to promote and enforce both rural and urban social order
(Wong 1997, 105-26). When growing numbers of bandits and rebels
threatened domestic social order in the second half of the nineteenth
century, increasing numbers of villages and towns built walled
fortifications. In other words, the Chinese had no culturally based
opposition to military defenses. Their response to insecurity was
indeed very similar to that of Europeans in the waning days of the
Roman Empire. They built fortifications when and where they deemed
defense works desirable. For the vast bulk of the population across
the empire between 1000 and 1800 it simply turns out that city walls
were not necessary for the pursuit of economic activities, including
manufacturing.
Artisans in the late imperial empire chose to remain in villages with
little or no defense. Doing so certainly did not prevent the rise of
dense networks of markets for inputs and outputs. In fact, it appears
that such markets were central to the functioning of Chinese
handicrafts (Elvin 1973). It is also not the case that there was no
manufacturing in cities whatsoever, for jewelry, silk and other luxury
products seem to have been urban activities. In the lower Yangzi
region cycles of commercial expansion after 1000 created a
sophisticated marketing network and considerable amounts of
manufacturing, especially in cotton and silk textiles. The growth of
handicraft production was largely a rural phenomenon. Goods were
produced by agrarian households also engaged in agriculture or by
rural households who specialized in craft activities. Cities and towns
marketed more craft goods with a rural origin than goods of urban
origin (Elvin 1973: 268-84; Nishijima 1984; Tanaka 1984). As a
consequence, increased manufacturing did not lead to a corresponding
increase in urbanization.
The rural bias of craft manufactures does not mean that Chinese
entrepreneurs disregarded urban technologies when clear advantages
accompanied their use. Indeed, after the Industrial Revolution’s
techniques had diffused to East Asia, the Chinese predilection for
rural manufacturing waned. Neither then nor in China’s earlier history
can we find evidence for cultural or political hindrances to locating
enterprises in cities when new institutions and technologies made
urban-based production more profitable. The growth of urban-based
manufacturing in Shanghai during the first four decades of the
twentieth century makes abundantly clear that certain areas of the
country did shift from rural manufacturing to urban production. But in
China, as in Europe, these developments were unanticipated—in 1500
much less in 1000 no one knew that mechanization would succeed. There
were no reasons to create large industrial centers in China before the
nineteenth century. Furthermore, as in continental Europe, rural
manufacturing remained competitive, especially in labor-intensive
activities and where entrepreneurs could respond to urban innovations.
A good example of this phenomenon comes from the northern cotton
textile-producing county of Gaoyang where rural weavers purchased iron
gear looms to install in their homes. (Grove 2006)
For Europe the relationship between urban manufacturing and war is
complex. At first glance, one might even think that the dominant chain
of causation involves war causing destruction of both cities and
manufacturing. After all, the Roman Empire was based on cities. In
places like Gaul, Britain, and Germania new cities grew under the
imperial peace. These cities collapsed and many disappeared during the
Great Invasions only to revive slowly during the Middle Ages. It was
during this revival that the pattern of urban, capital-intensive
manufacturing came to become an integral part of the European economy.
By the Renaissance, the most urbanized areas of Europe were also those
where conflict had raged the most often: the band of territories from
Flanders to Rome, including the Burgundian estates, Western Germany
and Northern Italy.
From Charlemagne forward, as cities slowly reemerged, rulers focused
on providing security for skilled artisans. Strife, however, continued
and it made rural manufacturing a risky proposition in Europe, thereby
droving a larger range of manufacturing activities into cities where
protection was available. In contrast, the countryside was open
terrain for provisioning, thievery and wanton destruction. J.R. Hale
leaves little doubt that, "in terms of personal impact the burdens of
wars certainly afflicted the rural more than the urban
population."(Hale 1985:196) Although the images of towns sacked by
conquering armies have a great hold on our imagination we must bear in
mind that all military campaigns ravaged the countryside whether or
not they succeeded in capturing cities. Evidence is abundant that in
Europe the countryside was ravaged by warfare and that cities were
relatively spared (Gutmann 1980). While Parisians may have thanked
Saint Genevieve for protecting them from Attila, it is more likely
that the city was able to repulse invaders because of its walls. Paris
maintained its walls, and they would also defeat the Normands, Joan of
Arc, and Henri IV.
The history of Italian cities like Sienna and Padua highlight the
value of urban residence in times of conflict from the late Middle
Ages to the Renaissance (Caferro 1998, Kohl 1998). Padua faced both
civil war and the threat of outside invasion; Sienna had to defend
itself from the attacks of Florence and the raids of mercenary
companies. In both cases strife devastated the countryside but it
typically spared the city (Sienna was never conquered and Padua only
fell twice in one hundred years of conflict). Each invading army
seized whatever it could find in the fields and the villages.
Historians have noted the deleterious effects of such raiding on
agriculture because little could be done to protect farmland. In areas
like Italy, even villages were fortified in fear of localized raids.
But walls that were not supplemented by a large body of soldiers did
not afford much protection against a determined foe.
The siege warfare that prevailed in Flanders and the Low Countries
more generally, from the Hundred Years’ War until the peace of Utrecht
in 1713, also points to war’s differential treatment of town and
country. What made the sack of Antwerp in 1685 so surprising was that
the Spanish armies visited the kind of destruction on an urban
population that they and their foes normally imposed on peasants, but
it was certainly not the first instance in that conflict of armies
taking civilians’ property. From the point of view of merchants, the
sack itself was not a signal to give up trade, or to set up in the
countryside, but rather to seek a new, safer location in the Northern
Netherlands (Gelderblom 2000). That location, Amsterdam, quickly
became the largest city in the region. In manufacturing the movement
was less concentrated, but what the Southern Netherlands lost was
gained by Dutch cities (De Vries and Van der Woude 1997 279-334).
The opposing forces of war acting to reduce the scale of the economy
and of war pushing manufacturing in cities have made tracing out the
interaction between warfare and manufacturing difficult. In
particular, Acemoglu, Johnson and Robinson (2005) find no relationship
between war and the growth of cities. That negative result provides
support for the kind of balance our argument favors. Had cities’
provided superb protection or been systematically destroyed one would
have found either a clear positive or negative relationship. We are
interested in a more subtle and slow moving effect: how war
reorganizes the supply of manufactured commodities. This process may
well not affect the scale of cities.
If the general pattern of warfare and urbanization holds in Europe,
Britain presents somewhat of an anomaly. It is one that we must
consider because, after all, that is where the Industrial Revolution
occurred. With the departure of the Roman legions in 407, cities
collapsed and did not reemerge for a long time. The Saxon period as
well as the two centuries when the Danelaw was in effect could hardly
be called peaceful. While the Norman conquest may have been the last
successful invasion of England up to 1688, the throne of England was
hotly contested (including landings from Normandy) throughout the
Tudor period. Moreover, the border lands to the North were subject to
Scottish raids well in the seventeenth century. During this time
England appears to have been a heavily rural frontier of Europe (De
Vries 1984). It was not until the Tudors that English cities, and in
particular, London began to grow. Even then, as Wrigley, has pointed
out urban centers were few and small (Wrigley 1985). They were largely
administrative and commercial centers. Urban craft industries by
contrast, remained undeveloped because England was an economic
periphery whose main export was wool. London’s rise as the largest
city in Europe can hardly be attributed to insecurity in England since
there was little of it after 1600. In a country that was protected
from its enemies by a fleet rather than a standing army, manufacturing
did not have to locate behind city walls. The singular genius of the
British Navy may well have been its capacity to afford equal
protection to city and countryside—thus destroying the long standing
advantages of cities. Thus London did not afford better protection
from war than other towns or locations in Britain. Not surprisingly,
much of the early growth of manufacturing in England was carried out
in the North, areas favored by endowments of coal and where wages were
lower than in London. The pacification of England did not set off
urban industrialization; but rather a dash for cheap labor. As many
have pointed out, the early growth of manufacturing in England was as
much a rural as an urban phenomenon. But by the mid seventeenth
century the technological impact of centuries of urban manufacturing
was already large and England’s rural population was too small to
change the path of technical change.
Beyond, England, there is abundant evidence that in Europe the
location of manufacturing was indeed a set of marginal decisions that
varied over time. The key drivers to such change were the evolution of
technology, changes in capital labor ratios, and changes in military
technology. As a result the history of manufacturing location is one
that is different across the different polities of Europe.
Let us begin with the Low Countries. Although Van der Wee does not
detail the effect of the wars that ravaged that area from the
Renaissance to the 1720s, he does identify urban and rural activities
(1988). Three points are worth emphasizing: first, over time, urban
activities tended to become rural as entrepreneurs made every effort
to find methods of producing goods with fewer skills and less capital.
'New' industries were therefore urban but, as they matured they tended
to become rural. Thus prior to the industrial revolution, the urban
nature of manufacturing was not a foregone conclusion. Second, in the
absence of any urban response we would anticipate a fully rural
manufacturing sector and in some periods there were real declines in
urban manufacturing. At other times urban workers reoriented their
activities towards higher quality goods (implicitly higher skill and
higher capital). Third, during the period of the Dutch revolt, "the
armies ravaged the countryside, occupying and sometimes plundering the
towns and disrupting communications. For reasons of security [emphasis
added] and in order to have easier access to raw materials and
markets, many rural industrial workers migrated to the neighboring
towns (Van der Wee 1988: 347-8)." This last point emphasizes both the
negative impact of war (town and country suffer) and its differential
effect (people seek refuge in towns).
In the northern Low Countries the spread of putting-out industries
seems to have followed the vagaries of warfare. De Vries and van der
Woode document the spread of rural manufacture in Holland after 1720.
They view the near doubling of the proportion of non-agricultural
households as a result of population pressure but to our mind the
timing, after the end of the wars of Louis XIV, when the Low Countries
had been under constant threat of invasion is telling (de Vries and
van der Woode 1997:55-7). After peace ‘broke out’ in the Low Countries
entrepreneurs could more easily rely on a cheaper rural labor force
than in the uncertain times of the late-sixteenth and
early-seventeenth centuries. The pattern we see both in the Southern
and in the Northern Low Countries is not the inevitable march of
manufacturing towards capital intensive urban production. Rather, we
observe a secular competition between two modes of craft production,
one rural with low wages and low capital, the other was urban high
wage and more capital intensive.
The same story can also be told for England. While it may have been
the cradle of the industrial revolution it was first an area of
widespread putting-out, and that activity grew rapidly during the long
period of internal peace that followed the end of the Civil War.
Further as shown by Berg (1994), the putting-out industry remained a
strong competitor to urban/centralized manufacturing. In the case of
textiles, at least part of the expansion of industrial manufacturing
was a rural expansion, driven by the search for cheap waterpower and
cheap labor. The long period of institutional stability that followed
the Glorious Revolution (and in terms of violence largely runs from
1660) reduced cities’ security advantages so that the competition
between urban and rural manufactures was quite fierce between 1730 and
1830. The first response was the rise of the putting-out industries.
Later in the eighteenth century a similar phenomenon seems to have
taken place in the Low Countries (Gutmann 1980, Ch. 3) and France
(Vardi 1993).
For many the spread of putting-out industries in Northwest Europe was
a precursor to industrialization. Scholars have, in fact, dubbed it
proto-industrialization. From the technological point of view,
however, putting out was an altogether different path than the
industrialization that followed. Putting-out relied on the spatial
division of labor to produce large quantities of goods of moderate
quality. Inherently the organizational innovations that allowed the
putting-out industries to flourish were labor rather than capital
using and thus following a path that was quite different from those
that characterized the Industrial Revolution. Contrary to those who
see proto-industrialization as a step towards modern manufacturing, in
the light of our model, putting out was making Europe more like China
not less so. Moreover the Chinese evidence argues against any notion
that sophisticated rural manufacturing networks
(proto-industrialization) were critical precursors to sustained
growth. Both China and Europe had a significant labor force in rural
manufacturing, but only one region went on to develop industrial
technologies
The historical evidence strongly supports both the assumptions and the
implication of our model: warfare mattered and made European
manufacturing more urban. The effects of violence were contingent on
its intensity, on technology, and the urbanization of manufacturing.
Thus while over the long term they pushed entrepreneurs into cities,
these effects could easily be reversed. In the secular interplay
between warfare and manufacturing a surprisingly subtle rule emerges:
too much violence (as during the Great Invasions, the Thirty Years’
war, and other brutal conflicts) and manufacturing collapses; too
little violence and manufacturing runs to the countryside.
Long-term History through the Industrial Revolution
We must now move from asking how entrepreneurs adjusted to changes in
violence to investigating the consequences of these adjustments for
the path of technical change. So far, to keep the analysis simple, we
have developed a model that is static; it takes as given technology in
each industry and allows entrepreneurs to chose their input mix (how
much capital per worker) and where their shop or factory operates. Now
we turn to the consequences of choices of location on technical
change. To do so we borrow from the literature on induced innovation
that has derived how factor scarcity might affect the pace and
direction of technical change (Allen 2009, Habbakuk).
The argument is simple: where labor is relatively cheap (in our case
in the countryside) entrepreneurs will prefer to adopt new techniques
that are labor using rather than labor saving. Thus the demand for new
techniques that increase or decrease capital per worker depends on
relative prices. To be sure entrepreneurs are happy to adopt any input
saving techniques, but the relative demand will be greater for new
techniques that accord with relative prices.
The relative demands for different technologies translate into
technical change through one of two mechanisms. First, learning by
doing: in an industry that is capital intensive, entrepreneurs are
more likely to discover new processes that improve the productivity of
capital than that of labor. Second, conscious directed change:
investments in research and development that lead to new machines are
more likely to be undertaken where the price of capital is low
relative to that of labor. That is not to say that in the process of
industrialization there were no labor-using innovations—but rather in
Europe a larger fraction of all innovations were associated with
capital deepening than in China.
These two pathways are reinforced by external economies: Indeed, in
economies in which the bulk of manufacturing relies on little capital,
there are few capital intensive industries from which entrepreneurs in
other activities can learn about the value of machines. There will
also be fewer skilled workers who can build equipment and deploy a
varied set of solutions for adopting capital using methods in a
particular industry. On the other hand, in the same economy there are
many industries that manufacturers can observe to develop labor-using
improvements in their firms.
The importance of factor costs in inducing technical change has been
noted by many. Kenneth Sokoloff’s work on a radically different
distortion—agricultural seasonality—is particularly relevant because
of its spatial dimension. Sokoloff emphasized the importance of firms’
incentives to adopt and create capital goods (Sokoloff and Dollar
1997, Sokoloff and Tchakerian 1998). He argued that the need to bring
in the harvest created seasonality in rural wages as workers were
drawn out of other activities to work on farms for a few weeks in the
summer. Firms could either raise wage or shut down for the summer.
Where seasonality was intense firms had little choice but to shut
down. In turn, they avoided deploying costly machinery that would lie
idle for part of the year. Seasonality in his framework is a cost to
capital that acts exactly like Δ in our model. Because Sokoloff was
primarily interested in the contrast between the U.S. and British
economies he did not emphasize urban-rural issues. But other scholars
(e.g. Postel-Vinay 1994, Magnac and Postel-Vinay 1997, Van der Wee
1988) have done so and noted the lower capital levels of rural firms
even as late as the later nineteenth century and its close connection
with the variation in rural wages over the months of the year.
Sokoloff concluded that the U.S. deployed more machinery in
manufacturing early on than England precisely because agriculture was
less seasonal in America than in Britain.
More recently Robert Allen (2007) has put forth the argument that
relative prices played a fundamental role in the development of the
key machines of the Industrial Revolution. Only where capital costs
were particularly low and wages high did it pay to invent machinery
that would increase capital intensity several times over. These
conditions, he argues, prevailed in England after 1650 or so but
nowhere else. Allen demonstrates that after 1650, wages in England
(and particularly in London) were the highest in Europe. Conversely,
the cost of energy was remarkably low after 1700 because the English
were reaping the rewards of several centuries of technical adaptation
that transformed coal from a dangerous product into one that could be
easily used for home heating and in manufacturing. While differences
in capital costs may have been less, they too favored England. By
1700, Allen concludes that the rewards to adopting mechanized
techniques were highest in England, and that is why they were
developed there.
Our question does not involve the path of technical change during
industrialization, or why the key inventions were developed in England
but instead why the structure of manufacturing was so different
between Europe and China. In our view war’s concentration of
manufacturing behind city walls produced a series of biases that
raised the cost of labor and in particular unskilled labor, and in the
long run would lower the cost of capital by making capital markets
more efficient. These relative prices induced individuals to seek to
substitute capital for labor. In turn, urban entrepreneurs provided a
steady demand for specialized tools and later machines. Thus cities’
higher capital intensity was an important source of demand for
machinery and provided incentives to make more machines. In the
countryside such incentives did not exist.
Prior to 1400, the relatively high cost of capital throughout the
world combined with the limited supply of skilled artisans, made the
path of innovation daunting because the machines many inventors
imagined simply could not be built. In contrast, innovation achieved
by transforming a production method from using skilled labor to less
skilled labor and moving it to the countryside promised considerable
savings (This dynamic remains an important element of economic
activity to this day as the migration of world manufacturing to China
bears witness). No one in China or Europe could forecast in 1400 the
tremendous success we have had at creating capital using technologies.
Thus the Chinese path of rural handicraft is eminently reasonable. And
it should be no surprise to see that much of Europe’s manufacturing
followed the same path as we have seen for a long time Europeans
themselves were attracted to low wages/ labor intensive
manufacturing—after all the putting-out system is nothing more than
outsourcing beyond city walls. Hence, China’s technological path is a
very common process in economic growth; the deviation was that of
Europe.
Again, the development of European manufacturing highlights the
intensity and length of the competition between the two approaches.
The best evidence for this comes from French industrial surveys
carried out in the middle of the nineteenth century. At that time
seasonal manufacturing was so widespread that the agents of the French
Ministry of Industry gathered data about the phenomenon (Postel-Vinay,
1994 Magnac and Postel-Vinay 1997). Here two facts stand out. First,
urban manufacturers faced intense competition from rural firms. That
competition endured into the twentieth century, in particular in
labor-intensive product lines. Nevertheless capital/labor ratios of
rural enterprises were significantly lower than those of urban firms.
Within France, the regions where the seasonal variation in
agricultural wages was largest had the highest share of rural
industrial firms that shut down during summer months. It was also in
those areas that capital labor ratios were smallest. Over time, France
saw a co-evolution of agriculture and manufacturing, as increased
specialization in wheat in the eastern and central regions encouraged
seasonal manufacturing to locate there, while in the West
specialization in livestock did not provide many part-time industrial
workers. It was not until France began to mechanize harvest tasks for
its very large grain production that labor could move into permanent
industrial employment.
The second fact that stands out from the French data is that the rise
of rural manufacturing pre-dated the advent of severe seasonality in
agriculture. In the eighteenth century such seasonal labor migration
was small, and strictly local, because local agriculture was quite
diversified. Rural manufacturing may have begun to spread under Louis
XIV. Such an early start suggests that, for a large country like
France, the location of manufacturing was more sensitive to internal
disorder than to war. Indeed the Sun King came to power after the last
major revolt, the Fronde, had been put down but wars with other
countries raged almost continuously from 1620 to 1713. Those wars were
mostly not on French soil, and internal peace was largely maintained
until the Revolution. Interestingly, the number of rural weavers in
Northeastern France seems to have grown significantly as early as the
1690s, even though their expansion did not come into full bloom until
after the treaty of Utrecht (Vardi 1993).
Warfare thus proved to be a valuable irritant for economic progress.
By changing the share of crafts that located behind city walls war
encouraged the adoption of production techniques that were friendly to
further machine improvement. This included skilled artisans capable of
making parts accurate enough to avoid the crippling burden of friction
(Landes, 1983). For most of European history the center of these
developments lay in the continent. It began in Italy and over six
centuries spread through parts of Germany and the Low Countries before
coming into full bloom in England as the Industrial Revolution. To
examine the conditions that prevailed in England after 1700 alone
requires us to assume that the growth of skills and technical change
that occurred before was somehow different. Only those who are
terminally Anglophile would suppose that the forces behind improved
water wheels, the printing press, the pistol, or the knitting frame
are somehow different from those that led to the spinning jenny or the
steam engine. The key difference between these latter developments and
those that occurred earlier was economic value: the demand for cotton
textile and motive power is simply massively larger than that for
pistols or woolens (Clark). Although the magnitude of demand for coke
or cotton textiles explains the visible success of the new
technologies, it masks the fact that they developed in ways that were
very similar to those older, less economically rewarding,
technologies.
The technical breakthroughs of the Industrial Revolution are but one
step in a long process—one that was far more European than it was
English. Thus the study of England will allow us to answer some
important questions: for instance why was it that technical leadership
moved to England after 1650? But such a narrow inquiry will lead us
astray in considering why Europeans discovered the importance of
machines. To our mind the narrower question has largely been answered
by Allen (2009). As Allen has argued the relative price context goes a
long way towards explaining the specifics of the miraculous inventions
of the Industrial Revolution. Yet high English wages in 1650 do not
seem likely to explain structural changes whose most intensive locus
varied over time and that began in Italy in the late Middle Ages.
We see Allen’s analysis of the sources of high English wages to
indicate politics and warfare as major forces explaining
capital-intensive technical changes. Two key elements in his account,
the rise of the New Draperies (a more versatile and lighter wool
fabric) and the massive expansion of English trade were in fact the
result of political change. The rise of the New Draperies in England,
did depend upon a series of technical changes (that moved from carded
to combed wool to produce a lighter fabric), but one wonders why such
an industry grew up in a land abundant, labor scare economy that prior
to this period exported most of its commercialized wool to the Low
Countries. Given that the Low Countries were the dominant producers of
woolens and had all the infrastructure to weave and finish cloth, one
would have expected the new techniques to be deployed there rather
than in England. But as John Munro has observed, an English industry
arose because wars interfered with the market. On the one hand, wars
on the continent tended to reduce the demand for English wool, while
at the same time reducing the supply of high quality textiles in
Britain. Worse yet, the Crown had long relied on taxing English wool
exports, in effect protecting English artisans (Munro 2005). Finally
the move of artisans from the Low Countries and northern France to
England in the late sixteenth century is likely to have been spurred
by the instability provoked by the Dutch revolt and the French Wars of
Religion. Had England and the Low Countries been in the same polity
(as would have been the case in a China-like empire) the rise of the
new draperies in England would have been unlikely at the very least.
The second key element of Allen’s explanation is the capture of an
ever increasing share of international commerce by the English
commercial fleet. Yet the economic logic of London becoming Europe’s
entrepot seems farfetched since any goods unloaded there would have to
be reloaded onto a ship to cross the Chanel. Amsterdam would seem
better located. Of course the competition between Amsterdam and London
was not simply economic but also political. That there were two
Anglo-Dutch wars precisely at the time that London forged ahead is not
mere coincidence. That Rotterdam rather than London emerged as the
largest port in Europe after World War II is simply further testimony
to the distorting impacts of political competition on the economics of
geography. Rotterdam (like its forbears Antwerp and Amsterdam) is
simply much better situated to serve the European hinterland than
London. Its not such a surprise that the city on the Thames declined
as a transshipment point once the Royal Navy lost its relevance.
In fact one would do well to ponder just how long high English wages
would have persisted if politics had not made it difficult for English
entrepreneurs to locate their enterprises on the continent rather than
in Northern England. It is not much farther from London to Mons in
Belgium or Maubeuge in France than it is from the same city to
Manchester or York. It seems doubtful that English entrepreneurs would
have deployed their textile devices in high wage Northern England
rather than in the cheaper continental settings had they had that
option. Even more likely they would have avoided the costs of
developing such devices if they could have relied on the cheaper wages
that prevailed on the continent. Such traitorous outsourcing was
precluded by politics.
Just as one should not take the English pattern of technical change in
the eighteenth century out of its longer, European, context, one
should be wary of lessons learned by restricting the comparison of
political systems to China and Europe exclusively.
While there is no doubt that political competition altered the
location of manufacturing in Europe, it is also abundantly clear that
reaping the benefits of this alteration is difficult. In most points
of time and in most places, the destruction brought about b war simply
outweighed the positive benefits coming from either war’s relative
price implications or government spending on technology. A glance
around the globe will find many places beyond Europe where political
fragmentation endured and warfare was endemic. Southeast Asia,
Meso-America, and Africa between 500 and 1500, all come to mind. Yet
by 1500 when European contact occurred, none had embarked on the
transformative process that would produce the Industrial Revolution.
On the contrary, although abundantly endowed with valuable resources,
most of these territories were relatively poor. For Southeast Asia at
least, the evidence is consistent with the notion that warfare when it
occurred was very intense and very destructive of both persons and
private capital—much like the periods Europeans know as the Dark Ages.
We should also bear in mind that the expansions centered in Italy and
the Low Countries were brought to a halt by warfare and that the
Thirty Years War so devastated Germany that its economy spent much of
next century and a half in recovery. We conjecture that further
research may make more precise just what kind of political competition
is tolerable if one seeks to produce economic change.
Coda: China and Europe Diverging Greatly
The model of economic change analyzed above is not the first to argue
that political economy is essential to understanding why the
structures of the European economy departed from those of China
starting the Middle Ages. Many authors (Deng 1993, Mokyr, 2002 Diamond
1997, Jones,1981, Landes 1998) favor of Europe because political
competition there avoided costly and abrupt policy reversals as
occurred under the Ming. They also put politics before economics. Our
conclusions are starkly different: political competition unlike
economic competition is no panacea; the benefits of warfare, were
indirect, contingent, and secured at tremendous cost.
The narrative we construct from the model has several advantages over
traditional narratives. Because it is based on a very small number of
parameters, investigating whether its assumptions are reasonable and
its implications consistent with the historical record is easy. For
instance, if the cost of capital in cities and the countryside were
the same, we would have been hard pressed to maintain the argument.
But, as we have seen such cost differences did exist, and war
exacerbated them.
From a dramatic narrative point of view our approach has severe
drawbacks. It fails, for example, to point to specific actors as
responsible for failure or success: neither politicians nor culture
are responsible for China not taking the path towards mechanical
innovation. Indeed, in our view, China failed to do so because its
entrepreneurs had no reason to forego the advantages of handicraft
labor in the countryside. Similarly, Europeans can take little credit
for the countless discoveries that led up to the Industrial
Revolution. Ours is a tale without heroes or villains, in which the
unintended consequences of political conflict are what matter most. A
second drawback of our narrative from a dramatic point of view is that
it is not deterministic. War made it both more likely that Europe
would be poor (if war was too destructive) and more likely that it
would embark on the path towards capital deepening earlier than China.
In contrast, China was more likely to stay an agrarian handicraft
economy, but less likely to experience the Dark Ages or the
devastation that followed the Hundred Years War for instance. As
Needham and many others have shown, technology was far from static in
China, and it may well have been that given another several hundred
years or so machine invention would have sprouted there too. From our
point of view, the political economies of the far ends of Eurasia made
it significantly more likely that such processes would emerge at the
western end of the land mass than at its eastern end.
What makes for poor drama, though, might actually make for good
economic history. Indeed, it would be remarkably unjust to expect
Chinese governments of the Early Qing to implement policies promoting
a kind of economic change that Adam Smith, the foremost economist of
the eighteenth century did not even perceive. The Wealth of Nations is
not an ode to the Workshop of the World, it is far more an apology for
light taxes and unfettered trade in an agrarian economy. Those are
precisely the policies pursued by the Qing emperor. They were not
those of European rulers because the fiscal requirement of war
interfered with trade, an issue we will take up in Chapter 6.
If removing lead actors makes sense so does accepting contingency. And
this would be true not just for us, but for authors who advocate the
importance of endowments (Pomeranz, 2000 Jones 1981), or culture
(Landes 1998). Consider culture, it is the same social norms, religion
and ideas that first made China the most advanced economy by 1300,
then held China back before 1900 that must be permitting it growth by
leaps and bounds in the last three decades. How can a culturally
deterministic approach account for all this change?
This chapter has linked political economy with relative prices over
the very long term. There are other accounts of the impact of politics
on relative prices that also focus on the long term. Unlike our
framework which emphasizes differences in relative prices within a
particular geographic area, these tend to focus on differences in
relative prices across regions. The most eloquent exponent of these
arguments has been North (North 1981, North and Weingast 1989). In his
view, capital costs were lower in certain parts of Europe than
elsewhere on the continent and elsewhere on the globe because
political arrangements like representative government reduced the risk
of expropriation. The idea that growth was precluded in China by the
cost of capital has such a long lineage and its interaction with
political economy runs so deep that we devote the next chapter to this
problem.
BOX 4.1
Costs of production are C=wL+rK, if the entrepreneur hires L workers
at wage W and K capital at cost r. The Leontief technology is linear,
so the analysis can be carried out on a per worker basis. Costs are
then w+rk where k is capital per worker. As discussed above, wages are
higher in cities so wc < wu where the subscript c denotes the
countryside and u denotes urban areas. Capital costs are higher, so rc
> ru. A manufacturer seeks the lowest cost location. He compares Cc =
wc+rck versus Cu = wu+ruk. He picks the countryside if the fall in
labor costs (wu-wc) more than offsets the increased cost of capital
((rc-ru)k). This is equivalent to k< wu-wc/(rc-ru). Let k*= wu-wc/(rc-ru).
If capital per worker is less than k* then this manufacturer is in the
countryside.
Denote the unit increment in rural capital costs due to war as Δ.6
Rural capital cost in the warfare-prone economy will be rcw= rc+ Δ.
Now the manufacturer who decides where to locate examines not k< wu-wc/(rc-ru)
but k< wu-wc/(rcw-ru) or k< wu-wc/(rc+Δ-ru). This implies a threshold
capital intensity of kw*= wu-wc/(rc+Δ-ru). Clearly kw*< k*.
BOX 4.2
What is key is a, the factor share of capital. It is a measure of the
underlying capital intensity of the industry (if a is 1 then all
expenses are made on capital and if a is 0 then all go to labor). If
we look at entrepreneurs in a peaceful economy (labor cheaper in the
countryside, capital is cheaper in cities) we again find a threshold
value of a, a*, such that industries with a< a*are in the countryside.
Industries with a> a* are urban. We also find that industries with a
less than aw* will be in the countryside in the war-torn economy and aw*<
a*.
The first result follows by letting firms choose where to locate in
the war torn economy. Then, if we fix each industry’s factor
proportions to what it would be with rural relative prices, that
determines a first threshold value (aw*’) for moving to cities. If we
now allow firms to adjust their factor proportions when they move to
cities, all those who already wanted to move to cities will still want
to, and some who did not will, hence aw*< aw*’. The war-torn economy
(Europe) has an even larger urban manufacturing sector relative to the
peaceful economy (China) when factor proportions are adjusted to
reflect relative prices than when they are not.
Chapter 5
Credit Markets and Economic Change
By 1700, as we saw in Chapter 4, the seeds of a capital intensive,
machine using economy were sprouting in Europe. While this new sector
remained small through the eighteenth century, it was growing, most
notably in England. For the next hundred and fifty years China, unlike
North America or the European continent, made little effort to either
adopt or develop capital intensive methods of production. By the
eighteenth century, divergence had clearly set in and it would grow
for a long time. While some readers may be willing to accord that
differences in political structure played some role in moving Europe
towards machines and keeping China focused on its rural labor, more
would invoke differences in capital markets to explain both the
divergence and its persistence. As we shall see while there were and
remain important differences in financial institutions between the two
regions there is little evidence that credit markets failures were
responsible for the path of the Chinese economy.
Such a thesis may strike the reader as folly. After all China has
either been unwillingness or unable to develop financial institutions
that resemble those of the West (e.g. banks, equity and bond markets).
And the Chinese approach to financial institutions has often been
invoked to explain the Middle Kingdom’s failure to sustain the catch
up process in the waning days of the Qing dynasty or in the Republican
period. Regarding more recent history, some scholars have laid the
blame for the failure of the command economy under Mao on centralized
capital allocation and the consequent inability of anyone to
discipline firms. Since the reform period, Westerners have forecast
the collapse of the growth process because of looming financial
problems several times. In contrast, Europe’s early affection for such
institutions provided an infrastructure that was critical to the
spread of the new technologies. Hence, even if capital markets were
not responsible for the onset of divergence, their abundance in the
West and their scarcity in the East mattered beyond divergence. A
careful look at these argument will help us focus our inquiry
Economic historians, most recently Robert Allen, argue that
differences in the cost of capital were important in understanding the
location of innovative activity within Europe. As many have noted, by
the 1750s England had the most developed financial system in Europe.
Financial economists and economic historians have long argued that
differences in financial institutions imply differences in the cost of
capital. To understand the evolution of China and Europe before the
nineteenth-century, we must to consider three possibilities: the first
is that the cost of capital was lower in Europe because of the early
development of a specific set of financial institutions whose ideal
manifestation flowered in England after 1700. A second possibility is
that capital costs were lower in Europe than in China because the
region as a whole had developed a cornucopia of financial
institutions, starting in the Middle Ages, which lowered the cost of
securing investment resources. The third is that the cost of capital
depended simply on the high rate of savings in Europe. In each case,
the cost of investment resources would be low enough in Europe that it
would pay to deploy labor using technologies. The institutions that
affect the formation of capital thus represent the primary alternative
to our political economy thesis in explaining both the rise of the
divergence between China and Europe and its persistence.
Of course it could also be that the cost of capital was rather similar
between the two world regions. Certainly China’s ancient and extensive
water control projects are consistent with a long-standing willingness
to make and sustain long-term investments. Moreover one should bear in
mind the last thirty years’ record of massive rates of investment and
savings in China—a period during which Chinese financial markets have
certainly not taken a form that Westerners recognize as their own
(Brandt and Rawski 2008 Chs 1 and 14). These kinds of facts serve as
warnings to anyone who would accept the capital cost thesis without
careful inquiry.
At this stage of research, our knowledge of the institutions governing
capital in Qing China is at best embryonic. This is in part because
until recently, scholars seem to have been content to document that
China either lacked some specific Western institution or that it
failed to deploy such institutions efficiently and speedily once they
became available for adoption from abroad (e.g. Ma 2006, Goetzman and
Koll 2006). Yet new evidence about Chinese methods of capital
formation has come to light when scholars have been willing to look
closely at firm-level financing. Nowadays, as we shall see below,
scholars are looking far more broadly than before. In this chapter we
will review that evidence and suggest that the capital cost thesis
lacks much basis in fact. We will begin by examining the cost of
capital, including what seems to be damning evidence against China.
The key, we argue, is to compare the same kinds of transactions across
regions, because the rates we observe involve costs to borrowers
rather than returns to lenders. We examine the diversity of credit
market institutions in Europe to show that these responded to changes
in the demand for capital. We then turn to China to show that similar
patterns held there as well. In particular the second half of the
nineteenth century saw the rise of a variety of credit institutions.
Having considered supply, we turn to demand to show that, consistent
with Chapter 2, we expect demand for credit transactions to be lower
in China than in Europe and the borrowers systematically more risky.
We then return to political economy because one major source of credit
demand in Europe was public agencies (sovereigns, sovereign bodies,
and other public institutions) but in China such organizations played
virtually no role in credit markets until the mid-nineteenth century.
We look both at the positive benefits of government demand as well as
at it costs in particular the fact that sovereigns were risky
borrowers and that the resources they received were invested in war.
In closing, we return to a broader set of questions and suggest that,
absent political constraints, growth drives capital deepening rather
than the other way around.
Credit markets and the price of capital
By most accounts, interest rates in Europe were high in the Middle
Ages when credit markets were small. Over the long term, European
credit markets grew, and interest rates fell. The best evidence comes
from very long term mortages variously known as rentes, renten, rent
charges, or perpetual annuities (Schnapper 1957, Epstein 2001, Clark
2007, Van Zanden 2007). The decline that we can chart from the Black
Death forward led to a rough halving in the interest charge on
long-term annuities (from about 10% to 5% per year or less). This
trend began long before Europe industrialized and continued into the
nineteenth century when mortgages could be had for as low as four
percent and for terms of more than a half century.
There is little data on interest rates from the late imperial era and
historians of China often refer to conditions in the early twentieth
century to make more general claims. In this era pawnshops and money
lenders typically charged between 2 and 4 percent a month, though
examples of far higher rates have also been noted. The absence of
cheaper credit was cited by R. H. Tawney in the 1930s as a major cause
of indebtedness and ruin of peasant families (Tawney 1966: 58-63).
More recently, Philip Huang has echoed earlier concerns about high
interest rates saying that peasants engaged in borrowing to ensure
survival and thus tolerated far higher rates of interest than any
capitalist would. He further suggests that modern enterprises paid
higher rates of interest because of the situation prevailing in the
countryside of both north China and Jiangnan (Huang 1985: 189-90, 301;
Huang 1990: 108-110). Rates of 2 to 4 percent a month made borrowing
money extremely expensive, some ten to twenty times higher than the
costs of credit at the same time in Europe. If these rates held
throughout the economy and over time, then the conclusion is
inescapable: China was starved of capital and as a result, the path
taken by Europe was closed.
Yet one should not accept these conclusions at face value. In a market
without transactions costs the price paid by any buyer is the same as
the per unit income of any seller. But, as we know, there are no such
markets and what a farmer earns for a pound of rice is a far cry for
what we pay for it. Credit is no different. There are transactions
costs and they matter. Consider first what we might learn from the
data above if there were no transaction costs (in this context that
would mean no asymmetry of information and no differences in default
rates across borrowers, and that competition prevails). Then from the
demand side we know that the price of capital must equal to its
marginal product, thus a very high rate of interest implies that
capital is extremely scarce and thus productive. From the supply side
an individual who makes a loan foregoes using the resources and thus
the interest rate must be related to his impatience (mathematically,
if the individual discounts the future at the rate d, then the
interest rate, r, must be such that d=1/(1+r). If interest rates are
very high then individuals are very impatient.
For Europeans the decline in interest rates to 5% corresponds to a
period of significant increase in durable goods in probates (De Vries
2009) and in the capital stock of the economy (livestock, buildings
and so forth). Thus the quantity of capital seems to increase as its
price declines. For China, the evidence on quantities of capital does
not match the putative high price data. In a no transaction costs
context, high Chinese interest rates (100% a year) would imply massive
capital scarcity. Yet such a scarcity is not consistent with the
observed patterns of consumption or of investment. On the consumption
side China did not lack in the production of luxury goods and elite
households at least could have redistributed some of their consumption
to the future to take advantage of a doubling of their wealth each
year. Moreover we know that the Chinese made large scale long-term
investments. Water control expenses were large. Beyond the costs that
local officials and people bore to keep irrigation networks in repair,
larger efforts were needed to maintain canals and major dikes. In the
mid-eighteenth century the government would turn to the salt merchants
and expect major “contributions”(juan) to pay for especially large
repairs (Zhou Zhizhu 2002: 22). More general efforts to solicit
contributions for water control repairs were made in the 1820s (Tang
1987: 35-37). Mounting to several million taels on occasion, these
costs could reach some 2-5% of the normal annual expenditures of the
government. Such persistent investment suggests that the interest
rates quoted above are not a good indicator of the impatience of the
society. Even if interest rates in credit markets had been as high as
100% a year making borrowing very unattractive, there are other ways
to invest. In particular families will invest their own resources in
their own enterprises. If individuals are not very impatient, then
capital accumulation will proceed no matter whether formal credit
institutions do or do not exist. To be sure, investment will be larger
with capital markets than without, but very high rates of interest
should not persist for centuries unless there are fundamental threats
to the security of property.
Once we abandon the idea that there are no transactions costs and
allow transactions costs to be present, the interest rate received by
the lender becomes the interest paid by the borrower minus the
transaction costs. In this case, high interest rates paid by some
borrowers can persist even with large scale investment and low rates
of social impatience (consider interest rates on credit cards as an
example of such high rates). This is no surprise as we can see in
credit markets in modern economies where scholars measure spreads: the
difference between the price at which a bank borrows (what the lender
receives) and the price at which it lends (what the borrower pays). A
bank charges different interest rates for different types of loans to
different types of borrowers. But it is making these different kinds
of loans from the same pool of money hence the expected return across
types of loans must be relatively similar. We can cast these ideas
back into history to better understand credit markets.
Interest Rates Reconsidered
To explain why historical interest rates quoted for China were
somewhere ten times higher than those quoted for Europe we need to
understand differences between markets and differences among borrowers
within markets. Here what we mean by a market is involve a particular
type of loan defined by its collateral, duration, and any other
characteristic that might affect returns. Borrowers within a market
will differ primarily in their risk of default. Let us start with the
latter as it is simpler it is a necessary step to understand before we
move on to look at different kinds of markets.
Consider for an instant a lender in a given credit market who knows
his potential borrowers quite well—as might be the case in the
countryside or a small town in China or Europe. Some of these
borrowers are in safe activities and their current indebtedness is
low. Hence, they are extremely likely to make loan payments and to
repay. Others are engaged in riskier but more profitable activities
and thus, with some probability, they may be unable to pay off their
loans. As box 5.1 shows, the lender will charge these riskier
borrowers a higher interest rate than the safer ones. More generally
the lender will increase the interest above what he would charge a
perfectly safe borrower to offset any cost or losses that he incurs or
expects to incur as a result of default. Because part of the lender’s
costs are independent of the transaction’s size smaller loans will
face higher interest rates, and cost of capital, than larger ones even
if they are no more risky. This will be true whether the market is
competitive or not.
This is exactly what occurs in the contemporary mortgage market and it
is also something we observe in Early Modern Europe (Rosenthal 1993).
Hence one possible explanation for the differences in interest rates
between China and Europe is that the Chinese borrowers we observe were
riskier than those we observe in Europe. That is certainly so since
the rates we have been comparing are those paid for pawnshop loans in
China and mortgage annuities in Europe. In Europe at least individuals
who resorted to pawnshop were those who did not have access to
alternative sources of credit, while getting a mortgage required real
assets. That borrowers in pawnshops are riskier than borrowers in
mortgages stands to reason, and that must be part of the explanation
for differences in interest rates between China and Europe. More
importantly these are different types of loans and we must consider
the impact of such differences on interest rates charged.
Recall from the introduction of this chapter that until the last
decade or so the assumption that interest rates in China were
extremely high was received wisdom and no-one particularly bothered to
ascertain what kinds of credit markets existed there. If this pawnshop
and money lender situation were in fact the most important way that
farmers secured new capital, then the Chinese credit market could not
have been large and investment would have been severely restricted.
The argument about China’s financial market failure seems complete
even before we consider any further historical evidence. Recall that
in Europe, by the eighteenth century rates were between 4 and 8% a
year and loan costs of 1% a month were often cited as prima facie
evidence of usury. Those few observations that had been gleaned from
pawnshops seemed sufficient to verify the general thesis.
From our point of view, that is simply inadequate. Historically and in
contemporary Europe different credit markets charged different average
interest rate was accorded little significance. Indeed economic
historians have been more focused on the long run decline in interest
rates, and the differences in interest rates charged for the same type
of loan across locations, very little has been done to examine
differences across credit instruments. Nevertheless, European economic
historians would not dream of using pawnshop rates as an indicator of
the cost of capital. Doing so would seem absurd because that is the
type of market where the spread between the lender’s return and the
borrower’s cost is largest—most borrowers and investors could get
resources for less. If we accept for now that there are different
kinds of credit markets, we will have to be careful in distinguishing
between them. Box 5.2 sketches out a model in which borrowers chose to
enter one of two markets, one with collateral (mortgage) or the other
without (short term credit), depending on what kind of loan they need.
As we saw in the case of borrowers with different risk profiles,
interest rates vary systematically between the two markets because
they saddle borrowers with different types of transactions costs. In
our example mortgage markets have lower interest rates because the
collateral insures the lender in case of default (in other words he
does not have to impose higher interest rates to recoup his losses
from a default from those projects that succeed). However there are
fees required to lien the collateral so, for small loans, the borrower
will prefer to pay the high rates in the unsecured market even if he
has collateral.
So now let us return to our two examples: pawnshops and mortgages. For
China we lack interest rates for credit instruments that look like
mortgages. For Europe, however, we have evidence of both going back to
the Renaissance. Borrowers tended to turn to pawnshops when they had
exhausted alternative sources of credit like mortgages or reputational
credit. The reason was their high costs: while they carried a high
interest charge, pawnbrokers also added fees that made the annualized
cost of short term loans extremely high. These fees were required
because the lender also assumed control of the pawn, which was then
stored and no one could use it. Should the borrower want to retrieve
the pawn, the lender had to find it in this storeroom, otherwise it
would have to be sold. All the costs that involved handling the pawn
did not depend on its value. As a result pawnshop loans were always
very expensive and consequently few investment projects could be
plausibly funded with such loans.
In Europe, pawn broking was an opprobrious and regulated activity
precisely because everyone knew interest rates were massively higher
than for other types of loans. In renaissance Italy, for example,
Jewish pawnbroker regularly charged an interest rate twice as high as
what was legal for perpetual annuities (Botticini 2000). High pawnshop
interest rates sparked debate and in many places pawn broking was
taken over by regulated municipal organization like Italy’s Monte di
Pieta (Delille 2000), or Paris’ Credit Municipal (Hoffman et al
2001:255). Where such lending was left to the market, interest rates
were quite high. As late as the 1870s, interest rates for pawnshops in
England were well above 20% when mortgages were being had for less
than 5%. In the case of short pawns rate were massively higher than
even that (a 2% charge on a three day loan translates to a huge
annualized rate of interest) (Parliamentary Papers, report on
Pawnshops 1870). All of a sudden interest rates in China look quite
similar to those in Europe. Most likely the remaining differences in
interest rates came from variation between rural and urban pawn
broking. The rates we have for Europe all come from urban areas where
unredeemed pledges were easy to sell and there was more competition
among pawnbrokers. The rates we have for China come from more rural
settings where pledges were likely more difficult to sell and
competition lower. When we compare pawnshops with pawnshops,
differences in interest rates are not so substantial.
Pawnshop loans in late imperial China were not a standard credit
agreement for production and trade. Borrowers tended to turn to
pawnshops when they had exhausted alternative sources of capital, in
particular from kin and associates. What then of more normal loans? We
do not have data on the interest rates implied by the sale of
repurchase loan contracts or other loans backed by real assets in
China. We do have some evidence for commercial debts; interest rates
again appear to be in a range near, though not equal to, European
rates. Huang Jianhui has documented the loans made by native banks (zhangju)
in 1844 in three northern cities. These banks charged monthly rates of
0.38%, 0.4%, 0.45%; and 0.55%. He also suggests that Suzhou rates
(without specific years) were in the range of 0.6-0.9% per month. He
argues that Jiangnan and the south of China had higher rates than in
the north (Huang Jianhui 1994: 38-39). The range of these rates once
annualized runs from 4.5 to 11%. We can compare these to what interest
rates were in England and France in the 1840s. The long-term rate on
public debt was about 3.2% for the U.K. and 4.59% for France (Homer
and Sylla 1991: 197, 222). The discount rates set by the central banks
were 4 and 4.1% respectively (Homer and Sylla 1991: 209, 230). Both of
these rates were below the Chinese commercial rates, but they refer to
the safest long-term bonds and the rate afforded to the very best
commercial paper in either country (the Bank of France required that
all commercial paper presented at its window be endorsed twice as
added security). The rate on letters of exchange between London and
Paris was also in the 4- 4.5% range (Boyer-Xambeu et al.). If Huang’s
data are to be trusted, they suggest that interest rates in Chinese
cities were up to twice as high as in Europe, but most likely a good
deal closer to those in Europe. It is important to note that although
Huang’s data come from 1840s well after the onset of industrialization
in Europe, China remained relatively untouched by economic change at
that time. Hence it is credible to think that interest rates in this
period were similar to those that might have prevailed during much of
the preceding century. These years are also the last in China before
the onset of a long period of institutional instability that would
only come to a close in 1949, hence rates may well have been
relatively low and risen significantly afterwards. At the very least
these data beg to be supplemented by future research because they no
longer allow us to be comfortable assuming that China had dramatically
higher interest rates than those prevailing in Europe once we control
for types of loans.
Interest rate data available at present is doubly inadequate—there is
not enough Chinese data to support any serious assessment of the cost
of capital for investment purposes, nor do the rates themselves tell
us much about the institutional contexts in which credit transactions
were made. Thus, the interest rate data is clearly inconclusive. As
was the case of wages in Chapter 2, we cannot make the assumption that
markets are perfect; interest rates must be placed in a context. To do
so we divide the effort into two parts: first, we consider the supply
side of the market, second we assess demand. The supply side will
borrow heavily from the ideas of Chapter 3 because we will eschew the
simple idea that one set of institutions is optimal. Instead we argue
for the importance of diversity and change over time. Understanding
differences in demand for credit will take us back to issues we
encountered in Chapter 2. We move our argument forward beyond those in
these chapters 2 and 3 by focusing below not just on sources of
divergence, but also on what happened beyond the initial parting of
ways between China and Europe.
From interest rates to credit markets in Europe
We start with Europe because of its vast literature on credit markets
before, during, and after the early phases of industrialization. Much
of that literature focuses on the slow diffusion of modern financial
institutions and blames government agents for failing to provide the
pre-requisites for financial development: safe property rights, sound
public debt, a central bank, easy incorporation of private companies,
and low barriers for new financial players. Yet much of that
literature forgets that these modern institutions were not put in
place instantly. Credit markets have an old history that long predates
the rise of even the simplest of banks. As we review that history we
will emphasize the diversity of financial institutions, the adaptation
of credit markets to local political and economic circumstances, and
their capacity for change. While there are vast numbers of examples of
political intervention to limit the spread of ‘modern’ financial
institutions or to dampen competition, there is a veritable cornucopia
of examples of local markets expanding as economic activity
accelerated at different points in time dating back to the Middle Ages
for Europe as a whole.
Credit contracts are both ubiquitous and elusive in European archives.
They are ubiquitous because, as early as the Middle Ages, they clog
the archives of notaries and disputes over their execution crowd the
rolls of lower courts. They are also elusive because not all credit
transactions were preserved. In many cases, only fragments of the
original body of contracts remain for historians to pore
over—discharged contracts were often discarded. Moreover, as we shall
see, well into the 19th century European credit markets were
characterized by a bewildering diversity of institutions and contracts
and much the same could be said for equity markets. Here we focus on
credit because such contracts were far more prevalent than stocks in
households’ portfolios and because European loan markets were large
before the divergence set in. Equity markets grew over time yet most
of their development follows industrialization (prior to that time the
bulk of equity was not tradable). Thus, although stocks and their
markets may be valuable institutions for growth, they cannot help us
solve the riddle of why Europe’s economy forged ahead of China’s.
Europeans’ debts can be broadly classified into four groups that
correspond to different legal categories. While the importance of each
of these categories has varied considerably over time and space,
evidence for each of these can be found all over Western Europe as
early as 1000 CE. The first and simplest consists of private IOUs.
These were unsecured loans between private individuals and the
conditions upon which recalcitrant borrowers could be made to pay
varied greatly. In England, a variety of means, including debtor’s
prison, aided enforcement while in France (and other Southern European
countries) such extreme measures could not be used as easily (Luckett
1992). These types of debts are very easy to find in probates and
merchants’ accounts and were a necessary lubricant to economies in
which the local supply of physical money was both limited and highly
uncertain (Brennan 1997). Because farmers fell under the civil rather
than the commercial law regime, their notes were considered private
unsecured IOUs, making this part of the credit system important. If we
trust merchants’ accounts, these types of debts appear essential to
their business. When we look at probates or loan registration
documents such debts are quantitatively numerous but their value pales
compared to that of mortgages.
The second type of debt was commercial debt. Although most European
countries gave debtors some protection from creditors who had lent
them money in an unsecured transaction, there was an exception:
merchants. Commercial debt was the realm of commercial law rather than
civil law. It generally emphasized the rights of creditors and speedy
resolution. Merchants who did not pay their debts were typically
imprisoned until they came to some agreement with their creditors.
Even so, institutional variation was extensive with dramatic
differences in the extent of endorsement, the standing of
non-commercial creditors in courts, and the share of such instruments
issued by formal financial agents (banks). This type of credit has
received a great deal of attention both because it was key to the
commercial expansion of Italy, the Low Countries, and England and it
was closely associated with the development of merchant banks that
would give us the commercial and investment banks of the contemporary
era (de Roover 1953, Muller and Lane, 1997, Neal and Quinn 2004). The
maturity of the instrument (inland bills, letter of exchange, local
commercial IOUs) were typically quite short (one to three months), and
the size of many of these loans was small. Starting as early as the
thirteenth century, commercial loans included letters of exchange.
These contracts allowed a merchant to purchase a note in one city
payable in another, thus avoiding the costs and risks involved in
carrying cash from one location to another. Depending on whether the
merchant paid for the note upon receipt or upon his return from his
travels, credit was either extended to the banker or to the merchant.
Letters of exchanges were accepted at distant locations, because
wholesale merchants and banks formed networks that spanned Europe.
Nevertheless merchant banks tended to concentrate in the most
economically active areas, thus creating a correlation between banks
and economic change. Again, prior to concluding that banks caused
economic growth, one must consider the fact that many of the merchant
banks were simply the result of the increased specialization of
wholesale merchants into credit operations. A bank was born when
someone shifted his primary focus from commodities to finance. Such
specialization depended upon the existence of sufficient demand for
financial transactions. Commercial debts were not registered at issue;
thus it is difficult to form an estimate of the size of this market.
Nevertheless what documentation has survived in insolvency proceedings
and the collections of commercial families make it clear that such
credit existed throughout Europe by 1700 (Kindleberger 1984 ch 3).
The third type of debt, collateralized debt (mortgages) generally
composed the largest set of loans, by value. These could, of course,
be used for the purchase of land, but often the funds raised were
raised for other purposes. The institutions that made these types of
loans possible were also quite varied. In some places, at some times,
mortgages were drawn up by notaries; in others they were registered by
town secretaries, or manorial courts; and in yet others they were
prepared by attorneys as purely private arrangements (Hoffman et al
2001, Anderson 1969a, 1969b, Gelderblom and Jonkers 2008, 2006,
Servais, 1982, Pfister 1994). In some places, even before 1800,
information about land ownership and liens on property could be
recovered from public registries. In others, such information remained
the private property of intermediaries. The legal consequences of
default also varied considerably. In England, title to the pledged
land simply passed to the lender, while in the parts of the continent
where the Roman law legacy was strongest, complex and expensive
procedures of expropriation and auction were required to punish
borrowers. Although the geographical reach of the mortgage market
increased with city size, lenders and borrowers did not live far from
each other. Better than 90% of all loans linked borrowers and lenders
living less than 20 kilometers apart (Hoffman et al 2009b) In Western
Europe at least, the density of small towns meant that these markets
overlapped, creating, in effect, an integrated market for mortgages.
Despite (or perhaps because of) all this variation the sums that could
be raised with collateral were significant and because terms were
typically quite long. Iin the case of annuities they were indefinite
and on average were paid off after 15 years or so, but some endured
for several centuries. Thus these markets involved far more credit
than unsecured debt, and the amounts grew where land was valuable and
cities were large (Hoffman et al. 2008, Brennan 2006). They were also
responsive to change—when economies boomed so did mortgage markets.
There is no strong evidence that initially large mortgage markets
accelerated growth (Hoffman et al. 2009a). Nevertheless in many
economies mortgages and other secured debts form the largest stock of
outstanding obligations in any economy after public debt. The reason
it outweighs commercial debt is simple. Commercial debts are used to
facilitate transactions related to current economic production on a
less than three months’ basis so they are a fraction of national
income. Mortgages, however, are a fraction of national wealth since
they are based on the value of real assets. Those differences were
even larger before 1800 than later because on the continent at least,
agriculture lay outside the realm ofcommercial debt. This part of the
credit system has been understudied first because sources for England
are scarce and it is the English financial system that has served as
reference point, and second because it can promise neither the roguish
adventures nor the transformational investment of commercial or public
debt.
For the private sector of the economy the existence of these different
types of loans meant that rich Europeans had access to a variety of
credit markets even in the seventeenth century. To be sure the scale
of each market varied from place to place. So did the specific legal
rules and information techniques that constrained and sustained these
markets. From the perspective of the Dutch, most of seventeenth
century Europe may well have seemed underserved by credit
intermediaries and they would not have recognized the legal practices
that surrounded mortgages either in England (where they were privately
drawn up and not registered), or Spain (where they were drawn up by
notaries but not registered) or parts of Germany (where they were
notarized and also registered with lien authorities). Nevertheless,
all these markets were ubiquitous: absent political constraints and
absolute poverty they were set up everywhere in Europe.
At any point of time, the map of European credit markets had one or
two sinks where capital was relatively abundant and interest rates
low. These were also the most economically dynamic areas within the
subcontinent. Thus, in the late Middle Ages Northern Italy was both
the most advanced economy in Europe and the most innovative area for
finance, later both leaderships would pass to the Low Countries and
later still to England. This evidence has been taken as confirming the
hypothesis that good credit markets are a source of growth. While it
is obvious that the absence of credit markets would have slowed
growth, a careful examination of what actually happened typically
shows that, except where politics constrains finance, private credit
markets evolved in response to economic change rather than caused it.
The fourth and final important kind of European credit involved the
debts and financial assets of the public sector, by which we mean
cities, provinces, corporations, religious institutions including the
Catholic church, and most importantly, sovereigns (Tracy 1985, Potter
and Rosenthal 1997, Altorfer 2004, Courdurier 1974, Epstein 2000).
Except for sovereigns, almost all the debts issued by public actors
were long-term annuities, most often with an indefinite term (the
borrower could repay at any time). Sovereigns dabbled heavily in both
the short and long-term debt markets (Drelichman, 2005, Pezzolo 2005,
Epstein 2000, Quinn 2004, Gelderblom and Jonkers 2006, 2008, Hoffman
et al. 2001). As was the case with the other three types of loans,
these debts also had medieval origins. The primary motivation for
borrowing was the prosecution of warfare, or the raising of defenses
around cities. Public institutions like religious institutions, or
guilds, relied on the capital market either as lenders investing
donations from the public, or as borrowers when they needed to meet a
demand from the sovereign.
Public borrowers were innovators in the long-term debt market. Even
prior to the advent of public debt markets, rulers relied on credit to
fund their political competition with each other. This competition
involved extremely expensive warfare that they could not easily pay
for because they controlled only a small fraction of their economies,
and their peacetime budgets were limited relative to the expenditures
that rulers wanted to devote to war (Hoffman and Rosenthal 1997,
Dincecco 2009). The way sovereigns managed their finances, as Epstein
has argued, was deeply entwined with their decisions to enter or exit
international competition. As long as they perceived that their
participation in international affairs as temporary, they tended to
rely on expedients such as short-term finance and made little effort
to develop long-term financial markets, with the consequence that
their long-term cost of finance was high (Epstein 2000). Over time
more and more rulers came to the conclusion that conflicts in Europe
were long, and thus took measures to create institutions that allowed
them to borrow (Pezzolo, 2005, Velde and Weir, 1992 for an exception
Drelichman 2004). These institutions included central banks (who
initially acted to provide short term funds to the state), long term
bonds and bond markets (Dickson, 1967, Muller 1997), life contingent
claims and underwriting (Hoffman et al. 2001), as well as debt for
equity swaps (Quinn 2008).
One important theme in the history of these markets has been that only
some political and financial institutions allow sovereigns to borrow
long-term at low cost (North and Weingast 1989, Dickson 1967, Neal
1993, Epstein 2000). For England it is argued that the political
changes of 1688 secured a political equilibrium between the crown and
parliamentary elites (North and Weingast 1989). This was followed by a
series of financial innovation (a central bank and long-term bonds
traded on an exchange) that by the 1730s allowed English monarchs to
borrow at the most favorable interest rates in Europe. England has
often served as the model, leading scholars to ask why other countries
failed todo the same (e.g. Stasavage 2003). Recent research, however,
again emphasizes institutional variety and suggests that English model
was far from the only path towards a large low cost public debt. Most
startling is the province of Holland where debt management practices
followed a path almost opposite to that of England. At the start of
the struggle with Spain, Holland issued primarily annuities. Over
time, however, it developed a market for short-term obligations that,
by the mid-17th century, composed the bulk of its debts. Further, the
primary actors in this market were the province’s fiscal receivers not
bankers (Gelderblom and Jonkers 2008). Like private debt markets,
European public debt markets seem to have developed in response to the
demands for war. Their particular structures did reflect some
political constraints, but by the eighteenth century few states could
do without public debt markets and few did.
Precisely because some important innovations in debt markets began
with public bonds it has been argued that public debts, and in
particular those of sovereigns, were an important element in fostering
the aggregate growth in credit markets for several reasons. (Neal
1993). Their huge scale favored the rise of intermediaries who saw to
the short-term needs of the king, who found individuals willing to
hold the long-term bonds, and who created markets where these
long-term debts could be traded. Later, the same intermediaries
expanded their activities to cover the private sector. Although we
could just accept this argument as an additional element of a broad
thesis that political economy drove the divergence between China and
Europe, we want to be more cautious. To be sure, cheap credit from
modern intermediaries may have been important to the continued
development of the European economy after 1750, just as cheap American
cotton favored English factory production in the late eighteenth and
early nineteenth centuries. Prior to the mid-eighteenth century, one
would have to carefully quantify the costs of sovereigns’
interventions in credit markets; including the diversion of more
resources to warfare before coming to any conclusion.
This account has emphasized institutional variation and change over
time. This variation complicates the task of a comparative analysis
several ways. To begin governments were not concerned with
establishing statistics about private credit transactions. What data
were preserved depended on the legal requirements for registration
(either at time of signing or when the debt entered a judicial
proceeding). These requirements weighed unevenly across space and
time, as a result we must be careful to avoid problems of sample
selection. Comparing the size of markets (say the value of mortgages
per capita in England and France) is difficult because of differences
in registration practices. We have good information about some markets
(in particular public debt and mortgages) in some places at some
times. For many others markets we have only qualitative evidence and
are tempted to use differences in prices to come to conclusions about
markets.
The second difficulty we already encountered. One cannot equate
information about interest rates with capital abundance, because we
most often observe rates charged on specific instruments to specific
groups of lenders. The market interest rate thus depends both on the
capital abundance which drives the riskless rate and whom the market
allows into these particular loans. More specifically, it could well
be that England had lower interest rates than France for mortgages
around 1700 because capital was more abundant in England and the
mortgage market was more efficient. The difference in interest rates,
however, could also be due to entirely different reasons. It is quite
possible that at least part of the difference could come from the
concentrated property structure of England (Allen 1992:102-105,
199-200). Such a structure meant that little capital would involve
mortgages to small farmers. Because such small farmers have higher
risks of default than large ones, interest rates would be higher for
them, even if they were to draw on the same pool of capital (Rosenthal
1993). Lest one think that this is a purely theoretical concern, the
range of interest rates charged to different borrowers in one town in
the South of France (4.5 to 7%) was larger than the average
differences between England and France (4 to 5%) in the same period
(Clark 2005, Hoffman et al. 2001). Moreover, when tenant farmers
wanted to borrow in England they could not pledge land and thus never
appeared in the mortgage market. Thus the riskier part of the French
mortgage market simply has no counterpart in England. That interest
rates would be lower in England is thus not much of a surprise and
tells us little about the supply of capital.
The third difficulty we also encountered when trying to account for
differences between China and Europe. It comes from the fact that
different types of instruments play different roles in different
economies. The relative importance of annuities and short-term
obligations in English and Dutch public finance is an example of this
phenomenon. But the problem is quite general. Consider the more rapid
development of merchant banks in England after 1660 relative to its
continental rivals. This is precisely the period of London’s rapid
commercial growth and the spread of putting out in England, to be
followed by the development of manufacturing (Neal 1994).
Entrepreneurs who wanted to raise capital could not rely on the
mortgage market because so few of them owned any land. The alternative
was to use short-term debt to fund firms, including some long-term
activities. Contrast this situation with that in continental countries
where agriculture remained more important and landholdings were more
dispersed; under these conditions, there was a smaller demand for
commercial debt and a greater supply of mortgages—including to
manufacturers. As a result of these differences, should we expect
interest rates for mortgages or commercial debts to be similar in
England and continental Europe? To the extent that we want to make the
comparison based on prices we must compare like with like, and that is
a much harder comparison than has been undertaken to date.
Rather than conclude that Britain owes the Industrial Revolution to
its capital market, we might consider an alternative. Perhaps Britain
owed its capital markets to the combination of its initial
distribution of land and its process of economic change. In any case
credit markets did change and grow over time. For instance, the City
banks of London of the eighteenth century were preceded by
seventeenth-century goldsmith bankers. The kinds of business enjoyed
at a later date by country banks was most likely taken up initially by
merchants. In each case we can point to causes of change. The City
banks arose in response to the greater role of London in public
finance and international trade (Neal and Quinn 2003). Many of the
developments that we associate with financial development occurred
concurrently with industrialization rather than before. Notably this
included the spread of a network of banks throughout the country
linked to the key merchant banks in London after 1750. Industrial
equities were quoted on either regional or national exchanges only
after their firms had achieved substantial scale: the key investments
had already been realized (Mitchie 1989). To argue that financial
innovation caused the Industrial Revolution is hard to support
empirically. At the same time it is equally clear that had financial
innovations not emerged, the Industrial Revolution could hardly have
proceeded as swiftly as it did.
Absent major political obstacles, we observe responsiveness to changes
in demand for credit across Europe similar to what is first seen in
Britain. As industrialization spread to France, the country’s network
of banks grew (Hoffman et al. 2008a). Although the banking network did
not approach British density, that is only because notaries provided
important alternative services—the value of outstanding loans
intermediated by notaries was between a fifth and a quarter of French
GDP in the nineteenth century. While it is true that there were strict
restrictions on formation a joint stock banks into the 1870s and on
listing on the Paris stock exchange (Bourse), there were important
escape valves. Individuals could freely enter into private banking and
there was an active curb market for shares. If France did not have the
best financial institutions, it certainly avoided the worst and when
demand for finance increased there was a significant supply response.
Much the same tale could be told for the Low Countries, Germany, and
Italy.
Yet the remarkable variation in financial institutions has provided
fodder for inward looking narratives of national economic success.
Germany for example, has been said to have caught up with England at
the end of the 19th century because of universal banks (Gershenkron
1957, Calomiris 1995). While it is true that this statement cannot be
refuted since England did not have universal banks and Germany caught
up, it overlooks other salient facts. That these banks were a rather
small fraction of the whole credit system should have raised some
questions (Guinnane 2002). Similarly France’s failure to build a large
stock exchange and the slow diffusion of banks there has been taken as
a good explanation for its slow industrialization. That there were no
real obstacles to the creation of private commercial banks in France
and a rather low demand for their services in the countryside had to
be neglected for this narrative to be persuasive however. Fortunately,
recent research has been more nuanced, uncovering ways in which two
very different systems actually provided resources to enterprises.
Thus, the English merchant bank- stock market equilibrium was just as
capable of providing resources as the German universal bank
equilibrium, but it did so in different ways (Collins 1991). To be
sure, for a given industry or a given firm, one of these two solutions
may well have been preferable to the other but the optimum is unlikely
to have been the same for every industry. Similarly, at any point in
time one of the two equilibria may have been more efficient. Growth
occurred in Europe because each country’s system evolved to meet new
demands, not because any country was able to consistently achieve the
most efficient practices.
Rather than the national comparison so sharply focused on drawing out
the responsibilities of any particular financial system to economic
success, we prefer a different perspective. We believe that in a world
of intense competition, the survival of different institutions over
centuries should induce scholars to consider that they are probably
equally efficient or, at the very least, useful. Moreover, the
institutional differentiation of Europe did not peak in the 1750s on
the eve industrialization but more likely in 1913 on the eve of World
War I. As financial systems got larger and added savings banks, curb
and formal stock markets, central banks, pension systems, mutual
banking organizations, insurance companies, and yet other financial
intermediaries, each country chose its own path. Some were very
centralized; others maintained multiple exchanges. In some cases, the
share of the financial sector under public control was larger; in some
countries the non-profit sector was important. Hence element by
element comparisons are easy but rarely informative, because what one
part of the system could not accomplish another likely did.
Financial markets grew with demand; thus the level of credit activity
at any given point in time actually says little about the future
capacity of the system to provide loans when needed. What was true
before 1700 was also true after 1850. Industrialization was capital
deepening and it did make new demands on the financial system. The
response in terms of new financial organizations was slow but steady.
It was not until the 1820s, and by accident, that the Dutch king
founded what came to be known as the Belgian Société Générale—the
first financial institution designed to promote industrial development
(Van der Wee 1997). Yet, when industrialization began to raise capital
requirements for firms, a variety of formal capital markets were
already present. Thus although before the 1850s few firms took on the
joint stock form and even fewer raised capital through public
offerings European manufacturers could rely on traditional financial
intermediaries (merchant or commercial bankers) for short-term loans.
They also raised both equity and long-term loans from more traditional
sources of capital (business associates, friends, and family). As
Gerschenkron (1957) noted, firms were small in the early phases of
industrialization thus the financial sector could grow with economic
development. Because, prior to industrialization, a large fraction of
the flows of capital went unrecorded, focusing only on formal
institutions like banks will tend to overstate the growth of credit.
Finally, national politics played a critical role in shaping the
contours of a country’s financial system. In Western Europe at least,
that influence led to a diversity of institutions that was far greater
than their differences in efficiency. As we prepare to turn to China
we must be careful not to simply search out organizations that we can
compare to a German universal bank or an English country bank.
Instead, we must seek to find out if mechanisms existed whereby
individuals with excess capital could make it available to individuals
with demand for resources whatever form these mechanisms may have
taken.
China: Do Credit Markets Exist?
As we turn to Chinese credit markets we wish to see how economic
actors similar to those we have encountered in Europe financed their
activities (e.g. production and trade) in quite different social and
institutional environments. To begin we must ascertain what kind of
credit transactions and institutions existed. As we shall see, an
important difference with Europe was the absence throughout China of
any systematic registration of debt or equity contracts at the moment
of their execution. In that sense late-imperial China was more like
early modern England where debts are only visible in the archives of
some long-lived organizations or when they entered the courts.
Initially, investigators of credit in China largely confined
themselves to studying the twentieth century. Indeed scholars could
not imagine there was any economic growth of note before
nineteenth-century Western merchants began to connect China to large
scale inter-oceanic trade. Yet in the last two decades considerable
new material has emerged that challenges such simple assumptions. As
this section shows, there is no longer any reason to doubt the
existence of a variety of private financial transactions in China
before 1800.
That said, scholars have found far less evidence of financial markets
and credit institutions in late imperial China than in early modern
Europe. Nevertheless, if we reprise our typology from Europe, there is
abundant evidence that private debts (between two individuals) were
ubiquitous. Work on Qing dynasty legal sources suggests that debts are
one of the three large categories of disputes brought before county
magistrates by people along with disputes over land transactions and
marriage. (Huang 1996, Macauley 1998) The large number of cases
relating to debt in the Chongqing Municipal archives for the early
twentieth century confirms that such debt relations continued to be
important (Dykstra n.d.). Beyond private debts we come to those
between merchants. Institutions to provide credit in long-distance
trade also developed early on, though along different lines than in
Europe. As we discussed in Chapter 3, the spatial scale of the empire
gave an early impetus to long distance trade, particularly along the
coast and the major rivers, but also along the Grand Canal and
overland. This trade expanded after the fifteenth century through the
formation of complex merchant networks. These networks fulfilled
multiple functions but one of the more important ones was to provide
institutions to facilitate trade in an environment where space alone
made the formal enforcement of contracts quite difficult. Merchant
networks, which overlapped with kin networks, functioned as internal
capital markets and thus dramatically reduced the demand for formal
debt contracts between relative strangers. This hardly means, of
course, that formal debt arrangements did not develop in China, only
that they would, in relative terms, be less important than in Europe.
In both China and Europe, long-distance trade in the late medieval era
began with merchants traveling with their goods over long distances.
Those who achieved some measure of success established resident
operations in several of the cities where they did business. In
Chinese cases there was a head office (zonghao) and branch offices (fenhao)
(Niu 2008: 251-60). Each had its own management team and could raise
its own capital through issuing shares. In what ways the accounts we
have of their transactions included either a calculation, or even a
recognition, of implicit interest costs associated with the gap
between the time that money was spent on purchases and when it was
received following sales remains unclear. In contrast to Europe, where
detailed studies of account books are quite common, far less has been
published on this matter for China. Firms were composed of people
sharing native place, and often kinship ties. It is, therefore, likely
that they relied upon informal and poorly documented mechanisms to
support the credit required for their transactions. This is a second
important feature of the conditions for Chinese commercial credit that
deserves consideration.
In the eighteenth century Chinese merchants used promissory notes
drawn up in one city that could be redeemed at another location after
a specified duration of time—exactly like European inland bills.
Typically, these do not appear to have been taken by strangers, but by
people who were part of the regional merchant networks defined by
native place. (Ye Shichang and Pan Liangui 2004: 148-152) Some notes
were issued by what we call “native banks” in English, (qianzhuang,
piaohao), some of them were redeemable on demand and others had a
specified duration. A group of 23 such notes dating from the 1680s was
discovered in 1985; their stipulated lengths were as short as 20 days
and as long as 210 days. Almost all of the notes were between one
family and three other families of merchants from their locale doing
business in different parts of the empire. (Huang Jianhui 2002: 7)
Huang Jianhui posits the possibility of more of these credit
instruments existing, but notes the absence of direct evidence, which
he suggests means either knowledgeable scholars of the era didn’t
bother to write about such practices or that we simply have yet to
discover further examples of still extant notes. (Huang Jianhui 1994:
21) At a minimum, we know that the Chinese did have ways of providing
credit for long-distance trade, even if we cannot say how widely
utilized such practices were. Arrangements for credit were even
developed between Chinese and Western merchants at eighteenth-century
Canton (van Dyke 2005: 150-56).
By the nineteenth century we also have considerable evidence of native
banks in China (qianzhuang, piaohao). They took deposits from
merchants, officials and other wealthy people and arranged funds
transfers and credit for both commercial transactions and consumer
loans. The literature suggests that these networks of banks spread
after the 1840s to more and more localities but what is not clear is
how the same transactions were carried out in the centuries before.
These financial intermediaries were also needed to facilitate exchange
of copper coins for unminted silver of varying degrees of purity which
together made up the empire’s bi-metallic monetary system. When
foreign banks entered the scene in the second half of the nineteenth
century they worked successfully through Chinese native banks to put
their capital into the Chinese financial market. At the end of the
nineteenth century, however, Chinese financial institutions become
increasingly vulnerable to pressures in the political and fiscal
turmoil preceding the collapse of the dynasty in 1911 (Ye and Pan
2004: 190-203).
Turning to our third category, evidence of mortgages is, to be blunt,
non-existent in China. This does not mean, however, that legal
mechanisms for either full or partial alienation of real assets were
absent. As we observed in Chapter 3, there was active buying and
selling of land. Land was traded most often in rent-to-buy contracts.
These transactions allowed both the seller and his heirs to repurchase
the sold land provided they paid fair value for improvements, making
the transaction equivalent to a loan secured by the land. This type of
transaction was open to opportunism when land prices changed abruptly
because the terms upon which the land could be repurchased were
ambiguous and buyers at times found themselves making subsequent
payments to sellers. Qing law tried to make a clear distinction
between sales that were final and sales that were subject to
repurchase in order to simplify and bring order to the diverse
practices that began to multiply after distinctions between
conditional sales and outright sales of land began to be made in the
tenth century, if not before. Local officials ruled on disputes
between buyers and sellers of land regarding the conditions under
which the land could be redeemed and the amounts of additional money
required from the buyer to terminate the seller’s claims to the land.
Between the sixteenth and eighteenth centuries, disputes over
additional payments to original sellers increased. These cases were
especially common in the lower Yangzi region where commercial
expansion was greatest. They also can be found in other southern
locales. When a contract did not stipulate the finality or
completeness of a transaction, it was possible for the seller or his
heirs to redeem the land with some additional payment that represented
the market value of the land. For the buyer to complete his purchase
he would have to make some supplemental payments. Problems emerged
among parties regarding the period during which such redemption could
take place, as well as the number of supplemental payments to be made
by a buyer seeking to gain complete ownership. Magistrates sometimes
resolved these disputes by getting the buyer to pay the seller a small
sum, apparently in recognition of the increased productivity of the
land and commercial value of crops, even when there had been a
contract of outright sale. In such cases the magistrate aimed to
promote local social harmony and order and called upon a sense of the
party benefiting from market prosperity to share his good fortune with
the person who had previously sold him the piece of land (Kishimoto
2007). While scholars offer different views on the coherence and
effectiveness of eighteenth-century legal efforts to adjudicate
disputes over land transactions (Zelin 2004, Bourgon 2004), for our
present purposes the key point is that these disputes affirm the
existence of inter-temporal markets for land.
Lest one think of these contracts as utterly foreign to Europeans, one
should remember, that as late as the nineteenth century across a large
swath of Europe, the land market also included a right of repurchase
of land that had been transmitted through the line of descent. In
France this was known as retrait lignager (Diderot and D’Alembert
1751, Dyson 2003). To the narrow minded, these contracts appear
inefficient but are they any odder than leases based on three
consecutive lives, or the 99 -year leases with subtenants that were,
until recently, common practice in urban Britain? Cash rent on short
leases (3 to 9 years) only replaced life tenancy in England after the
structural transformation of the economy was well underway (Allen
1992:87-102). Other contracts seem to have functioned as a sale, with
a repurchase option known in France as vente à réméré. The ‘seller’
transmitted his land to the ‘buyer’ for a fixed number of years in
return for a capital sum. If the capital sum was not repaid in time,
the ‘buyer’ became the owner. Whether one considers these contracts
sales or loans, they are inter-temporal contracts. Customs such as
retrait lignager may have raised transaction costs but they did not
eliminate the market for land. The reader will surely notice that such
contracts are not as effective as mortgages for several reasons, the
most important being that the borrower had to give up control of some
of his or her assets to secure credit. Thus, if credit markets aim to
augment able entrepreneurs’ capital resources, contracts with
repurchase options are worse than mortgages. In a mortgage the farmer
can combine his entire holding with the cash he raises from the loan,
while in a contract with repurchase option he loses land as he adds
capital.
Our fourth category of credit also shows major differences between
China and Europe. Prior to the 1840s one would be hard pressed to find
much evidence of public credit in China. Except for episodic loans
from salt merchants, neither the Emperor’s central treasury nor local
administrations had much recourse to credit markets. When a region was
in need of resources for infrastructure projects, the imperial
bureaucracy simply changed tax flows. Part of a province’s taxes could
be redirected from Beijing to another part of the empire or adjoining
provinces could be asked to transfer resources to a needy neighbor.
Prior to the nineteenth century even military expenditures were met
with current revenues. As we shall discuss further below, China’s
capital markets did without the gains (and losses) from sovereign
debt.
While the evidence from China remains far less abundant than that for
Europe, it also shows that credit transactions were long-standing and
diverse. Future scholarship, less beholden to theories of economic
development that require European style finance is likely to add more
dimensions to Chinese credit markets. With the exception of public
credit, it is clear that in China and in Europe were firms that
distributed their equity claims in a manner more complicated than
small simple partnerships. The evidence is presently still thin, but
if subsequent research includes more systematic study of these
markets, we can expect that markets once thought not to exist or to
exist solely to channel land from marginal peasant households to rich
people will prove to have broader functions. We should also get a
sense of how these markets evolved over time as the economy changed.
Thus, in the absence of the political problems that sprang up in the
nineteenth century it could be that China would have found
‘indigenous’ financial institutions to speed its industrialization in
ways complementing those it might have adopted from the West.
Certainly, in the late twentieth century China has experienced a
massive rise in domestic investment with capital markets that are
quite different from those in North America, Europe or even other
parts of Asia. Therefore, we should seriously consider the possibility
that in the past as in the present, the Chinese met investment demands
through markets of their own design.
The view that the Chinese could develop capital markets in response to
demand is bolstered by taking into account evidence from after 1850.
To be sure, prior to the 1890s, there were no banks in China, at least
banks that a European could recognize. There was no obvious mortgage
market or securities exchange and the multi-owner firm had dubious
legal standing. These absences could well have been major stumbling
blocks to growth because by the 1880s when industrial firms began to
form, their scale was radically larger than that of private firms in
earlier centuries. Here, the reader will notice an important
distinction between credit and labor markets as discussed in Chapter
2. In that case industrialization simply made household structure
irrelevant as an ever-growing fraction of the population becomes
employees rather than entrepreneurs, all this could occur with
‘traditional’ labor markets. For capital, however, entirely new
structures were need if China’s firm size was to grow in response to
technical change.
Despite the political difficulties of the late nineteenth century,
China did create such credit markets in some locations, most notably
Shanghai. Some of the new organizations closely copied Western
examples but others reflected the importance of institutional
adaptation. Chinese financial institutions both emulated some foreign
traits and remained different from European practices. Beyond new
institutions, not all traditional structures of investment were
incompatible with the development of industrial production.
An obvious place to begin our inquiry is with textiles, the dominant
industry in terms of employment in most developing economies. Chinese
textile mills, at first spinning and then later weaving, were owned
under a variety of legal devices. One form involved corporate charters
granted by special decree similar to those granted European firms
before the 1850s. Other firms were owned in more straightforward ways
as sole proprietorships or corporations. In the later case, however,
much as in Europe, a single family tended to exercise control over the
business (Goetzman and Koll 2006). From 1890 to 1922 (after that date,
Japanese investment in Chinese textiles surged and the data no longer
represent domestic initiatives), the number of Chinese textile firms
grew from 1 to 95 and capacity expanded from 35,000 to 1.2 million
spindles. The story for weaving factories is similar: by 1922 there
were 27 factories operating more than 7,000 looms (Ding 1987). More
than half of these factories were Chinese owned. One might think that
the textile industry developed rapidly because of legal innovations
that transplanted the corporation to China, or because of the rise of
new financial institutions (Ma 2006). Indeed, Shanghai, the center of
the textile industry, in the late nineteenth century looks a lot like
an emerging market. As in many other places in the world, a stock
market opened in the 1880s. Its growth was hampered by a crash
following a bubble and the market did not re-open until the 1920s. The
failure of the Chinese stock exchange echoes the failure the first Sao
Paulo exchange. Both institutions were created in a boom, but shares
were so closely held that once the boom collapsed there was no
business on the exchange (Hanley 2005). In the case of Sao Paulo, the
market reopened within a decade as shares and in particular bonds
became more widely held. In the case of Shanghai the market remains
shuttered for three decade. Ma and other are right to point out the
failure of some European institutional transplants. The massive growth
of the city, however, suggests that there were likely Chinese
alternatives institutions and. means of securing finance.
Similarly, both native and foreign banks provided short-term loans to
manufacturers during this period. The introduction of modern banking
in particular has been claimed to be of great importance to economic
growth in early twentieth-century China (Rawski 1989). The importance
of imported financial institutions has been supplemented by arguments
that Western law and the more general institutional environment for
economic growth in Europe were needed to create modern growth in China
(Ma 2006, Goetzman and Koll 2006). But there are reasons to be
skeptical. First, a corporate code was enacted in 1904. While the
number of firms in the industry was growing and so was their use of
finance, the corporate code provided little stimulus to the creation
of new corporations (Kirby 1995). The failure of the corporate code to
have much impact may have Chinese explanations, but in a comparative
framework, it is not surprising. Unlike in the contemporary period,
when entrepreneurs rely heavily on incorporation, in the early
twentieth century many firms in many countries opted not to
incorporate. Limited liability was not a major issue for several
reasons. For one thing the desire to retain control tempered the
temptation to issue stock to raise capital. In addition there may have
been far less great need for the new form since the well established
partnership form may well have made access to loans easier (Lamoreaux
and Rosenthal 2005, Guinnane et al 2007).
The crucial importance of Western institutions for twentieth-century
economic growth is also brought under question by the financial
practices of firms outside of Shanghai before the twentieth century.
Shanghai may be a good locus to study how the Chinese adopted European
technologies and how they adapted native institutions. But the
important role of foreigners begs the question of how
industrialization might have proceeded in areas less affected by the
presence of foreigners. It implies that such possibilities were few if
not absent altogether. Beyond Shanghai we have two excellent examples
of industrial development. The first is in salt mining. The industry
required digging deep wells (a form of fixed capital) to collect
brine; it also required much working capital to evaporate the water
from the brine and yet more capital to market the salt. The industry
also required some skilled labor and management services. Nevertheless
these firms appear to have been relatively small. In this industry,
the Chinese deployed partnerships with shares (Zelin 2005: 342) that
resemble private limited companies in many ways. These organizations
were first enacted in Europe in the early nineteenth century
(Lamoreaux and Rosenthal 2005). European private limited companies had
joint stock attributes, but because they were not traded on exchanges,
their by-laws typically included additional provisions about control
and income. Chinese salt enterprises were strikingly similar. Because
the specific contracts that have come to light in China are heavily
concentrated in the last decade of the 19th century or the early
twentieth century, it is not clear to what extent the clauses they
contain draw solely upon native legal tradition or rely on legal
imports to deal with the changing circumstances in China.
Nevertheless, it seems that such multi-owner, often lineage-based,
firms had been in existence at least since the eighteenth century.
Moreover, the technical nature of salt-making seems to have changed
little. The Zigong salt-making firms were very successful--they
endured, they invested and their output grew rapidly (Zelin 1990).
There seems to be little specific to salt mining that would explain
the choice of organizational form, save that the investments were
durable and large. To the extent that industrial investments were of
the same kind, lineage-based firms were an available response—neither
the corporation nor capital markets were necessary. This industry also
demonstrated an ability to bring in Western steam engine technology
within the management and financial practices that had existed before
(Zelin 2005, Ch 7).
The second non-Shanghai case of adaptation also takes place well
outside the lower Yangzi region. It is the remarkable tale of the
Yutang pickle factory in Jining, Shandong (Pomeranz 1997). Like the
Zigong salt mines, it was initially a family firm. Like salt mines it
was remarkably long lived, having been founded in the 1770s. Unlike
many eighteenth century condiment makers, it grew to be very large by
the early twentieth century. Its history, as recounted by Pomeranz,
contains much valuable detail that allows us to push the analysis
beyond what the Shanghai textile mills or the salt mines of Sichuan
have established. Founded by migrants from Jiangsu, it was sold to a
partnership of locals in the early 19th century that grouped
individuals from at least two lineages. Further in 1827, management
was turned over to an employee, and it remained managed by a
non-family member for the rest of the century. In the 1870s the
general manager had to find new equity partners and issued interest
bearing notes to raise capital as one of the lineages decided to
reduce its investments in Yutang in order to buy some land. Then
around 1900, the two original lineages took the firm ‘private’ by
buying out all other investors, and one lineage assumed control. The
firm then branched out of the pickle business into local finance. If
we replace all the Chinese location names with English or French ones
and pickles with textiles, the Yutang story suddenly looks
unexceptional in a European context. From Shanghai to Jining, what we
know of the history of Chinese manufacturing leads us to conclude
that, absent the political difficulties that engulfed China in the
first half of the twentieth century, the financial system would have
evolved to fund its manufacturing growth.
There is of course an alternative reading of each of these three
cases, one that puts more emphasis on the political connections of the
players. Such connections were important both for early textile mills
and for the Yutang Company. One could also point out that nearly half
of all investment in Chinese industrial textiles was foreign by 1922
and that these investments were heavily concentrated near Shanghai
where foreigners had both the financial might to make the investments
and the military might to enforce their property rights. Only those
(foreigners or Chinese) with proper political connections could and
did invest. But such qualifications could lead us to miss the central
point of the examples for our topic of financial markets. China was
not an enterprise desert, nor was the legal structure truly limiting
to the formation of large enterprises. It may not have had a capital
market prior to the 1880s, but each of the examples above suggests
that there were important pathways for investment. As has been the
case in the last three decades, these pathways can act as very
powerful motors for investment when politics allow. We can see
however, from the very troubled history of China from the 1850s to the
1940s, that circumstances were rarely favorable to capital market
development. That Chinese entrepreneurs accomplished as much as they
did in the decades between the Opium War and the Japanese invasion is
quite astounding given the political weakness of their governments.
Since the mid 1970s another transformation has been taking place, one
that again features financial markets that most Western scholars view
as unstable if not outright dangerous to China’s growth. We refer of
course to the massive expansion of investment since the
liberalizations of the late 1970s. The critiques are multiform, but
the most damning involve excessive control over access to credit, and
the inability of banks to discipline borrowers who do not perform.
Yet, during the same time the Chinese government has created equity
markets, consumer banking, and allowed non-bank intermediaries to play
an important role in the provision of credit. China is entering its
third decade as the fastest growing economy in the world fuelled by
extremely high savings and investment. We might want to balance our
concerns about these credit institutions with some consideration for
the speed of their evolution. Clearly what was suitable for the
economy in 1980 when most manufacturing remained state owned and rural
reforms had just begun to pay off is radically different from what
China needs today.
A key difference between China and Europe: levels of demand for credit
Our analysis of Chinese credit markets suggests that we have yet to
uncover the full range of credit markets in pre-industrial China. Yet
the recent change in the scholarship is striking since it has stopped
seeking European forms in pre-industrial China and started to find the
indigenous solutions to common economic problems. Some demand for
credit was met informally within business forms and networks that did
not calculate interest rates as a basic part of their business making
practices. It is even more likely that much of the demand for credit
in the early modern Chinese economy could be satisfied within social
institutions and networks that did not require new specialized
institutions with particular efforts to document credit transactions.
It is also clear that investment resources flowed in other ways. The
state, as we will consider more fully in chapter 5, took a much more
proactive role than nearly everywhere in Europe investing in
infrastructure. The economic cost to the Chinese of not developing
more formal credit markets was likely quite small, especially when we
recall that much of the resources raised in early modern European
credit markets went towards the political and not economic purpose of
improving the arts of war. Nevertheless, as European economies began
their structural transformation away from agriculture, the role of
capital markets increased. If capital markets are not responsible for
the initial economic divergence, could they nevertheless be
responsible for China falling further behind in subsequent decades?
China had credit markets, but its formal markets were less developed
than European ones. We believe that a key reason for this state of
affairs was a difference in demand for credit despite comparable
levels of economic development. Demand for credit was lower in China
for both political and economic reasons. In this section, we consider
briefly why the Chinese state did not accumulate a public debt prior
to the nineteenth century as European states did, returning to the
subject of public finance in Chapter 6. We then look more closely at
economic sources of demand for credit.
While European empires were founded and survived on oceans of finance,
the Chinese empire was largely debt free until the intrusion of
Europeans into its internal affairs. The empire had three kinds of
expenses, all of which might have led to borrowing. These were
expenses to fund military campaign to preserve or enlarge the borders
of the realm, domestic administration, and economic development
projects. All of these would have led to debt in the European case,
yet none of them did in China. Over two millennia, Chinese rulers
faced two sets of military expenses, steady ones that involved the
defense of the empire and extraordinary ones that arose when the
empire had to be defended from an invader or reassembled after a
collapse. Most of the time, the empire was able to maintain a distinct
military advantage with current levels of spending, in part because
outlying populations were thin and not organized to put serious
pressure on the empire. Periodically however, the people living beyond
the Great Wall mobilized armies that could threaten major disruptions.
These types of threats typically brought dynasties to their knees but
they occurred very infrequently, separated by long periods of stable
rule. The incentive to turn to credit was probably quite strong when
regimes were tottering but in such dire circumstances few would have
been willing to lend and there was no credit market infrastructure.
Hence, Chinese rulers who were in need of quick revenues resorted to
similar actions as ‘despotic’ rulers in Europe. They manipulated
currencies and preyed upon individuals who had large amounts of liquid
wealth (Von Glahn 1996: 175-8). During the eighteenth and nineteenth
centuries specifically, they solicited “contributions” (juan) from
wealthy people to help meet extraordinary expenses, which became in
the later period increasingly military in nature (Tang 1987: 35-37;
Zhou Yumin 2000: 41-42). When a dynasty was stable, however, the value
of credit to military affairs was small. Then the Chinese emperor,
like his Roman, Ottoman, or and European counterparts (Charlemagne,
Napoleon) it preferred to run his campaigns out of current revenues.7
Not surprisingly, domestic administration in China was also funded out
of current revenues. Given that these costs over the whole of the
empire were likely to be quite stable, there was no reason to shift
their burden over time, in particular given the glacial pace of
growth. As long as episodes of civil unrest, environmental catastrophe
or other types of disturbances were local or provincial rather than
empire wide, borrowing made little sense. Instead, the empire could
easily shift resources from peaceful or prosperous provinces to
unstable or famished ones. Using space (transfers) rather than time
(loans) as a means of providing insurance was sensible and had the
desirable goal of binding the provinces together.
When we turn to the demand for credit for private investment in China
there are two aspects to highlight. First consider agriculture.
Irrigated rice was probably as capital intensive as any farming
activity undertaken in Europe, but most of the capital was in
improvements to water control and thus in a mix of public and private
hands. While Chinese farmers may have contributed all the labor to
maintaining ditches and dikes, they did not have to make all financial
investments. The major investments in Europe (draft animals) were
actually less prevalent in China. This is not to say that Chinese
farmers had no demand for credit.
A final difference between Chinese and European demand for credit lies
with handicraft production because, as we saw in Chapter 4, rural
locations figured more prominently in China than in Europe. Until the
very late eighteenth century rural sites of production were typically
more efficient then their urban counterparts. Rural industries in
China were typically pursued by households also engaged in farming;
they were therefore both very small scale and labor intensive. In some
cases the incomes from craft pursuits could equal or exceed those from
farming, but across much of the empire craft production was a
supplement to farming income. Some of the craft production was
processing crops, such as curing tobacco and tea, refining sugar,
extracting the blue dye from indigo plants or tanning leather from
animal hides. But a lot of production involved handicraft production
of final goods, in textiles and housewares (Zheng 1989). These were
labor-intensive and appear to have required little capital investment
in tools. For silk production, where we do know that rural households
likely needed credit to purchase needed inputs, Ming-te Pan has
carefully constructed a plausible household scenario based on
available data to suggest that even paying high rates of interest, a
peasant family was likely to be better off engaging in sericulture
than in their next best alternative (Pan 1996). For cotton textiles we
know that the division of labor meant that weaving households could
buy their needed thread and sell their finished product at local
markets (Elvin 1973). The efficiency of these markets for small-scale
transactions meant that credit transactions for cotton textile
production were less necessary than for silk. The rural location of
craft production also affected capital labor ratios. Beyond the fact
that Chinese putting-out markets may well have been dense enough to
minimize the capital invested in inventory, the structure of Chinese
manufacturing also reduced demand for capital. Indeed, as we have
argued in Chapter 4, the overwhelmingly rural nature of manufacturing
in China encouraged labor intensive methods of production. In China as
in Europe, rural manufacturing was less capital intensive and thus
less likely to be a source of demand for the credit market than urban
manufacturing. As we noted in Chapter 3, the Industrial Revolution
made it clear that capital intensive methods of production had a much
greater potential for raising productivity in the long run than labor
intensive methods of production, but this was not obvious in 1650 or
even in 1700. In any case, the demand for credit was low because rural
production took place near sources of input supply (so inventories
were limited), because all workers were family members (so no wages
were paid) and because rural artisans used less capital than their
urban competitors.
Both in agriculture and in handicraft production, individuals who
needed a loan for productive purposes, could either turn to the
market, or rely on their lineage connections. If as we expect, kin
members were well informed about each other, they would have been
happy to fund promising ventures. In this context, large kin groups in
China that we saw in chapter 1, would be superior sources of capital
for most individuals relative to the market. Within these kin groups,
the ability to enforce contracts (including implicit expectations of
future contributions to group activities), tilted capital flows away
from the market towards less formal intra group inter-generational
transactions. As long as the scale of activities remained relatively
small, the costs of using kin groups rather than credit markets, could
not be very large.
Because there was little public credit, lineage groups and clans
provided investment resources to their members, and the structure of
manufacturing reduced the demand for capital, credit markets were
smaller in China than in Europe. This would have been true even if
China had had the identical credit institutions as, say, England or
the Low Countries. But making such a large investment in financial
services when demand was low would have been inefficient.
Conclusion
We began this chapter by considering the commonly held view that
capital market structure is critical to economic outcomes and we have
now come, in effect, full circle. Rather than finding that structure
is paramount, we advocate instead greater recognition of different
types of markets in different places and different mechanisms for
producing investment. While it is likely that some financial
structures are more efficient than others, the lens of history is not
clear enough to allow us to discern which ones these are. To the
extent that we have wanted to explain the key differences in capital
markets across space we have had to move to more fundamental
processes, including differences in politics. For example, the fact
that traditional empires do not borrow has important consequences. But
there is also inequality in the distribution of wealth—highly unequal
societies are unlikely to create mortgage markets and more likely to
create reputational debt markets. Equally important are relationships
between households; extended kin groups can and do act as internal
capital markets. Finally, demand for credit is very important. The
Chinese empire with its internal peace and agrarian emphasis did not
have much demand for credit markets. Europe, whose violent politics
drove governments into debt and pushed manufacturing into cities, had
a higher demand for capital markets. When industrialization began,
Europe’s advantage over China would most likely have been
shorter-lived except for the tragically difficult dozen decades from
1850 to 1970 that China experienced. This is in part because China
could and in many ways did imitate the West, and it is also in part
because China could deploy different mechanisms to create structural
changes in the economy, none of which depended on capital markets.
While it may be that (as Robert Allen would have it) the relatively
higher price of capital in China discouraged machine invention and
innovation in the eighteenth century, it did not stop the adoption of
these machines at a later date. (Allen 2007) Thus, the explanation of
a difference at one point in the economic histories of China and
Europe may or may not have significance for explaining changes at a
later point in time.
To conclude, financial structure seems to have been of limited
importance to economic growth, at least before industrialization.
Whether one has large or small banks and large or small capital
markets, what matters more is the aggregate size of the financial
market. Moreover, finance most often follows rather than leads growth.
When processes of structural change arise they create demand for
financial services, and where political constraints are not
overwhelming, these demands are met either because old intermediaries
adapt or because new ones arise.
Box 5.1 Pricing credit
Individuals are different and borrowers know that. So it is an
interesting question to ask what kind of interests might prevail in a
market.
On the lender’s side we will assume risk neutrality. On the borrower
side we assume each individual has a idiosyncratic risk of default and
Q is the probability the loan is repaid while (1-Q) is the probability
of default. Some individuals will warrant larger loans (because they
have more collateral, or businesses with a higher expected cash flow).
The lender has access to a low risk borrower (the state or some local
institutions that serves as a reference point). Thus he can invest L
into a loan that has an expected net return of rs. In dealing with a
borrower he must invest C + Δl to learn about the borrower and what
she is going to do with the loan. In other words the borrower’s
expected return is  
Because the market is competitive, the lender is indifferent across
investment opportunities so  . The interest charged is then 
. Or  . In this context individuals who are more likely to
default, or get smaller loans, will pay higher interest rates. It is
easy to show that individuals who borrow for longer terms also get
lower interest rates.
Lenders all get the same expected rate of return independent of whom
they lend to. Borrower pay different interest rates depending on their
qualities, those are the interest rates in the contracts. If in a
given market Chinese borrowers take out smaller loans and are riskier
then they will have higher interest rates than their counterparts
elsewhere.
Box 5.2: From one market to many
Assume an individual has a project for which he needs a loan of size l
and that he will succeed with probability Q. In this case his return
gross of capital costs is R. He has the choice among several sources
of credit (family, network of friends, pawn, mortgage, bank) and has
to decide where to get financed. The borrower considers each market to
be defined by four characteristics (Li, ri, Pi ,ci,Qi). Li is the
maximum loan he can get in that market, ri is the interest rate, Pi is
the penalty he faces in case of default, ci is the transaction fee he
faces, Qi is the exogenous probability his project succeed if he gets
a loan of type i. The borrower only defaults for unanticipated reasons
(the project fails) with probability 1-Qi. The borrower may be shut
out of a given market (for instance because Pi=0 in a mortgage markets
when the borrower has no collateral).
For simplicity we consider a borrower who faces the choice between two
markets. One of which has no upfront fee because it’s a reputational
market (family or network of friends), the other is a mortgage market
where lenders must verify the collateral. When the borrower chooses to
fund his loan in the reputational market he will face the following
expected profit:  . If instead he gets a secured loan he
faces a different expected profit  .
It is immediately obvious that because there are no information costs
in the reputational loan market, for the borrower to enter the
mortgage market one of three things has to be true rm< rp , Pm< Pr, Lm>Lr.
Because the penalty in the mortgage market is the loss of the asset
which is transferred to the lender, interest rates in the collateral
market will be lower than in the reputational market. Second, the rich
borrowers who can post a lot of collateral can get bigger loans in the
mortgage market than it the reputational market. Hence different
borrowers will sort themselves among the different markets.
What about differences between China and Europe? It all depends on who
shows up in what market. The logic of Chapter 2 suggests that in China
larger households, and in particular, lineages would function as
internal capital markets while in Europe, the more limited extent of
kin would make more individuals dependent on kin. Thus individuals
whose families have fallen on hard times, or whose projects are viewed
as too risky to fund would be the dominant participant in China. Ergo
market interest rates in China should be higher than in Europe even
though average returns including projects funded by lineages are the
same.
Chapter 6
Autocrats, Wars, Taxes, and Public Goods
In literature and social science the rapacious despot is hard to kill.
In fiction, he survives countless defeats because the hero demands a
nemesis. In the social sciences, the despot endures because he serves
as a perfect foil for the virtuous political regimes that allow their
people representation. Indeed, the despot’s rapacity leads to leave
his subjects destitute while more liberal regimes promote the material
well being of their citizens. It is no surprise then that the despot
is regularly pressed into service to drag absolutist France and Spain
down behind the Netherlands and England (North 1981, Glazer and
Shleifer, Acemoglu Johnson Robinson). He is also enrolled to explain
why China fell behind the constitutional monarchies and republics of
Europe (Mokyr 1990, Diamond 1997).
In each case, the argument is the same: autocrats levy far more taxes
than is necessary to provide public services. Relying on the despot is
tempting. For at least the last two millennia, European fiscal
institutions have more often than not involved formal constraints on
the executive’s capacity to raise taxes. The most famous of these are
various assemblies of subjects that have come to be known as
representative institutions. To be sure, European rulers repeatedly
attempted to throw off their fiscal yoke, and in many places they
succeeded for long periods of time. Nevertheless, the contrast with
China is striking; its emperors never faced formal limits in fiscal
policy. Much the same can be said of the half century of Communist
Party rule. It is thus tempting to ascribe differences in economic
performance both across Eurasia, and within Europe, to the fiscal
consequences of political regimes (North et al 2009). We often imagine
that only representative regimes have the light taxes that liberate
growth, while the economies of autocrats are hobbled by high taxes and
low infrastructure. It seems the despot lives on.
It may be time to bury the despot. Indeed, little of the evidence
about fiscal regimes unearthed in the past quarter century is
consistent with the comparative perspective outlined above. China did
not fail because of avaricious emperors and Europe succeeded despite a
public finance system that was onerous and distorted economic
incentives. We do not take issue with the logic of the traditional
argument, in fact it will form the core of our exploration of
comparative public finance, but our review of the history of China and
Europe requires us to impose additional constraints on both
representative government and autocrats. As we shall see, those
additional considerations will first moderate, and in the end reverse
the comparison so that China’s emperors may well have enacted more
favorable tax policies than any European ruler and his representative
institutions. To carry out this analysis the chapter departs from the
analytic mode we have carried out so far, whereby a single central
theory is used to reconcile what are initially disparate and
contradictory pieces of evidence. Instead the chapter unfolds as a
dialogue between theory and history in which the model evolves through
three stages.
We begin by reviewing the basic logic of a heavy handed autocrat. We
then compare that model with the historical evidence to conclude that
two critical political elements—war or international relations, and
the capacity of the disenfranchised population to resist taxes imposed
from above—need to be added to the model. Rather than take them on all
at once we start with the most essential: international relations. The
traditional model focuses heavily on domestic spending, but
historically international relations often swallowed the bulk of a
sovereign’s expenses. War, in particular, proved extraordinarily
costly. Because rates of conflicts were radically different between
China and Europe war cannot be subsumed into a general public
expenditure. Including war in the analysis allows us to show that
while fiscal differences across political regimes decline, autocrats
still extract the most from their subjects. It will, however, prove an
important force in moderating the public spending of non-autocratic
states.
To limit how much revenue autocrats want to raise implies changing not
just their demand for revenues but supply (the political costs they
face when they raise taxes). This leads us to think explicitly about
how subject populations can resist arbitrary taxation. We borrow from
and elaborate upon Albert O. Hirschman’s important insights into exit,
voice, and loyalty. Among the strategies the subject population can
deploy to limit taxation we distinguish exit from voice. In exit,
individuals affect the fiscal system because they either migrate out
of the polity or move to the informal economy if taxes are too high.
Exit combined with war can indeed lead unfettered rulers to moderate
their taxes, in particular when the international scene is peaceful.
But this mechanism is inefficient when fiscal needs change abruptly.
We also consider voice in which the population influences the ruler’s
fiscal decisions either through revolt or formal institutions. We show
that if exit and voice are effective, then peaceful autocracies can
have lower taxes than any regime in a war torn region. We then use the
lessons of theory to understand how the fiscal structures inherited
from the Middle Ages evolved in China and Europe.
The long march of an idea
Throughout history, thinkers have considered and analyzed the nature
of despotism. By the era of the Enlightenment, Europeans had come to
associate the type of despotism practiced by Asian rulers as
repressive, backward, and heavily burdensome to their people (e.g.
Montesquieu, 1748, book XIII chapter XIII). The ‘philosophes’ also
entertained the notion of an enlightened despot who wisely made his
subjects’ well being paramount, an ideal with which to criticize their
own rulers. But by the beginning of the nineteenth century, Europeans
had conceived an alternative to despotism (enlightened or otherwise).
According to the new thinking, ‘modern’ and good regimes were
parliamentary monarchies that gave voice to some of the ruler’s
subjects. The rise of twentieth-century totalitarian regimes breathed
new life in the study of despotism. In a comparative context Karl
Wittfogel’s (1957) hydraulic societies may well have had the most
impact. Wittfogel’s despot owed his power to the environment,
especially its water resources. In economies whose prosperity depends
on water control, Wittfolgel argued, politicians can amass unbounded
power because they alone can guarantee the smooth flow of water. This
efficiency of rulers comes from fundamental features of water control.
Indeed this activity has extensive economies of scale and economies of
scope. The cost of water control does not rise with area covered and
it is best for the same agency to organize its different facets:
irrigation, drainage, flood control, and water power. According to
Witfogel, providing water control requires a bureaucracy that can
efficiently decide where new investment is needed and where
maintenance is required. This bureaucracy then becomes the primary
mechanism for repression because it is both well informed and designed
to detect deviations from normal behavior. The ruler can thus extract
massive tribute as long as life in the hydraulic society remains at
least slightly preferable to life on the unirrigated periphery.
Economic historians have reprised Wittfogel’s arguments and extended
them to cover cases of despotism not created by environmental
conditions. Douglass North (1981) famously argued that despotism could
arise simply because the technology of violence privileged specialists
who could use their weapons to gain control of productive assets.
Given this danger, one of the key purposes of representative
government in Europe was to force rulers to respect property rights
and refrain from confiscatory taxation. Hence, despotic Spain and
France were fiscally more oppressive and irresponsible than England or
the Netherlands. Comparative scholars, like E.L Jones, concurred with
North when they ascribed much of the differences in economic
performance across Eurasia to the failure of emperors to provide
adequate public goods (Jones 1981). Jones and more recently Jared
Diamond have stressed the abrupt halt of Chinese maritime voyages in
the 1430s due to a single central government decision (Diamond 1997).
In these various accounts, each author comments on the power of the
emperor and of the central Chinese government and offers a variant of
the imperial despot thesis, as if they had each penned a different
episode of a serial novel. The emperors, in these accounts all tax
excessively, fail to invest in useful public goods, and greedily
divert public resources to their own purposes like building palaces,
enjoying luxury goods, and enriching favored friends—all of which can
simply be called private consumption.
While there are many comparative models of public finance, one
formulated by McGuire and Olson (1996) is particularly suitable to our
analysis because it is framed around a dictator who is both rational
and rapacious. Rather than model simply a dictator and his polar
opposite of democracy, McGuire and Olson start with a simple premise:
political regimes can be analyzed based on the size of the faction in
power. A dictator is a faction of one, and regimes get more liberal as
the faction gets larger. The faction in power wants to maximize the
private returns to its members. In the extreme of a faction of one,
this model thus features a despot maximizing the net revenue from
taxation. The dominant group has the same two instruments with which
to achieve its goal: taxation and investment in public goods.
Citizens’ incentive to work in the formal sector declines with
taxation. In the most pessimistic case, when taxes increase,
individuals will simply consume more leisure; a slightly more
optimistic scenario has individuals evading taxes by moving to the
informal sector. In either case the economy suffers. Public goods make
the economy larger. It is important to note that the size of the
faction in control, affects the trade-off through a single channel, as
the size of the faction increases it owns more of the economy and thus
bears more of the distortionary costs from taxation and gets more of
the reward. All regimes thus face the same decline in the size of the
economy as tax rates increase and the same growth of the economy as
more public goods are provided.
The details of the model are provided in Box 6.1. The key results are
first that given any public good expenditure a dictator will choose
the tax rate that maximizes revenues (the rate at the top of the
Laffer curve). The intuition for this result is straightforward: the
dictator’s return is earned through the difference between his
expenditure on public goods and his total tax extraction—all else
being equal, the dictator wants to maximize tax revenue. As the size
of the faction that is in control increases, taxes fall because
members of the faction bear more of the cost of taxation (reduced
total output). The second conclusion involves how much to spend on
public goods. For the dictator the last dollar spent on public goods
must do much more; it must increase the size of the economy to such an
extent that it generates another dollar in tax revenue. As the size of
the faction in control increases, it realizes two gains from more
public goods, increased tax revenue and a larger private economy, thus
public goods increase as the size of the faction increases. Corruption
(the difference between what the government takes in and what it
spends on public goods) must fall as the size of the faction
increases. McGuire and Olson further showed that once the faction’s
share of the economy reaches a critical point it spends all the tax
revenue on public goods (corruption is eliminated). Increasing the
size of the faction beyond that point has no effect on taxation or
public goods. Because autocratic societies have higher taxes and fewer
public goods, their economies will also poorer than those of
democracies.
The contrast suggested by the theory fits nicely with the eighteenth
century evidence of the prosperous economies of England and the
Netherlands relative to the poverty of Austria-Hungary, Spain, Naples,
or France (North 1981). It also fits nicely with the nineteenth
century evidence of the prosperous economies of Europe and the poverty
of China. Finally, it also fits nicely with an ideology that more
representation reduces the extent of confiscation by elites to the
extent that taxes can fall at the same time that public goods spending
increases. But the theory’s predictions are certainly not consistent
with current evidence where repressive regimes have underdeveloped
fiscal capacities and prosperous democracies have very large public
sectors. It also does not fit as we move backwards in time. Nor does
the theory account for the range of historical evidence we consider in
this book. In Brief, china has not always been poorer than Europe and
the economic rise of both the Low Countries and England seems to have
begun long before their political transformations in the late
sixteenth and late seventeenth centuries respectively. Furthermore as
we shall see shortly, the rulers of the Middle Kingdom seem to have
spent considerably more resources on public goods than any European
ruler.
History strikes back
McGuire and Olson’s model is a mathematical exposition of the received
wisdom of political economists from the philosophes to the 1980s—yet
it is wrong. When scholars have investigated either tax rates or
expenditures on public goods the model has simply failed to stand up.
Attacks have come from all sides. China scholars have radically
revised Wittfogel’s thesis in favor a Qing imperial bureaucracy that
sought legitimacy in the provision of key public goods. European
scholars beginning with Mathias and O’Brien (1976) discovered that
taxes were actually lighter in countries like France and Spain than
they were in England and the Netherlands. Their findings conformed
with another of Montesquieu’s observations, “Taxes may be heavier in
proportion of the liberty of the subjects” (1748 XIII-12). In this
light the confiscatory behavior of absolutist monarchies is an
expression of their lack of power to tax. Had despots had full
capacity to choose their own fiscal rules, they would not have
resorted to inefficient confiscation. Because the evidence about the
burden of taxes is so often forgotten and some of it poorly known, it
is useful to review the history at some length.
Here is what we know about taxes and expenditures before 1800.
Government expenditures for states of the early modern era can be
divided among military expenditures, civilian administration, public
goods, and sovereign consumption. While the costs of the sovereign’s
consumption could look dramatic from the perspective of a single
household or an extended family, but these costs never loomed large
enough in any of the major European states or the Chinese empire to
exert a significant effect on fiscal decisions. Nor did the costs of
civilian administration weigh heavily, even though officials often
used their offices for their private gain. Public goods caused little
fiscal distress in Europe because there were so few and almost none
were financed from central government revenues. In China, as we shall
see there were more public goods and more of them were paid from the
imperial purse. These expenditures, however, were financed in ways
that made costs bearable and thus acceptable.
Chinese rulers viewed public goods as an important element in
maintaining social order and control. They were well aware that social
stability translated into political longevity. Social order, in turn,
was understood to depend on popular material security. These criteria
for political success made more sense in a large polity where external
enemies were few relative to the domestic challenges of sustaining
social order. In contrast, European criteria for political success
made sense because the spatial fragmentation of the region produced
relatively more external threats. Despite their understanding of the
value of domestic social order, European rulers had to face the fiscal
consequences of war. The costs of competition with other states
occupied a proportionally larger amount of government finances than
domestic rule.
The Chinese logic for successful state maintenance in the centuries
under consideration in this chapter was quite different from that
necessary for successful state formation in Europe. At heart it
emphasized light taxation and generally tried to avoid interfering
with commerce. For instance, there were few if any transit taxes or
other tariffs within China. Although the state nominally regulated
international trade in a restrictive fashion (leading to implicit if
not explicit tariffs) the reader should bear in mind that China’s
internal market dwarfed that of Europe as a whole for millennia.
States do, of course, need taxes to survive and in 1500 the Chinese
central government levied taxes on peasants in two main forms, grain
and labor service. Over the next three centuries both of these were
converted into monetary payments which made the movement and spending
of revenues far easier and more flexible.
Sixteenth-century Chinese taxes were collected in each of the Ming
empire’s roughly 1,100 counties. Agricultural taxes were divided into
two main categories, those that remained in the county to meet local
administrative expenses and those that were forwarded to the capital
or diverted to another part of the empire. Taxes sent to the capital
paid for central administrative costs; they were joined by additional
levies in grain from the rice rich provinces along the Yangzi River
that were sent up the Grand Canal to help feed the capital. Despite
its lack of a comprehensive accounting system, officials were able to
keep track of most of the revenues sent to the capital and those that
were left at local levels. Moreover, Ministry of Revenue officials
were able to move revenues around the empire among locales in order to
meet extraordinary needs, thereby reducing the need to tax at higher
levels within locales to provide resources (Wong 2009).
During the eighteenth century officials collected routine taxes
roughly amounting to some five to ten percent of agricultural output
and yet the Chinese state managed to help maintain waterways, manage
water control works for irrigation, and build massive granary reserves
among other projects that helped promote material security and
economic growth. To achieve these goals the Qing dynasty forged a more
tightly integrated bureaucracy to improve the flow of information up
and down the official hierarchy. For 1766, land taxes collected in
monetary form accounted for some 68 percent of routine revenues, salt
revenues about 12 percent, commercial taxes 11 percent, and
miscellaneous sources the remaining 9 percent. If taxes collected in
grain are expressed in terms of the monetary value and added to these
totals, land taxes account for 73 percent, salt 10 percent, commercial
taxes 9 percent and miscellaneous 8 percent of total revenues (Zhou
2002: 29). Land taxation rates were higher in the richer provinces and
portions of the revenue raised in these provinces were sent not to the
center but to poorer provinces. The two provinces of Jiangsu and
Zhejiang within which lay the Jiangnan delta, the empire’s wealthiest
region, accounted for roughly a quarter of the total agricultural tax
revenues of the empire but had probably less than twenty percent of
the empire’s population (Wang Yeh-chien 1973: 89-90). These revenues
were used to support military operations, build up granary reserves,
establish schools, and pay for general civilian administration
expenses. Although exact figures are hard to come by, the overwhelming
evidence is that taxes per capita were much lower in China than in
Europe
If the Chinese empire can be characterized as one of restrained
taxation, and limited fiscal innovation, nothing of the like can be
said for European polities. European rulers were always eager to
increase taxes; they relied on an amazing array of taxes and
ceaselessly invented new ways of squeezing revenue from their
subjects. Prior to the Dutch revolt, direct taxation, as in China, may
well have been the backbone of most European sovereigns. By the
seventeenth century, however, in France, Spain, and England the fiscal
frontier was indirect taxes (duties, excises and stamp taxes). In
fact, European sovereigns were ceaseless taxation innovators.
One striking European innovation was public borrowing. Indeed,
European rulers were never content to live off current revenues;
rather they sought to escape the unpleasant consequences of balanced
budgets by bringing forward future revenues. Here we will explore the
fiscal consequences of rulers’ desires for credit, having already
explored the consequences for credit markets in Chapter 5. We must
emphasize that, as economists know well, more borrowing does require
more tax revenue. Thus the long run development of public debt was
undergirded by the growth of public revenues. To take but one example,
that of France, the rate of growth of tax revenues per capita is
extraordinarily steady over the centuries from 1550 to 1850. Hence,
tax burdens which may have started out as modest levies rose
inexorably over time (Bonney 1995, 1999).
Beyond these generalities, one critical point stands out: taxes were
generally higher in regimes with representation than in regimes that
called themselves absolutist (and are generally understood as
dictatorships). Since Mathias and O’Brien’s 1974 seminal article
focusing on the comparison of England and France in the 18th century,
other scholars have verified this finding for a broader set of
countries and longer periods of time. The essays in Hoffman and
Norberg’s 1994 volume, for instance, make this point for the main
participants in the wars of the seventeenth century (England, France,
Spain and the Netherlands) from 1600 on. Richard Bonney and others
have assembled much more detailed fiscal data from a large set of
countries and have come to quite similar conclusions (Bonney 1999).
More recently the focus has moved back to the early Renaissance, in
particular due to the work of S.R. Epstein (2000). Epstein down plays
the achievements of parliamentary England by noting that Italian
republics had achieved remarkably good public finance by 1300. The
reason for this, of course, was that these republics had been
remarkably innovative in finance and in taxation, even though
representation was often limited to a very select few. European
economic historians have found time and again that representative
governments simply taxed more than authoritarian ones. Representative
governments were also more innovative in public finance: they extended
taxes beyond traditional sources like land, to move aggressively into
indirect and commercial taxes; and they were pioneers in issuing
long-term debt and creating markets where their bonds could be traded.
Many of these innovations were imitated by absolutist regimes so that
the advantages of any particular innovation dissipated over time.
Nevertheless the fiscal inventiveness of smaller, more representative,
states gave them an important advantage over their larger rivals.
China’s less aggressive fiscal policies do not mean that the emperor
and his staff were content to sit in some isolated palace and let the
population fend for itself. On the contrary the central government
aimed to influence local conditions through policies implemented by
both provincial and county officials. Within each of more than 1,300
counties in the eighteenth century, officials depended on local elites
to help them implement a Neo-Confucian agenda for local social order,
which included the repair of roads, bridges, and temples, the funding
of granaries and schools, and in some areas an even broader spectrum
of benevolent activities, such as orphanages and the care of widows.
At the core of the local elites were individuals who had studied for
the civil service examinations and consequently learned the same
principles for promoting social order as those advocated by state
officials. Because local elites made significant contributions to
local welfare, taxes could be collected in smaller amounts, and only a
fraction of these were kept at the local level. Considerable services
were nonetheless provided when elites met their Confucian duties to
fund and manage various local institutions (Wong 1997: 105-26). On
occasion it seems that elites preferred to manage local welfare
efforts without official participation (Mori 1969), but the more
common norm appears to have been a mix of official and elite efforts
with both finding the joint shouldering of these expenditures
reasonable. In some cases, imperial officials monitored local
activities in a routine fashion, for example, community granaries, but
in other cases there was little direct oversight (Will and Wong 1991:
63-69).
Little of these kinds of Chinese efforts can be found in Western
Europe. At their best, European states allowed local agencies the
capacity to raise resources in order to improve communications, or aid
public welfare (Bogart et al 2009). But European states were rarely at
their very best. From the end of the Middle Ages to well into the
twentieth century, rulers waged a relentless campaign to limit local
freedom to tax and to borrow, and to appropriate as many local
revenues as possible. The rise of a parliamentary regime in England
after 1688 did nothing to loosen the power of the center to control
investment in local public goods like roads, docks, and canals (Allen
2009). Even after 1688, local parishes had to seek parliamentary
approval to change their tax rates. Hence higher central government
taxes did not translate into higher rates of public goods provision.
The only exception is the Netherlands where the political structure
allowed cities and provinces to promote water control without much
interference (to be sure this is in part because there never was much
of a central government in the Netherlands). If public goods had been
any indication, taxation should have been higher in China than in
Europe. But taxes were lower in China. As we shall see the contrast is
not due to European kings thirsting for more palaces and personal
consumption than their Chinese counterparts, but to military
expenditures.
International Relations
In our first model, taxation serves two purposes, increasing the
incomes of the faction in power and producing public goods. One can
think of these two activities separately because their only
interrelation comes from the fact that they share a common revenue
source. When we consider war, the problem is rather different. On the
one hand war involves consumption for the sovereign; up to the French
Revolution kings liked to fight wars (Hoffman and Rosenthal 1997). On
the other hand, war has dramatic effects on the size of economy
independent of the fiscal burden it imposes. The economies of the
losers in war tended to shrink while those of winners tended to
expand. The other reason to take war seriously is that the military
dominated public expenditures everywhere but to an extent that was
radically different in China compared to Europe
To bring a framework like McGuire and Olson’s closer to the fiscal
histories of Europe and China we must we must take in the extent of
war, which, as many before us have observed, explains Europe’s high
taxes (Boney 1999). A less obvious consequence of war also matters:
war breaks the independence of the public finance decisions of
different polities. We can make the assumption that welfare spending
in France, for instance, is independent of that in England, but when
considering war that is not possible. Indeed a despotic regime
fighting a representative regime must raise approximately the same
amount of resources. Thus we cannot assume that France’s taxation
evolved independently of that of Spain or England (its key rivals); we
also cannot assume that the Netherlands’ fiscal system provided no
incentive for innovation in Spain when the two polities were at war
with each other. As we shall see war created pressures on regimes of
all types to raise taxes in Europe. In other words, the contrast
between Europe and China is twofold. On the one hand there was a
millennium of different experiences in the extent of warfare that
explains why in general taxes were lower in China than Europe. On the
other hand there is a contrast between representative and absolutist
regimes independent of war. How do these two realities come together?
We can begin to answer the question by considering that war is a
public investment for which states seem to have had to devote a
minimum of resources in order to continue to exist. The technical
details are exposed in Box 6.2. With no investment in the military the
economy simply disappears (it is taken over by someone else). Indeed,
an economy with no public expenditures on roads, school or welfare
still has some level of output. But, as the history of Poland makes
clear, a state without an army is doomed to be absorbed by another
regime. Then as military investment increases, the economy first
shrinks less, and with some sufficiently high investment it may
actually grow. Of course investment in the military also affects
individuals’ willingness to pay taxes and the returns to public goods
investment. What happens to the property of the inhabitants varies, in
some cases like colonial America, conquest was associated with very
high levels of confiscation by the conquering population. In the case
of Poland, however, the redistribution of private property was not
nearly so severe. Local elites who accepted their Russian, Prussian or
Austrian rulers kept their estates. In any case, since we find few
instances of conquest by popular request, we can assume that both the
ruler and subjects would prefer to conquer than be conquered.
Paying for the army thus requires that the faction in control to
adjust its spending on other priorities. First consider an autocratic
government. Recall that in a peaceful economy the autocrat was already
maximizing tax revenue. When war becomes an important consideration
the tax rate does not change. Investments in public goods, however,
will decline from their already low levels in polities where war is
not profitable and it will increase in economies where war is
profitable. On net, however, the autocrat’s diversion of public money
declines. In our model, as many others have also argued, political
competition does reduce corruption but only to the extent that
resources are diverted to the military. Beyond the case where there is
a dictator we must consider regimes where a fraction of the population
is in control. As the initial model, Increasing the size of the
faction in control increases still leads taxation to fall, while
public goods and war spending will increase. Finally the critical size
of the faction such that private diversion is eliminated declines
relative to an economy without war. Indeed for any faction size the
sum of resources devoted to war and public goods is larger in the
war-economy than it is in the peaceful economy and thus the total
expenditures curve must intersect the tax revenue curve (which is
still declining with faction size) earlier than it did in the peaceful
economy. For all regimes in which expenditures on public goods and war
are equal to tax revenue, war increases taxes, its effect on public
goods depends on just how effective the military is at protecting the
private economy.
War offers a first avenue for thinking about the fact that a peaceable
autocrat might provide more public goods than a more liberal regime.
To the extent that wars are very costly and unprofitable (the war torn
economies must spend a lot just to survive), then investment in public
goods will be very low. This helps understand why the Chinese state
could devote a larger fraction of its budget to public goods than
governments in Europe did. In Europe military expenditures were always
a very important element of the budget consuming, anywhere between 70
and 90% of the government’s revenues (Hoffman and Rosenthal 1997). The
reason is simple-- major European powers were about equally likely to
be at peace as at war. And even peace was uneasy and required
significant military expenditures. In fact, the history of the rise of
the state in Europe is written against a backdrop of military
expenditures. In particular the fiscal innovation we discussed in the
previous section were largely motivated by the relentless drive of
rulers to secure the resources necessary to pay their armies. European
rulers were granted tax increases and new sources of revenues for
specific military campaigns but then did their best to turn temporary
(extraordinary) taxes into permanent (ordinary) sources of income. War
not only explains high taxes in Europe it is also the central force
behind fiscal change.
Though more peaceful than Europe, imperial China was not without its
foes. Consequently, the emperor fielded an army that may well have
been the largest in the world for centuries. Under the Qing, the
military continued to consume a very large share of total public
revenues. The celebrated Great Wall was clearly a military
expenditure. The late imperial Chinese state maintained a military
presence along its borders and within parts of the empire. Moreover,
the eighteenth-century campaigns that took the armies of the Manchu
rulers into Inner Asia and led to the fuller incorporation of a far
greater territory than was held under the previous Ming dynasty were
expensive. As a result, the central government may have devoted a bit
more than fifty percent of its routine resources to the maintenance of
its army (Figures are difficult to assemble but in the mid-eighteenth
century, total annual military expenditures were around 23 million
taels and annual revenues around 41 or 42 million taels (Zhou Yumin
2000: 36-38). While the lax principles under which fiscal accounts
were maintained all across Eurasia mean that we must take these
figures as indicative rather than as exact numbers, they do suggest
that the fraction of a sovereign’s resources that were devoted to the
military was much higher in Europe than in China and that the
differences, once taken as shares of national output, were larger
still.
While half of a government’s resources going to the military may seem
large in today’s world, at least it left a significant amount to be
spent on domestic public goods. That was more than what Europeans
managed and, indeed, the contrast of fiscal resources available for
public goods in China and Europe was significant in the centuries we
consider in this chapter. While European rulers had to fund all
non-military activities out of less than a fifth of all tax revenues,
China’s emperor had a full half of his revenue to devote to public
goods, his bureaucracy or to his personal welfare. Moreover, until
1830 at least, China did not experience Europe’s ever-increasing
military needs. In fact, for long periods of time dynasties faced few
outside rivals. If Chinese subjects came to expect a relatively high
level of public goods, trouble on the frontiers could cause problems
of domestic unrest. War therefore seems an essential component of the
comparison.
It is also important to note that early modern European rulers’ thirst
for higher taxes did not come from a difference in ideology. Ideas of
good governance in Europe, as in China, emphasized low taxes, balanced
budgets and the provision of public services. Henry IV and his prime
minister, Sully would attain a hallowed mark of good governance in
seventeenth century France for restoring public order, encouraging the
revival of economic activity, guaranteeing the property rights of
religious minorities, and most amazingly, for keeping taxes light. A
century and a half later, the Burgundian estates attempted to decline
requests for further resources from Louis XV’s minister by gently
reminding the powers at Versailles that the only reason the crown’s
finances were precarious was that it had gotten involved in wars and
that prudent economy was the only proper way to run the country
(Potter and Rosenthal 1997).
Yet these French attempts to live up to ideas of good governance were
frustratingly rate. European rulers craved revenues because they
wanted to fight wars. To raise the resources for war they had to
confront the opposition of their subjects. Subjects had local
allegiances and were not always easily persuaded that defending some
other of their ruler’s possessions was in their interest. More
importantly, as Hoffman and Rosenthal (1997) emphasize, rulers seem to
have been the primary beneficiaries of successful wars, while their
subjects bore most of the burden of unsuccessful ones. The absence of
an imperial peace in Europe therefore meant that military expenses and
their attendant taxes were a continuous source of tension between
rulers and their subjects.
Nevertheless the existence of an empire with low taxes, and low levels
of military effort are neither particularly nor intrinsically Asian
traits. South-East Asia, in particular, remained a competitive and
unstable system well into the seventeenth century. Thai, Vietnamese,
Malay and Burmese rulers fought with each other as well as with
smaller groups of people across mainland and island Southeast Asia
into the early modern era. China’s equilibrium was only one among many
in Asia. Conversely, European kings were hardly the only rulers to
depend greatly on military activities. Empire builders like the
Ottomans and Mughals conquered additional territory through military
force and organized their territories in ways that allowed them to
support substantial armies. In fact, one could argue that the
political organization of these empires gave strong priority to
continuous political expansion. To be sure the Qing, like other
dynasties before them, expanded the territorial reach of China, but
their true genius lay in improving upon the system of internal
administration they had inherited to provideda broad range of public
goods at low cost.
European rulers were also well aware of the principles that promoted
good governance in China, even though they did not associate them with
the Qing dynasty. As noted in the French case earlier, such ideas had
a good deal of currency in Europe. Had European rulers followed the
policies recommended by the early economists, Adam Smith most famously
among them, they would have taxed their subjects at far lower rates
and thereby met some of the goals more actively pursued by Chinese
policy makers. They would also have eliminated many of the impediments
to trade so as to extend the market, and such reforms would have made
their domains more like China. Unfortunately for European rulers, the
revenue they needed to finance military and bureaucratic expansions,
so basic to state making of the period, meant that meeting Smith’s
hopes for low taxes was impossible. Perhaps the best that could be
done was what Smith propose—recognize the constraints on rulers to
achieve lower levels of taxation and think of ways to raise
unavoidable taxes with as few negative impacts on the economy as
possible.
Moving from a comparison of political regimes whose sole priorities
are domestic spending to one in which they also face external
competition offers several general lessons. To begin with, including
war in the analysis helps explain why European rulers persistently
spent more than Chinese emperors. But taking the despot as a fiscal
maximizer implies that tax rates should be higher in authoritarian
regimes like China than in representative governments like the
Netherlands or Britain. Indeed as long as we assume that autocrats’
greed for revenue is unchecked, and that they perch at the top of the
Laffer curve, democracy should levy similar taxes only in the most
extreme of circumstances. Having maximized tax revenue in a peaceful
economy, the despot must fund war out of resources that would have
gone to private consumption, and the tax rate will be similar in peace
and war. A second lesson concerns spending patterns. In any political
regime, war reduces spending to domestic public goods. Thus, that
China’s emperors may have allocated more of their revenues on domestic
public goods than European rulers becomes more understandable. Yet,
one final and major fact conflicts with the argument: China’s
autocratic rulers simply taxed less than their European counterparts.
Our framework will accommodate this fact if we introduce additional
costs to raising revenue. We will do so in the next section, but
before doing so we want to detail some additional implications of
interest.
The first of these implications is that political competition as it
played out in Europe in the early modern period was extremely costly.
Very little, if any of the higher tax revenues in Europe relative to
China and Europe went into useful infrastructure before 1700.8 In
particular, the beneficial effect of contemporary domestic political
competition mentioned at the start of this chapter cannot be assumed
to carry over to international relations. In fact, the very political
competition that was supposed to align rulers’ incentives had as a
first order consequence: very large military budgets. To be sure
political competition creates a check on rulers, because inefficient
politicians will be replaced or their polities conquered; but it also
requires resources. Although the number of independent polities in
Europe shrank over time, the size of the major powers, Great Britain,
France, Spain, Portugal, the Netherlands, Florence, Venice, Ottoman
Empire changed very little between 1500 and 1789. Central European
powers (Austria-Hungary and Prussia) did grow some and other countries
expanded their territorial reach outside Europe through colonial
empires. Nevertheless within Europe, few rulers eliminated despite the
fact that political competition grew more intense over time. To the
extent that we can find a return to political competition it can only
be indirect, say, in the development of capital markets or external
through colonial empires. War in Europe was expensive and proffered
little gain.
The second implication concerns democracies, or rather polities with
representative institutions: Their freedoms came at the considerable
price of very high taxation because they were small and because they
had to face larger, absolutists, opponents. Prior to 1800, the time
and expense of travel meant that regular representative assemblies
could only be sustained in small polities. It was easy to have regular
meetings of assemblies in Venice, harder in a French or Spanish
province, England or the Netherlands; but it was virtually impossible
for the whole of Spain. Given the limited variation of population
density and per capita income, small states had to levy high taxes
just to field armies that could hold their larger enemies like France,
Spain or the Hapsburg Empire in check. If we measure a country’s
economic success by the amount of resources available per person,
military success, to a large extent, depends on the absolute level of
resources one polity can bring to the field. Consider the case of
England and France at the beginning of the eighteenth century. The
population of the British Isles (England, Wales, Scotland, and
Ireland) was about 9 million while that of France was twice as large
(De Vries 1984). It may be that the British fiscal system was more
efficient (Brewer 1989) and British incomes were higher, but those two
effects were not sufficient to overcome the built in advantage of the
larger yet less representative territory. As a result in England, as
in the Netherlands or Venice, tax rates were driven by war and the tax
decisions made by autocrats who ruled large territories, not by any
domestic calculus. Even if their fiscal institutions had wanted to be
light, they could not but be heavy.
Considering war as an important element in the evolution of fiscal
institutions helps us to understand a final contrast in public finance
between China and Europe: public credit. China had, for all intents
and purposes, no public debt until the disastrous treaties that
followed the Opium War. Most of the money raised by Chinese debt
issues between the 1850s and World War I went towards paying war
indemnities; thus it had few benefits for the economy. In contrast,
European countries have been addicted to debt since the Middle Ages.
Overwhelmingly they turned to credit markets, not to finance public
infrastructure, but to pay for military expenditures. Debt was
attractive because it allowed rulers to bring forward future expected
peacetime surpluses and thus increase the scale of the military at the
time that it mattered most. This was particularly important for the
smaller ‘representative’ polities that fought bigger foes. It is
unlikely that the Netherlands would have prevailed against Spain or
England against France, without credit markets. For absolutist rulers
who had complete discretion on spending but limited leeway to increase
revenues on their own, the credit market allowed them to circumvent
constitutional limits on taxation (Drelichman 200x), As a result of
these differences in the importance of public debt, there clearly was
more financial innovation in Europe than in China. Yet war-driven
financial innovation was double-edged. War no doubt encouraged the
monetization of economies (simply because sovereigns preferred to get
their revenue in cash so as to pay for a professional military rather
than rely on feudal services). Such monetization certainly assisted
the spread of markets. War also no doubt encouraged the growth of a
market for public bonds and the shares of corporations whose primary
assets were also government obligations (e.g. the Bank of England) or
government-granted monopolies. Given the importance of scale in the
efficiency of financial markets, there were likely positive
externalities for private financial actors. Yet the downside of
war-driven financial policy is simply stunning. Europeans states
engaged in trade-distorting taxation on a scale unimagined in China.
They also quite frequently debased their currency, and in other ways
reduced the security of financial contracts denoted in official units
of account. Many of them also defaulted and engineered some of the
most spectacular financial crashes of all time, as occurred under
Phillip II in Spain or under the Regency in France in 1719-20.
Constraining the despot: Exit and Voice
While the development of credit markets can help us understand the
persistence of smaller, richer polities in Europe, they are not much
assistance if we want to explain why taxes were low in China. In fact,
economic forces alone are likely to be only a minor obstacle to rulers
who want to behave as revenue maximizers. Thus, the willingness of
individuals to take political action plays an important role in
limiting despotic taxation. Indeed, contemporary evidence suggests
that one can sustain hefty rates of taxes on labor (on the order of
30%) without a huge effect on formal work (in the U.S. the income tax
rate is above 20% state taxes are often on the order of 7% and social
security and other taxes add another 4 or 5%). If the formal economy
does not start to shrink much until taxes surge past 30 or 40%, then,
absent political constraints, a dictator would chose rates of taxation
at least that high. But except for resource rich economies today,
where dictators can tax heavily because they control exports, rulers
make do with much lower rates of taxation. In societies like China and
Europe prior to 1800; the bulk of revenues came from internal sources
and the cost of increasing taxation was substantial. The need to
secure at least the grudging willingness of the population to pay
taxes has profound implications (Levi 1989). Transposing Albert O.
Hirschman’s important insights about firms to government, we argue
that subjects and citizens can influence the fiscal plans of rulers
and officials in two broad ways that we call exit and voice (Hirshman
1970).
Exit for our present purposes refers to a variety of strategies that
deprive the ruler of revenue without confrontation when individuals
decide they do not like their leaders’ behavior. It includes people
migrating to another country that might offer a better mix of public
spending and private opportunities. It also includes moving into the
informal economy. A household might undertake either of these moves
for purely economic reasons, in other words to escape the burden of
taxes, but it might also do so if spending on public goods is too low.
This kind of passive resistance is unlikely to break a regime but it
proves costly in terms of resources and may be enough to persuade
rulers to keep their rapacity in check and provide more than the bare
minimum of services.
There are also other, more effective and more expensive, ways to get
the attention of the ruler: these we will call voice. They include
both institutions whose assent was required for new taxes like the
Spanish Cortes or the French Estates General; and organized protest
and revolts. To the extent that these kinds of mechanisms are
effective, tax collection and spending cannot get too far out of line
with the expectations of the population, regardless of whether the
regime is democratic or autocratic. The critical issue is that
individuals’ willingness to pay taxes depends not just on the tax rate
but on the political regime. In box 6.3 we model a situation in which
the decline in economic output due to increased taxation depends
directly on the size of the faction that controls the economy (the
dictator is a tiny fraction and full democracy would involve a faction
that includes at least half the population). While there are many ways
to model the impact of exit and voice on a fiscal regime, this turns
out to be the simplest. It also has the advantage that the gains from
political change are not exhausted as soon as the size of the group
that controls taxation and spending is large enough that it ceases to
steal part of the tax revenue. In doing so we forego the analysis of
possibly more subtle political interplay whereby autocrats may secure
more revenue by promising and delivering more public goods. If the
willingness of individuals to pay taxes rises sufficiently fast as the
political regime becomes more inclusive then revenues, then public
expenditures, and war expenditures will both rise with the size of the
faction in control. Still, taxes will be higher in war torn economies
than in peaceful ones because the logic we saw in the previous
section. Our iterative approach to theory has produced a different
model with which to compare China and Europe. It is kinder to China
than the simple dictator approach because it recognizes that peaceful
economies can provide more public goods than war torn ones (even today
the rate of infrastructure spending would surely slow in China if it
were to spend an equivalent fraction of national income on the
military as the U.S. does on the military). It also reaffirms our
understanding that the European political and fiscal innovations
associated with representative governments were important.
For some, implicit threats of exit were precisely what persuaded the
Chinese emperors to show restraint in taxation and it was the same
threats that dissuaded them from raising the revenue to meet the
political challenges of the nineteenth century (Mu Li 2002). In China,
a combination of low taxes and provision of public goods defined good
governance. By the Qing dynasty the imperial bureaucracy devoted
considerable effort to advertising its adherence to this standard and,
in particular, to insuring that the population understood that any
deviation from these principles was undertaken to fulfill a public
need. These efforts are easiest to see when it comes to expenditures
for water control. These expensive infrastructure projects would have
been ideal arenas for corruption if the government had been capable of
diverting resources gained from expanding public spending. Instead,
the empire organized water control efforts as ‘campaigns.’ Major
problems were addressed by bursts of bureaucratic energy and
resources. Zhou Zhichu estimates that routine river conservancy and
water control projects cost more than 2 million taels annually in the
eighteenth century (at a time when routine revenues were roughly 41 to
42 million taels annually). Special projects of extraordinary repairs
amounted to another 1.5 million taels annually. Finally, he estimates
the combination of routine and extraordinary repairs needed for the
seacoast water works in Jiangnan Cost another half million taels
annually for a total of some 4 million taels annually, nearly half of
which was non-routine expenditures (Zhou Zhichu 2002: 27). Campaigns
had advantages over routine projects. For the emperor, they allowed
him to mobilize additional resources, while his subjects were
comforted by the fact that the resources raised would go to a very
specific purpose rather than be swallowed in the mysteries of the
public treasury.
Campaigns were in fact used for several purposes. The expansion of the
granary system depended on periodic campaign-like efforts to mobilize
and store additional amounts of grain. These succeeded between the
late seventeenth and late eighteenth centuries in amassing hundreds
upon hundreds of tons of grain (Will and Wong 1991). Unlike water
control issues, there weren't ecological dangers created by the
granary system's expansion, though some officials pondered the
possibility of public operations with interfere with the market. As
for water control, creating routine maintenance and supervision could
prove more challenging than mobilizing men and resources in
campaign-like efforts to establish granaries or augment their
reserves.
The long periods of quiet in the Chinese empire enabled rulers and
subjects to rely heavily on such strategies. At the other end of the
continent, kings’ extraordinary revenues showed some similarities what
was raised in China as campaigns. Every early modern European
potentate filled his coffers with both ordinary and extraordinary
revenues. Extraordinary revenues were earmarked for specific purposes
and set to expire at some pre-specified time. Yet any resemblance to
China’s campaign is deceptive, Europeans rulers failed to develop a
workable fiscal system based on an understanding the funds subjects
provided would be allocated to useful purposes. This failure had
multiple causes starting with the fact that extraordinary revenues
rarely went to public goods provisions but rather for warfare.9 Kings
could not keep their promises to spend money on domestic public goods
because they were at war with each other so often. The high cost of
war in itself constrained their behavior to the extent that the
resources they allocated to the military and basic administration
always dwarfed their private consumption even. Even such a sumptuous
monarch as Louis XIV could have afforded Versailles ten times over and
cut taxes had he avoided the wars of the League of Augsburg and of
Spanish Succession.10
Moreover, resources did not go to any special administration but
rather they flowed into the public treasury. It is not surprising that
populations were loathe to consent to significant taxes for public
goods because they were well aware that these would likely be
appropriated for military purposes whenever war broke out. Furthermore
given the already high levels of taxation in these poor economies, the
bulk of the population may well have preferred to keep what income it
had for basic necessities rather than for better road or later
schools. The surprising fact about European history is how little
rulers were punished for redirecting resources to their preferred
activity: war.
Consider that rulers tended to expropriate part of the resources of
welfare providers (most famously Henry VIII’s nationalization of
Church wealth in England). For instance, in France by the seventeenth
century the primary provider of education and charity, the Catholic
Church, had been ‘persuaded’ to assist the treasury by granting
subsidies (the so called ‘don gratuit’). These resources were
supposedly designed to assist the crown in fighting infidels (Michaud
1991). When the crown re-allocated the resources to the general
budget, the Church did not cease to pay. While the negotiations had
the trappings of an equilibrium based on exist, the threat to stop
paying taxes did not materialize. Clearly then, Europeans had plenty
of occasions to practice ‘exit’ on misbehaving rulers, but the
resulting equilibrium was not the same as in China: revenues grew
ceaselessly and expenditures on war were rarely popular.
The likes of Louis XIV were not kept in check by the threat of exit.
Instead, Europeans relied heavily on voice and ideas about good
governance became linked, not to low taxation per se, but to “no
taxation without representation” as it was famously put in the context
of the American Revolution. European monarchs, even the most
absolutist ones, were not fiscal dictators because elites had voice.
Rulers faced serious constraints on raising revenues and in some cases
on how money was spent. While in some domains they could set taxes
with little restraint, most tax rates had to be decided in
consultation with representative assemblies. While rulers could govern
for long periods of time without consultation, they found it extremely
difficult to expand their revenues without calling an assembly. In
fact, revenue and expenditure institutions are remarkably important if
we want to understand early modern European public finance.
To raise money early modern European sovereigns relied on a variety of
strategies. At first, taxes were collected by individuals who were
required to provide services in kind to the state. Theoretically, in
this initial feudal equilibrium, the king’s sole revenues came from
his own estates, and he could call on his vassals to perform periodic
military service—most often capped at sixty days a year. This
arrangement proved unsatisfactory—military campaigns tended to last
longer than military obligations and vassals had limited incentives to
provide high quality services. Rulers were thus soon in the hunt for
more effective sources of revenues. They could obtain these by special
arrangements with some of their subjects, or more practically, by
bringing together the important players in the polity to discuss
fiscal matters. These discussions were institutionalized as Parliament
in England, the Estates General in France, the Cortez in Spain, the
Diet in the Holy Roman Empire and so forth. Henceforth we will refer
to these institutions as Estates. While there was tremendous variation
in who was represented in these meetings (e.g. mostly cities in Spain
and mostly barons in England), one thing is clear, these institutions
had important constitutional privileges regarding tax revenue. When
representative institutions perceived the policy objectives of the
crown to coincide with their own, they could loosen their purse
strings.
In authorizing new taxes, representative bodies could grant either
ordinary revenue (taxes that could be levied as long as the ruler
lived) or extraordinary revenues (taxes that would expire after a war
ended, after some specified period of time, or after some specified
revenue was raised). These revenues could include either direct taxes
on land, people and capital or indirect taxes (sales or transit
taxes). Once taxes were granted, a variety of systems were used to
collect the attendant revenue. In some places and at some times, taxes
were collected by salaried government officials, in others by private
firms that bought the right to collect revenue. In either case, the
investment in fiscal infrastructure throughout Europe was nothing less
than astounding, a testimony to rulers’ dedication to secure
additional revenues (Tilly 1990, Hoffman and Norberg 1994).
Despite the institutional investments of rulers, taxes rose slowly
because they faced a number of problems. To begin with, sovereignty
was a personal matter in Europe. As a result when two independent
areas came to be ruled by the same person, this did not necessarily
mean fiscal unification of the areas. Although Castile and Aragon were
united by Ferdinand and Isabella, for instance, that action only
committed the sovereign to a common line of inheritance. When their
successors wanted to raise taxes, they had to negotiate separately
with the assemblies of the different provinces—not only with the
Cortez of Castile and the assemblies of Aragon, but with those of
Catalonia and the Northern provinces as well. The process of fiscal
unification was slow. It was not completed, in many parts of Europe
until the nineteenth century (Dincecco 2005). This was no mere problem
of Southern or continental countries. The same ruler reigned
separately over the kingdom of England and that of Scotland until
1707. Unification with Ireland waited until 1801. The result of this
divided sovereignty was two-fold. First, it created a problem of free
riding--each territory wanted others to shoulder the cost of the
common good. The best the crown could hope for was to negotiate with
some large entity and then persuade the other ones to go along with
proportional increases. Success depended on strong popular support for
the crown’s fiscal aims. The second consequence of divided sovereignty
was that it raised the cost of negotiations with outlying areas. The
basic constitutional principle was that the sovereign had to travel to
the province and meet with its estates to request revenues. Obviously
if the crown’s goals were popular then such meetings were unnecessary.
But most often some, if not many, of a ruler’s subjects had sufficient
reservations to refuse to provide funds. Then, the ruler faced a stark
choice: travel to the recalcitrant province or do with a smaller
grant. As a result there was wide variation in the fiscal burden
across regions. (Hoffman and Rosenthal 1997, Beik 1989 Elliott 1986,
De Vries and Van de Woode 1997, Van Zanden and Van Riel 2004).
The fiscal equilibrium of divided sovereignty coupled with a
requirement of consultation of elites was very unappealing to early
modern monarchs. They resented the time it took to negotiate tax
increases as well as the oversight it implicitly gave to other groups
in society. In each country the crown attempted to change tack and
sidestep estates. Whenever it succeeded an absolute monarchy was
established. Under absolutism the crown did without estates and thus
could spend its resources with complete discretion. Doing without
estates was feasible because their meeting was at the discretion of
the crown, but it was costly as no regular taxation could be imposed
without some sort of consenting institution. Doing without estates did
not imply that revenues became fixed. Rather, the crown had to find
alternative paths to raise revenues.
The crown had some leeway in securing revenue because it was not only
the executive but also the apex of the judiciary (North and Weingast
1989). Its executive function gave it discretion over currency matters
in at least some of its domains and it also gave it discretion over
the organization of tax collection. Its judicial powers allowed the
crown considerable freedom to secure revenues. Indeed the judiciary’s
regulatory powers could be used to raise revenue, most famously
through the sale of monopolies “for the public good.”
European history thus shows that rulers faced different levels of
constraint on their expenditure and revenue decisions. If we abstract
from these different situations, it makes sense to think the same
provinces have the capacity to limit rulers’ revenues without having
the capacity to limit their rulers’ use of funds. In contrast, the
reverse situation in which provincial elites cannot set tax rates but
do tell the ruler how to spend is revenues is highly unlikely. We can
therefore limit ourselves to the set of regimes where expenditure
decisions are less constrained than revenue decisions. For any given
province we can think elites may have an effective check on taxes, on
taxes and expenditures, or on neither. Because kings rule over
multiple provinces, each with its own institutions, the extent of
provincial fiscal independence can vary. At one extreme a ruler could
have complete fiscal control in all his provinces. At the other
extreme, he could lack fiscal control over any province—he would in
effect be a parliamentary monarch. But the king’s power could also
vary from province to province, leading him to face partial
constraints on revenues and possibly partial constraints on
expenditures. This form of government has often been called
absolutism.
Consider the domains of Philip II of Spain. In Castile the crown
wielded considerably more power over revenues than it did in other
parts of the Iberian Peninsula and its efforts at tax rationalization
in the Low Countries were viewed as an affront to provincial
liberties. Resistance to new taxes led to the sixteenth-century revolt
and ultimately to the loss of the Northern Low Countries. Within the
Iberian Peninsula taxes were much higher in Castile than elsewhere and
an attempt at tax rationalization there led to the permanent loss of
Portugal (Elliott 1986). In France, William Beik (1989) and others
have documented that even Louis XIV had to negotiate far more
strenuously to extract revenue from provinces with representative
assemblies (Pays d’Etats) than with those that did not have such
institutions. Equally notable, Beik documents that in the case of
Languedoc the crown was forced to leave substantial sums in the
province for the provision of public goods.
It is important to note that although representation in Europe has
very old roots, its ultimate adoption as the standard mechanism for
fiscal decision making was challenged every step of the way.
Representation was a key element of rule in many ancient city states
(Athens and Rome but not Sparta). It was also an important element of
the political structure among the populations that invaded the Roman
Empire. Nevertheless it was a structure that most rulers did not like.
From the Roman emperors, to absolutist kings, to ‘parliamentary’
monarchs like William and Mary, at best rulers tolerated the
institution. Rather than quietly accept representation, both Charles
II and Louis XVI lost their heads (Rosenthal 1998). Europe may have
had an efficient fiscal mechanism, but it was not easily adopted
because it represented a fundamental weakening of the power of
sovereigns.
Representation was as unique to Europe as good governance was to
China. China, until the 20th century had no formal institutions
outside the imperial bureaucracy that would structure negotiations
between ruler and subjects over taxes. However, the Chinese were not
without voice: an important element behind revolts was the burden of
taxation and beliefs about the diversion of revenue into the pockets
of corrupt officials. Tax revolts were in fact an element of tax
negotiation that was common to both Europe and China. Revolts were
costly in terms of effort and danger for people to mount but they were
also expensive for governments to put down. This latter fact no doubt
curbed government appetites for taxation to some degree, even if in
the case of Europe a diet of light taxes could never really satisfy
the ruler’s needs.
We can use these ideas to reconcile our understanding of the political
economy of taxation across China and Europe. Rulers in China could
keep the taxes low because they could field comparatively large armies
with low tax rates and a domestic political equilibrium that equated
internal fiscal and political stability with a limited amount
diversion of tax revenues into the pockets of officials and the
emperors. The emperor could and did buy further good will by investing
in an array of public goods important to his mostly rural subjects.
Meanwhile in the competitive political economy of Europe, taxes were
kept relatively high by the need to defray ever growing military
expenses. Wars made it impossible for rulers to commit to any
significant program of domestic public spending—to the extent that
these activities occurred they were discharged by the private sector
or by non-governmental organizations like guilds or the Church. While
European rulers and Chinese emperors lived in luxury unimaginable to
their subjects, political mechanisms kept their take of the fiscal
system limited. While perhaps politically costly, Versailles and the
Forbidden City were fiscally cheap. The rise of representative
government did not lead to any decline in fiscal pressure, in part
because such systems were expensive, but mostly because they too had
to face the burden of war.
Reinterpreting History: Equilibria and Unforeseen Consequences
In the universe of economic possibilities available before the
Industrial Revolution, the rate of technological change was exogenous.
The main sources of economic growth were either market-based or
related to agricultural productivity. Given these conditions, the
Chinese fiscal regime of relatively low taxes and relatively high
provision of public goods compares favorably to European fiscal
regimes of relatively high taxes and relatively low provision of
public goods. It thus seems the Chinese fiscal regime was the superior
one for the economy.
Although the motivation of rulers and their populations might have
been similar, the different intensities of warfare led to long term
differences in fiscal structure. Most prominently, Chinese emperors
preferred to face the threat of exit than the constraints of voice and
they could do so because they did not face constant warfare. By 1300
European rulers, on the other hand, had exhausted such a strategy and
they had to rely on voice. These differences had important
implications for fiscal structures across Eurasia and for how they
evolved.
The Chinese achieved their bureaucratic regime of rule without
representative institutions and despite dramatic geographic variations
and cultural diversity. While Chinese historians sometimes express
skepticism regarding the efficacy of Confucian ideology, there can be
little doubt that in contrast to European political ideologies,
Confucian thought succeeded in providing simple and relatively
persuasive rules for the behavior of peasants, elites, and the
emperor. Because of these norms, local officials were aided by local
elites in raising funds for various projects. Such activities include
those that initiated by proclamations coming from the emperor to his
provincial officials, who in turn passed on instructions to local
county officials. In China, the eighteenth century witnessed rising
levels of public expenditures—if there was a state that sponsored
economic development anywhere in the eighteenth century it was the
Qing state—not Britain, France or any other Europeans state.
The Qing recognized that the geographic diversity of the empire could
be as much a source of strength as one of weakness. Officials were
members of a larger vertically integrated bureaucracy. Those serving
in more developed provinces were required to coordinate decision
making with officials in other provinces and in the capital to create
both routine and extraordinary flows of resources to poorer areas.
Without these resource transfers it is difficult to imagine how the
Qing state could have succeeded in consolidating its frontiers. As a
result, officials at least implicitly divided the agrarian empire into
three zones. The most economically prosperous provinces produced
fiscal surpluses to be used in the poor and landlocked frontier that
formed a second kind of zone. A third (intermediate) zone consisted of
provinces that utilized resources and techniques to replicate many of
the practices initially developed in the first zone; it filled in the
spaces between the empire’s most commercialized core and the frontiers
(Wong 2000).
In Europe it was the development of voice rather than exit that
produced growth in the provision of public goods (and of military
resources). In the previous section we argued that constitutional
constraints forced rulers to accept voice however reluctantly. Given
that negotiations over budgetary matters were a repeated process, one
might assume rulers could have developed a reputation for producing
low corruption regimes and avoided either the sanctions of exit or the
unpleasant oversight of voice. Although this approach is theoretically
attractive it holds little promise to explain the development of
taxation in Europe. To be sure, reputation enlarges the set of
equilibrium taxes and expenditure shares that can be supported and
thus apparently weakens the differences across political regimes. This
is particularly true if subjects use trigger strategies to punish
misbehavior and rulers never deviate from the equilibrium path. Then
rulers and subjects who are patient enough can find equilibria that
support efficient (low corruption, high tax, high public goods)
outcomes regardless of the institutional structure. It is our
contention however that history is not consistent with the prevalence
of reputational equilibria. Institutions mattered in European public
finance because rulers could not adhere to behavior consistent with a
good reputation. While some may have behaved well (as Henry IV
appeared to do in France), inevitably either warfare or corruption
reared its ugly head and society returned to a more naked political
economy in which institutional constraints were binding. As we shall
see below, the history of China after 1850 supports the notion that
public finance based on reputation-exit institutions cannot be
sustained in the presence of repeated warfare.
Hence the European equilibrium may well have been less a consequence
of initial constitutional conditions than a consequence of political
fragmentation. Fragmented sovereignty was a necessary condition for
warfare, it also was important in coordinating resistance to tax
increases and paving the way for voice. It also had the consequence
that taxing resources at the boundaries of sovereignty was a much more
important source of revenue for European states than for China. Both
in absolute and proportional terms, more revenue was generated from
commerce by European states than in China. Tariffs were both domestic
and international. Indeed the European polities that existed in 1700
were the product of conquest, marriage and inheritance and as noted
above each province tended to maintain institutional autonomy long
after it had come under the sway of a particular king. Although such
autonomy was important for creating voice, it also made possible
internal tariffs, a fiscal benefit for the crown. During the
eighteenth century the elimination of such internal tariffs became an
important policy goal in France and Spain. Such reform was part of a
package of reforms that were intended to increase economic output but
were resisted by local elites because they threatened the fiscal
equilibrium. In France these internal barriers were swiftly removed
during the Revolution as representative institutions arrived. In sharp
contrast, the unification of Britain with Ireland in1801 came long
after the rise of representative government. There was thus no simple
causal linkage between institutions of political representation and a
state’s fiscal centralization and integration. More generally and key
to our argument the competitive state system that created warfare also
caused serious distortions to trade within Europe.
China’s fiscal regime was well suited for promoting the material
welfare of the general population under a pre-industrial set of
economic possibilities. The direct consequences of low taxes and high
public goods expenditures were economically positive. China’s fiscal
regime was well able to absorb a variety of natural and social shocks
that could have challenged the material security and economic
well-being of the people. The equilibrium between principles of
taxation and expenditure broadly stayed in balance. In Europe, any
attempt to exit the high taxation/low public goods equilibrium was
inevitably undermined by needs for raising taxes to fund warfare.
Over several centuries preceding 1850, Chinese rulers were usually
able to address a variety of economic opportunities and challenges.
Moderate taxation was a key component of a larger kind of balance that
the state aimed to maintain across the empire’s diverse regions and
along the social hierarchy. Maintaining the fiscal equilibrium and
social and economic balances more generally became increasingly
difficult over the eighteenth century as populations grew and the
territory under imperial rule expanded. Indeed, in so far as state
policies enhanced people’s abilities to reach the economy’s production
frontier, the state may have made some people more vulnerable to
short-term economic shocks from natural or man-made disasters.
In Europe the fiscal equilibrium of high taxes and high military
expenditures produced fewer direct economic benefits for ordinary
people. But despite the persistent underinvestment in public goods,
the European fiscal regime produced some unintended consequences that
were positive, while China’s did not. Governments as a source of
demand for military armaments, for instance, could stimulate
technological changes that could be beneficial beyond warfare.
Government spending stimulated investment and technical changes that
might otherwise not have taken place. These consequences for the
economy were indirect because governments were not seeking to improve
conditions in the economy generally. For instance, investments in
warships were designed to make them faster and more reliable. The
knowledge gained from these investments did diffuse and made all ships
faster. As a results goods and people, rather than just cannons and
marines moved more quickly over time. These suggestions have, of
course, been made long ago (Nef 1950). For our purposes it is
important to contrast these kinds of unexpected long-term consequences
from the more immediate goals pursued by government officials, which
in early modern Europe meant warfare and in late imperial China meant
social welfare. This point complements the one made in chapter 3 about
the unintended consequences of locating manufacturing in cities. For
both observations, we distinguish causal consequences from the
intentions or purposes of the people making the political decisions
that affected long-run economic growth possibilities.
The two cases of early modern Europe and late imperial China by
themselves might suggest that persistent political instability is more
advantageous than peace for promoting long-run economic growth. It is
helpful therefore to recall that much of the world, including Africa,
South and Central Europe, and Southeast Asia all had competing
political regimes and often warfare, without benefiting from the kinds
of windfalls that took place in Europe in the period preceding the
Industrial Revolution. Further research is needed to understand just
why those competitive areas did not embark on the path of capital
using technical change. Clearly though having political competition
and conflict is not enough to guarantee technological change and
economic growth.
China and Europe after 1850
In some ways the contrasting relationships between state finances,
warfare and economic growth in China and Europe after 1850 reversed
the basic patterns clearly visible a century before. China moved from
a low tax and high public goods equilibrium to a high tax regime in
order to finance military expenses—first to put down massive
mid-century rebellions and second to begin building armaments to
strengthen the empire’s defense against foreign predators. As a
consequence, the Chinese state’s abilities and willingness to invest
in public goods fell well below the levels it had maintained a century
earlier. Just as European states had earlier turned to commercial
sources of revenues, after 1850 the Chinese empire also expanded its
revenue base through new domestic and foreign trade taxes, and by
increasing the price at which the government’s monopoly sold salt.
Europe, however, ceased to engage in the scale of costly warfare that
had previously been the driving force behind fiscal expansion.
Instead, European states were beginning to supply more public goods.
This included financing railroads beginning in the 1840s and 1850s on
the continent and improving urban amenities later in the century with
sewage, plumbing, paved streets, street lamps and the like. Alongside
the growth of public goods provision, Europeans completed their moves
toward freer trade that had begun in the eighteenth century. These
took place both within individual European countries and between them,
1847 saw the unilateral abolition by the British of the Corn Laws,
while the British and French signed a major trade treaty in 1860.
These changes in the fiscal regimes of both China and Europe took
place after the Industrial Revolution. Neither played a positive
causal role in creating the Industrial Revolution. We might, however,
have expected the Chinese state in the second half of the nineteenth
century to play a positive role in mobilizing resources and directing
them into projects promoting economic growth. The state certainly was
able to raise much more revenue than it had been able to raise in the
eighteenth century. But the extremely large increases in taxes that
occurred after 1895 were forced upon the Chinese by foreign powers
wanting indemnities for military actions. Early modern European state
expenditures on war could not be expected to yield direct and positive
impacts on those economies. Similarly China’s payments of indemnities
in the late nineteenth and early twentieth centuries could do little,
if anything, to stimulate economic growth.
In 1849 the government raised some 42.5 million taels of revenue with
77 percent of this coming from agriculture and the balance from
commerce. Thirty-six years later revenues had climbed to more than 77
million taels, the increase largely due to a quadrupling of commercial
revenues. Expenditure levels had remained in the range of 30-40
million taels annually between the 1720s and the early 1840s. They
then doubled to 70-80 million taels annually between the 1860s and
early 1890s (Hamashita 1989: 66) The capacity to increase revenues and
expenditures in this manner is hardly the sign of a weak state—the
conventional portrait of China in this period. But it is an indicator
of a significant transformation. The state was able to mobilize far
more revenues than they had previously found desirable to amass. While
the state took in less than what was needed, its revenue was
sufficient to begin responding to foreign challenges.
Much of the increased revenue was raised through the Maritime Customs.
In addition to serving as security on foreign loans (which were used
to help pay for the 1867 Muslim rebellion suppression in northwest
China), customs revenues were used in the 1880s to build railroads
(Hamashita 1989: 68, 72). The development of imperial control over
customs revenues is a clear indication of the state's ability to
create new infrastructural capacities. When China’s late
nineteenth-century central government is not judged by its failure to
survive beyond 1911 but is instead compared with its
eighteenth-century predecessor, we can see just how much its fiscal
capacities had grown. But these Chinese increases were nothing
compared to the nearly 302 million taels of revenue gathered in 1911,
the final year of the dynasty. By this date agricultural taxes had
grown from roughly 30 to roughly 50 million taels, commercial taxes
brought in more than 207 million, and another 45 million came from
miscellaneous sources. Whatever the late Qing state's weaknesses,
raising money was not among them (Wei Guangqi 1986:227).
Unfortunately, the Japanese indemnity equaled a full year's receipts
and the Boxer indemnity was one and a half times as large. It was the
weight of international reparations that made China's fiscal situation
so precarious and ultimately untenable.
What if, however, China had been free of its international debts and
thus able to put newly raised funds to more productive uses? Are there
any indications that the government would have used the funds
effectively? Typically the Chinese failure to industrialize in the
late nineteenth century is contrasted with Japanese successes. Yet,
these contrasts between China and Japan may rest too much on a reading
backward from mid-twentieth century differences in economic growth to
nineteenth-century differences. An assessment of either what the
Chinese state did or might have done were it to have had more revenues
with which to pursue an economic agenda would clearly provide a more
positive conclusion. Benjamin Elman’s assessment of Chinese science
and technology includes analyses of some late nineteenth century
changes. He argues that the Chinese tradition of natural studies and
Western science developed together beginning in the 1860s and that
both highly educated literati, the social stratum from which officials
were recruited, and more modestly educated artisans, were drawn to
modern science and technologies entering the empire from the West.
Furthermore, these developments began a decade before similar changes
would take place in Japan. For their part, Japanese visited Chinese
arsenals and shipyards to learn how to develop imported technologies.
(Elman 2005: 283-395)
At the turn of the century, a time during which Tessa Morris-Suzuki
(1994) has found the Japanese were spreading technological knowledge
through various local study circles, we also find governmental efforts
in China spearheaded by provincial governments to promote the
formation of similar groups at the county level (Morris-Suzuki 1994,
Jiangsu sheng 1919). When we look directly at nineteenth-century
evidence it is less obvious than we once assumed that Japanese efforts
to promote economic growth were more serious than those taken up in
China. Had the Chinese had more fiscal and organizational resources to
put toward these efforts after 1900, they might have helped promote
industrialization and economic growth. But the fiscal difficulties of
the indemnities for the Sino-Japanese War and the Boxer Uprising
crippled the state’s capacities to spend money for projects intended
to yield economic benefits. Though the late nineteenth-century Chinese
state was unable to devote as much of its attention to economic
development as the Japanese did, officials were able to raise
substantially more revenues than they had previously done. If we
imagine China less threatened by foreigners, we could reasonably
expect a greater portion of Chinese efforts would have gone toward
promoting economic change. Had there been even less contact we can
imagine China possibly continuing for a longer period of time with its
earlier kind of fiscal equilibrium, and thus not likely to have had
either the opportunities or the pressures to make dramatic changes.
In Europe, the nineteenth century was also a time of rising taxes,
fiscal innovation and, more generally, of increasing government
involvement in promoting economic growth. Without going back to the
catch up framework of Gerschenkron (1962), states did become more
concerned about promoting growth. As a result and in contrast to
China, industrialization in Europe led to a decline in the share of
government revenue (and of the economy) devoted to the military.
Hence, the political transformation begun with the French Revolution
propelled Europeans in a direction opposite from China’s movements.
While the Chinese empire had to spend ever more on military activities
(or on war indemnities), continental European state were investing in
their economies. England is a somewhat anomalous case because the
maintenance of its empire was a very important factor in the
relatively high rates of taxation that fell on Britons. Yet because
per capita income was relatively high, England could afford both
public goods and the largest navy in the world (Davis and Huttenback
1986).
Leaving the consequences of colonial involvement aside, European
states increased their expenditures on public goods in different ways
and at different times in different places. The recent work of Peter
Lindert (2004) has focused on the growth of welfare and education
spending. It shows that representative institutions (voice) were
critical in the expansion of public services from the 1880s to the mid
twentieth century. Yet the growth of public spending antedates the
rise of the welfare state and the expansion of schooling. In the
nineteenth century, public spending included transport through the
direct financing of roads and canals, subsidies to railroad expansion,
or concessions for transport improvements (ports, bridges, and at
times roads). Such spending (admittedly by local rather than central
governments) also provided significant investment in local public
utilities, as urban areas expanded the demand for sanitation, clean
water, street lighting, market places, and local transport services.
The pursuit of increased public goods depended on the mechanism of
voice. Most obviously, countries with representative government were
more likely to devote a share of their central government budget to
these activities—that is, after all, how the expansion of railroads,
canals and other major infrastructure projects were funded. Equally
important was that the process of creating structures that favored the
local provision of infrastructure or education also depended on
mechanisms of voices. These included the bills of the British
parliament that authorized turnpike trusts and railroad corporations
(Bogart 200X). It also included the mechanism whereby municipalities
could concede the business of providing lighting, clean water and
other local public goods to the public sector or decide to provide it
themselves. Finally, mechanisms of voice were also present in the
myriad of not-for-profit organizations that bloomed in nineteenth
century Europe. Whether these were credit cooperatives, savings banks,
or agricultural improvement districts they all relied on the capacity
of individuals to form organizations whose governance gave their
members voice.
The end of the Old Regime in Europe marks a transition to a higher
level of public goods provision. Nevertheless, not all was new in the
nineteenth century. Military concerns continued to loom large. Thus
states did not promote trade or railroads solely because of their
economic benefits but even more importantly because of their
implication for the balance of power in Europe. For similar reasons,
central governments were reluctant to let sub-national units (in
particular municipalities) run their own fiscal affairs. The main
motivation for strict restriction on cities’ fiscal independence was
to avoid competition over revenues between central and local
governments. Because they were fiscally constrained, local governments
had little choice but to turn to the private sector for the
development over local utilities. War continued to limit the provision
of services by the public sector in Europe. Public goods provision did
indeed depend on the mechanism of voice but, as the Chinese case makes
clear, the European political logic is by no means a universal
one—Chinese had more developed public goods provision at an earlier
point in time because of attention to issues of exit rather than those
of voice. (Wong 2007)
Conclusion
This chapter has examined the differences in public finance using a
simple model of political economy. In comparing this model with
history we have both been required to make it more complicated and
more subtle and we have also come to some better understanding of the
key differences between the pre-industrial public finance regimes that
prevailed in China and Europe.
Perhaps the most controversial implication of this chapter is that
Chinese political economy, in particular its fiscal regime, had more
positive direct consequences for economic growth than did European
fiscal regimes before the Industrial Revolution. This was because the
emperor was far from being a predatory despot. Rather, the Chinese
regime focused on a significant production of public goods with
moderate taxation—in particular under the Qing. This regime produced a
successful expansion of the agrarian and commercial economy—precisely
the kind of Smithian growth that European economic historians are so
fond of. But Smithian growth need not beget industrialization (Wong
1997).
In Europe by contrast, the political pressure to raise taxes was a
central element of conflict between rulers and elites. While in the
short run perspective of an agrarian commercial economy, these
pressures had little good to offer, in the long run they may well have
been an irritant that increased the likelihood of both political and
economic change. The military after all was an important source of
demand for manufactured goods. Yet one should bear in mind that
competitive state systems, like those in Africa or Southeast Asia,
have more often than not produced warfare rather than economic growth.
It is also apparent that the direct benefits of political systems with
representation only came to be realized after the onset of the
Industrial Revolution, and after the consolidation of fiscal power
into a unified central government. In the early modern period, all but
the smallest states labored under inefficient and fragmented fiscal
regimes. In these regimes, while taxes were high relative to China,
public goods provision mattered little. China reminds us that although
representation played an important role in nineteenth and twentieth
century in Europe, one cannot conclude that representation drives
public goods provision everywhere and at all times.
By the mid-nineteenth century, China was laboring under increased
external and internal pressures and its failure to develop large-scale
fiscal capacities was becoming increasingly costly. Adaptation to a
world in which warfare was a major problem proved difficult. The
history of China since 1949, however, suggests that the path to higher
taxation and public goods provision need not go through representative
institutions. The Communist party both in its Maoist and reform guises
has been quite successful at providing public goods—be they primary
schooling, health systems or infrastructure. A case in point is the
generation of school children whose education was ruined or at least
severely impaired by the Cultural Revolution. For such a generation of
students to have their education destroyed there had to be schools in
China in the 1950s which enrolled very large numbers of pupils. Had
Mao behaved like the proper despot of economic theory after 1949,
there would have been no need for a Cultural Revolution because few if
any children would have been going to school.
Since 1978, the Chinese government has embarked on series of reforms
that have privatized much of the economy. To be sure serious problems
remain in factor markets, most notably capital markets. Yet an
important part of the gains from growth have been invested in
infrastructure and other public goods. Although the public sector has
shrunk dramatically, investment in more classic public services has
been increasing (though probably not at a rate commensurate with
demand). This is despite the fact that the Communist Party has not
released its grip on the political process. Thus, while China may open
up the political process over the next decade or two, it is not going
to be because they had previously failed to provide public goods.
In our analysis we emphasize variations among the fiscal institutions
of European polities and only hint at the variation in fiscal
institutions across the Chinese empire. We have taken the spatial
scale of polities as given and contrasted empire in China with a
competitive political system in Europe. In several previous chapters
we have noted that very few competitive political systems have led to
the peculiar equilibrium favorable to economic change as experienced
by Europe. A similar caveat is in order for empires, China’s
bureaucracy focused on public goods and agrarian prosperity was
unique. Of the three other empires the Mongols ruled (India, Persia
and Central Asia) none had administrative structures like China’s. The
Mongols were either unwilling or unable to export the political
structure that they imbedded themselves into in China. Hence the
Chinese imperial equilibrium is but one possible outcome for a
territorially large entity. We do not claim that Europe is typical of
competitive state situations or that China is a typical empire.
Rather, they represent two especially successful examples of a
competitive state system and an empire respectively. It is now time to
show how the divergent political equilibria at either ends of Eurasia
came to be sustained.
Box 6.1: taxation and public goods
This box reproduces the McGuire and Olson results in a simple setting.
Let Y(G) be total income, given investment in public goods of G.
Y(0)=0. Let t be the tax rate and the reduction to R(t) . There is a
faction that controls the net proceeds from taxation. That faction
account for F of the economy, (F=0 corresponds to a dictatorship).
The faction must decide the tax rate and public goods

First assume the constraint does not bind.





Thus starting with dictatorship (F=0) and increasing F, t declines and
G increases. There exists an  <1 such that the constraint
binds (  ). In other words the controlling faction finds it
inefficient to redistribute income to itself,  . Past that
point, political structure no longer matters.
Box 6.2: Adding War.
Let war be a gamble W(w) such that W(0)=0. W() is increasing and
concave. If F is sufficiently small, then the faction in power set
taxes exactly as if there was peace: It funds war and public goods out
of the fraction of revenue that it would have confiscated.







Again there exists  such that the constraint binds (
). In other words the controlling faction finds it inefficient to
grab tax revenue for itself when it includes more than   of
the population. As in the peaceful economy, taxes are highest,
spending on public goods is lowest in the dictatorship (F-0). War
expenditures rise with F.
Public goods spending depends on the returns to war, thus for a given
F, they may be higher (if war is very profitable W(w)>1, or lower if
war is very costly (W(w)<1).
If wars are an even bargain (at  W(w*)=1); then public good’s
spending is the same as in the peaceful economy for all political
regimes where F< . For regimes where F >  . It also
follows that for all regimes where  taxes are higher, and
public goods spending less than in the peaceful economy.
Box 6.3: adding Exit and Voice
The key in establishing the results we saw in box 6.1 and 6.2 is that
taxes damage the formal economy at the same rate independent of
political regime or what they are used for. In such a context, where
the dictator is a tax maximizer, he will always choose the highest tax
rates of all. Yet we know that the extent of evasion depends on
whether people feel about taxes are raised for legitimate or corrupt
purposes. Judgments about the legitimacy of taxation concern both the
degree of representation (F) and the level of diversion that the
controlling faction carries out.
These considerations mean that the share of the economy that is lost
to tax avoidance (1-R(t)) depends on F. In other words R(t, F) such
that  
The faction in power continues to maximize the same objective
function.

Claim If the willingness to pay taxes increases sufficiently rapidly
with F ( ). Then  .
Clearly as in the earlier cases there must exist   such that
 . For any F>  The constraint will continue to bind
but because the taxation is less costly, taxes will be increased to
produce more public goods or spending on the military.
For F below   because the constraint on revenue does not
bind, spending on the military and public goods will expand as they
did in the earlier cases
The key therefore is the comparative static of changes in F on changes
in t*.
 =0’
 =0

The first term is negative, between 1 and 0 and involves the standard
tax avoidance response of the population. The latter two terms are
positive and relate to the greater willingness of individuals to pay
taxes as the franchise expands. A necessary condition for taxes to
rise with the franchise is thus  .
.
Chapter 7
Political Economies of Growth, 1500-1950: Chinese Empire and European
States
We have seen in previous chapters that economic growth due to gains
from trade was more easily achieved in China than in Europe during the
early modern era. Despite differences among the kinds of economic
institutions most typical of China and those most typical of Europe,
we can find no evidence that such differences made for significantly
different likelihoods of economic growth taking place in one rather
than the other. Nor do differences in the representative nature of
political institutions play the often anticipated role of serving
economic growth. Yet it is easy to be suspicious that these claims
must somehow be specious for surely the economic and political
practices preceding the Industrial Revolution must have influenced the
manifest divergence between the economic trajectories of
nineteenth-century China and Europe. We do not, however, claim that
different practices preceding the Industrial Revolution had no
significance for nineteenth-century patterns of economic growth.
Rather, we suggest that some of the most important differences between
China and Europe that mattered to nineteenth century economic growth
emerged several centuries before that time. Unlike many previous
observers we do not find that the differences were due to either a
particular cultural genius of Europeans or to political and economic
circumstances that endowed them with advantages from a very early
point in time. In this chapter we argue that early modern Chinese
political economy was more explicitly intended to foster economic
growth than European political economies. Moreover, Chinese officials
succeeded in part because they had created political peace and social
stability for more people across far more territory than their
European counterparts could realistically imagine, let alone pursue.
At this point, the nineteenth-century economic divergence is not
merely a European success story and a story of Chinese failure to
emulate those successes. It is also a story of the Chinese’s loss of
an earlier era of political economy, in part due to the political
challenges created by Westerners and the Japanese. This history may
seem very distant from a twenty-first century that has witnessed an
apparently relentless expansion of the Chinese economy, but the
abilities of the Chinese state to foster conditions making this growth
possible are in part explained by economic history.
Late Empire: Foreigners, Natives, and Chinese Strategies of Rule
While European princes, as well as rulers in the Islamic political
world, were advised about how to undo their princely rivals and
suppress internal challengers, many Chinese officials were reading a
text that was very different in substance and spirit from Machiaveli’s
The Prince. They studied the Supplement to the Exposition on the Great
Learning, by fifteenth-century Confucian scholar Qiu Jun, a work that
combined descriptions of statecraft policies popular in earlier
centuries with the author’s own commentaries. Widely distributed after
Qiu Jun presented it to the emperor who ordered the text to be printed
and disseminated across the empire, this work became a ready reference
for officials considering a variety of statecraft subjects, including
water control, grain storage, taxation, and local administration of
minority populations among many others. Indeed it continued to be
influential after the Manchus established their Qing dynasty in 1644
as the men who became emperors were taught from the supplement by
their Chinese tutors. They learned what the practice of benevolent
rule across an agrarian empire could concretely mean. From the vantage
point of the empire’s sedentary population alive in the late
seventeenth and eighteenth centuries, people who accounted for well
over ninety percent of the empire’s total, the Manchu emperors
advocated and implemented an agenda for managing society that was far
more energetic and ambitious than their Ming predecessors’.
The emperor’s commitment to Neo-Confucian strategies of rule was by
itself inadequate to create the conditions for Ming and Qing dynasty
successes at ruling the agrarian empire. To make a difference elites
and commoners alike also had to consider Neo-Confucian priorities and
policies expressed in works like the supplement to be sensible and
beneficial. At a minimum they had to believe that their interests were
better served within this political order than by undertaking the
costs of exiting the empire. We do not mean to suggest that people
were constantly evaluating the relative benefits and costs of staying
within the empire in late imperial times, but simply that if people
had been actively dissatisfied they would have sought to reformulate
their relationship to the state through some combination of voice and
exit; instead they remained loyal for the most part. Why? Because most
subjects had little incentive to bear the costs of inventing an
alternative way of organizing political order outside of empire when
enjoyed considerable social space within which the state weighed
lightly on them and they could enjoy its material benefits.
Given that both elites and commoners accepted imperial forms of rule,
how did Neo-Confucian strategies of social order deflect and defuse
the challenges that strained and often fragmented other empires?
First, the core of the social elite was composed of literati educated
to seek official positions from which they gained their social status.
Second, unlike either the early empire of the Han or the middle empire
of the Tang, the late empire of the Ming and Qing dynasties did not
have to confront great magnate families. Third, commercial elites were
not pressed so hard for resources that they considered mounting major
opposition to the center. Landed and commercial elites were instead
effectively delegated the tasks of maintaining social order by the
bureaucracy, and as long as no serious troubles emerged, they were
largely left alone by the state. Elite interests were effectively
served by a partnership with officials.
For their part, merchant elites specifically benefited from state
policies that facilitated long-distance trade and their riches (unlike
the wealth of Italian and German merchants which was chronically
vulnerable to predations from princes anxious for resources), could
usually be protected from extraordinary state exactions. The state
could keep its direct costs of governing the empire relatively low
because it depended on local elites to shoulder much of the burden of
formulating and maintaining institutions of local order such as
granaries and schools, as well as insuring the upkeep of roads,
bridges and temples. Social order was the joint product of official
and elite efforts. When natural disasters or social problems emerged,
officials, elites and common people often expected that joint efforts
would solve the crises and when they didn’t think this likely they did
not imagine that some exit strategy from empire would improve their
conditions. The late Ming dynasty survived the kinds of domestic
threats from regional powerholders that undermined the effectiveness
of other empires. When they lost control, their ideology and
institutions of rule for society were largely adopted by the Manchus
who succeeded them.
The Manchu-led Qing dynasty that came to power in 1644 expanded the
empire’s borders once again into Central Asia. Unlike the Mongols, the
Manchus largely adopted the bureaucratic institutions of civilian rule
for administering the vast peasant population of the empire. They made
changes designed to improve communications, bureaucratic effectiveness
and, in particular, responsiveness to imperial orders, but the basic
institutional template and ideological justifications of rule followed
the principles and policies of earlier rulers of empires.11 As we look
at the role of the Manchus from the vantage point of the role of
outsiders either promoting the persistence or hastening the
destruction of empire, we can appreciate the degree to which Manchu
successes across peasant China depended upon their integration into an
ongoing bureaucratic structure of rule. The eighteenth-century
imperial anxieties about Manchus losing their martial spirit and
becoming assimilated into Han Chinese culture reflects the
considerable assimilation they underwent. The differences between the
Manchus and Han Chinese, important as specialists have shown them to
be, remain less stark than those between Mongols and Chinese a few
centuries earlier. The political similarities and connections between
Manchus and Han are even more apparent, and for us crucial, when we
put this Manchu-Han relationship into a common frame of reference with
the relationship between imperial Rome and its “barbarians.” In
contrast to the Western situation where large numbers of distinct
groups invaded portions of the Roman empire with none able to either
ally with or defeat the others, outsiders in Chinese history were
smaller in number relative to those already living under imperial
rule. By the time the Manchus appear on the scene, a demographically
small group from beyond the empire had available a repertoire of
policies that created benefits for both the rulers of empire and their
subjects.
The political economy of the eighteenth-century state generally
followed on principles articulated in the previous centuries, but
committed officials to a greater degree of intervention and activism
for longer periods of time than was typical under the Ming dynasty.
During the eighteenth century domestic commercial taxes were
deliberately kept minimal. Merchants largely regulated local markets
on their own. For its part, the state depended on markets, not only to
purchase the commodities consumed by the imperial household and the
bureaucracy, but also to purchase the construction materials and hire
the labor needed to build and repair government buildings. More
significantly for the population, the state also bought grain in times
of dearth to transport to places suffering the greatest subsistence
needs. These features of the state’s political economy contributed to
the expansion of long-distance trade and the importance of informal
institutions addressed in Chapter 3. The state also encouraged the
diffusion of handicraft production (that we evaluated in Chapter 4)
throughout China. Some officials disseminated information about craft
technologies as they moved from post to post across the empire, and
the state more generally chose not to tax the craft output of agrarian
households, limiting itself to taxing the household’s agricultural
output. Indeed, it would have been far more costly to tax widely
dispersed rural craft production than to tax urban-based production in
larger workshops, which makes the state’s decision to forego these
taxes more understandable than if the crafts had been concentrated in
fewer locations.
The state’s role in private and public finance also promoted economic
growth. The private credit market we examined in Chapter 5 was largely
informal with the state playing only a small role in regulating its
activities. Chinese business was able to develop informal mechanisms
to finance production and distribution without much recourse to
government intervention. The costs of doing business were therefore
lower than they would have been had more formal institutions been
established. In public finance, as we showed in Chapter 6, the
eighteenth-century Qing state invested far more in infrastructure
(e.g. water control for production and transport) than was possible in
Europe in the same time period. The state’s social spending overall
was higher and it stimulated and guided local government and elites to
fund granaries, schools, road and temple repairs, and social
surveillance against crime in their areas.
The contrasting spatial scales of the Chinese empire and European
states offers a splendid illustration that the trade-offs offered by
the theory of the firm are relevant to understanding the importance of
scale in political economy. The firm size (in terms of total
capital/employment or in terms of the number of tasks that it takes
on) is variable as technology changes, and a manager who wishes to
expand his firm must develop techniques of administration that make
internal management superior to that of the market. The Ming chose a
smaller empire but one in which the population was overwhelmingly
sedentary and thus receptive to the value of peace and internal trade.
They did so not because they could not muster the might necessary to
recover part of the western lands held by precursor dynasties. To the
contrary they limited their spatial ambitions to focus their resources
on internal growth. The Qing dynasty, building on that effort, was
able to simultaneously expand the set of services they rendered to
their peasant population and bring the empire to its largest scale.
Their successful strategy of rule provided the resources for
expansion. How different was the experience of state formation in
Europe?
State Formation in Europe from Charles V to Napoleon: European Lessons
Machiavelli’s prince was a guide for rulers who faced enemies from all
sides. As Machiavelli saw it, the ambitious prince wanted to enhance
the size of his real at the same time as he wanted to avoid being
beholden to his subjects. Pursuing his ambitions implied defeating his
external enemies while holding the rebellious tendencies of their
subjects in check. That The Prince was the main secular guide to
rulers’ behavior reminds us of the long history of European political
strife. European states could not be built from a core relationship
between subjects and rulers focused on low taxes and public goods.
Instead they had to be built in struggles that included conflicts
between dynasties as well as violent confrontations between subjects
and rulers.
Fragmentation in Europe ended haltingly but by 1300 the trend toward
ever smaller polities that began with the collapse of the Roman Empire
had reversed and states were generally growing. One reason for this is
that, by then, the external challenges to Europe were limited
geographically to the formidable threats represented by the Ottomans.
From Spain to Poland, Europe was expanding based on a military
technology that was radically different from that of the Roman legions
also unlike what had prevailed at the end of the first millennium. On
the defensive side the importance of fortifications made it possible
for relatively small states (like the Low Countries) to hold off
larger ones. Yet, fortifications required resources, and expenses did
not stop there. By 1300 feudal levies of troops had long been replaced
by soldiers who had to be paid (be they foreign mercenaries or
domestically levied troops like the Spanish Tercios). By 1400
artillery trains added to the cost of war. Only states with large
treasuries could continue to compete in Europe’s political contests.
Such large treasuries were possessed either by small but very
prosperous polities, like Venice or Florence, or by very large ones,
like France or Castile. Although many independent polities
disappeared, there were serious obstacles to the expansion of states,
most notably the general tendency of alliances to form against the
major power of the time. Throughout the centuries between Crécy and
Waterloo, international conflict was perhaps only somewhat less
pervasive than in the preceding millennium. The persistence of war had
several consequences. One of these was the development of a military
infrastructure that, by the end of the sixteenth century, had enabled
Europe to extend its political authority across the globe. The other
is that the demands of warfare in Europe would make the development of
Chinese like-strategies of rule simply impossible prior to 1815.
In 1516 Charles Hapsburg ascended to the throne of Spain. This is none
other than the Charles V we have already met in Chapter One. Charles
was the focal point of an extraordinary dynastic convergence. Through
each of his grandparents he inherited a formerly sovereign entity.
Along with the crown of Spain on his head, Charles also was ruler of
major parts of the Italian peninsula, Austria and its dominions, and
the Low Countries. Not content, he had himself elected emperor of the
Holy Roman Empire as well. Soon his domains included most of Latin
America after the conquests led by Cortez and Pizarro. Consequently by
the time of his abdication in 1556 he was the ruler of an empire of
nearly Chinese proportions and one that, even though it did not
include France, vastly exceeded dominions of Charlemagne or Napoleon.
Yet had been true for Trajan before him, Charles V’s capacity to
acquire territory exceeded his capacity to rule it. When he abdicated
he split the empire, carving out the Imperial crown and the Austrian
dominions for his younger brother and leaving the rest to his son
Philip II.
Charles V’s European empire was definitely un-Chinese, and for that
matter, un-Roman.12 Obviously it was far from compact, and it was not
the result of some persistent expansion based one group’s military
prowess over another. Furthermore, while Charles’ legitimacy as the
ruler of these lands was unquestioned, the extent of his authority in
any one of his domains was a complicated matter. Charles was hemmed in
by the liberties his forebears had granted to the different regions
they had acquired and some of these were quite extensive. More
important still, the administration of each region was sui generis and
changing any of the key institutions in a locality required the assent
of a local representative body, or the presence of the monarch, or
both. Because so many of his territories were quite small, Charles V
had much greater difficulty ruling his domains than his Chinese
counterparts did. Subjects in Castile fell under a relatively uniform
set of institutions but they inhabited only about two-thirds of the
polity we now call Spain. In the Netherlands there were nearly a dozen
separate provinces or territories where Charles V’s authority varied.
Moreover, while Castilians and Catalonians may have recognized more
connections amongst themselves than they did with the king’s subjects
in Naples or Vienna, they were far more like to emphasize their
differences, and to take political action to maintain these
differences. Hence Charles V’s efforts at creating coherence in his
European domains failed; his son Philip II would continue the effort
only to spark the Dutch revolt.
At the cost of a digression, it is worth noting that European rulers’
inability to gain more riches and territory at each others’ expense
propelled some of them to go overseas. In the Americas, large expanses
of land were claimed for European crowns and for a time massive wealth
for the Spanish crown. Trade with Asia, was organized around
monopolies that were supposed to make regular contributions to the
state’s coffers as merchants maneuvered to gain positions at new and
old ports from which they could purchase precious spices and luxury
goods. In both instances, European states built what historians have
labeled as “empire.” These territorial and commercial expansions do
not meet our criteria for empire in terms of population and territory.
Indeed, these empires were either purely commercial or heavily
extractive—in no case was there any effort to fold newly gained
territory into a larger, homogeneous whole. This fragmented strategy
persisted into the colonial rush of the nineteenth century. Thus,
there is a fundamental institutional contrast between Chinese and
European empires.
Nowhere is this more in evidence than in the deep concern of Charles
V’s subjects for their local privileges meant that, in Europe, larger
polities did not realize many of the gains that one might expect.
Indeed these larger territories were themselves institutionally
fragmented and constantly at war. From a Chinese point of view
Charles’ European dominions were small and not very well integrated.
To promote gains from trade, a peaceful empire was more advantageous
than smaller war making states could be. In this matter, the
difference between the Hapsburg empires and Britain was in fact less
than that between these empires and China. Neither Charles nor his
successor could sustain a program of institutional harmonization
because of the demands of war. Europe remained a competitive political
system. As we saw in preceding chapters, the economic advantages to be
realized from competing states came late and were unintended. The
dominant impression of political change from 1300 to 1700 must be one
of states fighting. To a lesser extent, polity size was growing. Yet
the rise of more uniform institutions was to come later—after the
onset of the economic transformation of the Industrial Revolution.
Europe’s Industrialization and Imperialism: State transformations and
economic growth
Social scientists often associate the conditions conducive for
economic growth with those that enable democratic political regimes.
Individuals who enjoy liberties and freedoms typically also benefit
from secure property rights. Those places in Europe developing
economically were also those formulating democratic political
institutions. The rise of representative government represents a
remarkable break in its political history not only for Europe but for
the world. Nevertheless, rather than the Glorious Revolution, it was
the French Revolution that was a watershed for European political
structures. In the eighteenth century no European country followed
Britain’s lead of parliamentary monarchy, just as in the seventeenth
century no country had followed the Netherlands by establishing a
federal republic. To a large extent such regimes were anathema to
Europe’s rulers. In the quarter century following the French
Revolution, however, Europe experienced a massive political
transformation--the creation of unified parliamentary monarchies in
France and the Netherlands, a significant reduction in the number of
independent entities in Germany and Italy, the creation of unified
authoritarian monarchy in Prussia, and attempts at constitutional
monarchies in Spain and Portugal. Napoleon’s attempt to forge a large
political entity failed, but many of the changes he initiated endured.
Most strikingly, none of the restored ruling houses in France, the
Netherlands or in Italy gave up on fiscal centralization. Moreover
these changes would spread, for instance, when Belgium became
independent in 1825, it immediately adopted a form of representative
government.13
The transformation brought on by the French Revolution has typically
received less attention than the rise of democracy in the later
nineteenth century for several reasons. Most importantly, this
transformation was quite likely to create or bolster conservative or
authoritarian regimes (as it did in Prussia and the Netherlands). As a
result, while this transformation seems to have been a complement to
the surge in infrastructure investment that spread through Europe
after the demise of Napoleon, it did not lead to the appealing
equation of liberalization and growth. Yet from the point of our
comparison this is the period when European states begin to look
‘Chinese.’ Rulers in Europe demonstrated a new emphasis on efficient
governance and providing prosperity. That said, it is remarkable how
little effect this major political innovation had on fragmentation in
Europe. It is true that the number of independent states continued
fall between 1789 and 1815 because Napoleon and the Congress of Vienna
redrew the boundaries of Europe. Overall, however, the most radical
attempt at reducing fragmentation—Napoleon’s gambit to create a single
state out of territories in much of Western Europe—simply failed.
While many populations might have welcomed the reforms that French
conquest brought in its wake they did not want to be ruled by
Frenchmen. Local elites were sometimes divided about reform, but they
were always opposed to the elimination of their power and to foreign
overlords.
The French Revolution and the regimes created in its wake typically
downplayed regional identities in favor of national ones. But these
new identities were no more favorable to the creation of a common
political space than the older provincial ones. From a Chinese
perspective the partial replacement of Breton identities with French
identities, for example, was not much of a step toward creating a
European identity.
European history up to the mid-nineteenth century makes it abundantly
clear that the fragmentation of the Roman Empire had tremendous
consequence for this end of Eurasia. Long after the Great Invasion had
passed, long after Europe had become an exporter of military violence,
political processes remained mired in a local logic. There were
economic, political, and military reasons for states to grow, and to
some extent they did. Territorial growth, however, was painful and
slow. The path to a unified Europe having been closed by Napoleon’s
defeat, after 1815, the surviving states could enjoy the benefits of a
significant reduction of the power of sub-national institutions. They
also tried to reduce the economic costs of political fragmentation
through trade and monetary negotiations. They articulated a political
logic of balance of power meant to acknowledge competition among
themselves, while reigning in any unbounded pursuit of power at each
others’ expense. These efforts extended and elaborated upon the
political sensibilities formulated in the mid-seventeenth century
Treaty of Westphalia. Had Europeans been able to do more politically,
they could have achieved a larger economic space with lower
transaction costs and greater gains from trade. Clear they believed
that such union was not feasible. Thus, the approach to regional
economic institutions in Europe that bypasses the problem of political
union has its roots in the nineteenth century. Clearly, building a
large economic space from the bottom up is different than building it
from the top down.
Beyond their own fragmented region, Europeans grasped the desirability
of pursuing power and wealth internationally. Consequently, at the
same time as the domestic regimes of European countries developed new
political institutions and fashioned new political ideologies, some of
them embarked on new overseas adventures. During the second half of
the nineteenth century much of the world that had not already been
settled by white European became formal colonies of European powers.
The military expertise they had gained in a dozen centuries of warfare
allowed Europeans to exploit the labor and raw materials available in
many Asian and African areas. More generally, an industrializing
Europe fostered an international division of labor within which,
industrial capital concentrated in Western Europe and North America,
bought raw materials and attracted labor from other parts of the
world. The British promoted free trade as a virtuous and efficient way
to benefit people and their economies. The degree to which these
economic principles dominated international exchange remains a topic
of disagreement. To be sure free trade and an international division
of labor based on comparative advantage and natural resource
endowments prove to be powerful engines of growth. The Europeans
pursuit of these economic possibilities grew out of an early modern
era in which Europeans for the most part competed with each other
overseas rather than cooperated with each other within Europe.
Nineteenth-century British domination of the world had more than four
centuries of European maritime exploration and conquest as its
historical background. The ability of the British and other Europeans
to exploit economically their international political position
depended on technological and institutional changes more likely to
occur in Europe than in China (or anywhere else for that matter).
War-making drew entrepreneurs seeking to defend their capital into
cities where relative prices ended up favoring capital investments,
agglomeration favored growth, and certain technologies improved from
advances made in military pursuits. In previous chapters of this book
we noted what we consider to be the key reasons for economic growth in
China and Europe in the early modern era. We have argued that China
was not obviously, or certainly, failing to grow as Europe did. But
key differences in relative factor prices, which we explain through
the impact of political differences on economic decision making,
explain the far higher likelihood of an Industrial Revolution
occurring in parts of Europe than anywhere in China. Once this
economic transformation was underway it no longer makes sense for us
to simply compare the dynamics of economic growth in China and Europe.
We must also evaluate the significance of European impacts on China.
It is possible that after the Middle Kingdom was forced to enter into
the global economy, politics either prevented it from doing so on its
own terms or made it much more difficult such a transformation to
succeed.
After 1850 we no longer can analyze China and Europe as two large and
important regions independently. Indeed it is not obvious that the
dynamics of political and economic change in one region can be kept
separate from changes occurring in the other region. In the European
case, the influence of China was probably limited, though relations
with the Americas and the wider worlds were of considerable
importance. In the Chinese case, the major focus of politics and of
economics cannot simply be domestic. Even though China does not become
a formal colony of any foreign power in the nineteenth century, the
political and economic influences of foreign entrepreneurs and
officials was huge. The growing presence of Europeans in
nineteenth-century China was accompanied by increasing signs of a
weakening central government unable to meet the twin challenges of
maintaining the virtues of eighteenth-century statecraft and
fashioning a new kind of state power able to manage new kinds of
foreign relations. But while for a century after 1850 the Chinese
government failed to maintain itself, let alone provide order across
the country, China emerged after 1949 as a sovereign nation that
comprised almost all the territories previously ruled by the Manchus.
The Chinese accomplished this feat by asserting the primacy of a
unitary state in which authority was vested in the central government.
As a consequence, the Chinese were in a position to benefit from the
economic advantages of spatial scale once again and in ways that
Europeans only began to approach concretely after World War II.
Chinese Empire: the Limitations to Growth in a World of European
Dominance
For two millennia starting with the Qin, the Chinese economic growth
that we have examined in earlier chapters of this book was possible
because of the country’s imperial scale. During this time the state’s
political economy helped to support the institutional practices and
relative prices that favored an agrarian and rural economy. Yet such
strategies proved increasingly difficult to sustain as foreign
political pressures created new demands on the Chinese state. We do
not believe that their ultimate failure in 1911 can be attributed to
the limitations of the earlier dynamics of growth. Instead, the
economic advantages of empire were lost in the nineteenth century when
the demands of managing both domestic space and foreign relations
became increasingly expensive and difficult.
The Chinese political economy of promoting trade across a peaceful
empire and supplying social services and goods with relatively modest
taxation was no longer feasible in the nineteenth century. Chinese
leaders had to invest in new political institutions and economic
efforts designed to strengthen the state. They had to raise more taxes
and were able to supply their subjects with fewer public goods.
Beginning at roughly mid-century, the Chinese state began raising
taxes on commerce. By the 1870s and 1880s it appeared that the state
was coping adequately with it new political agenda. A crucial turning
point came in 1895 when the Qing emperor was defeated by Meiji Japan
in a naval war waged around the Korean peninsula. The victorious
Japanese imposed a punishing indemnity. To pay the Japanese the
Chinese government was forced to increase its taxes and face growing
domestic dissatisfaction. The Qing state confronted an even more
difficult challenge after an eight-nation army marched on the capital
in 1900 to demand the Qing state put down the violence of the Boxer
movement against foreign Christians. The ensuring indemnity equaled
roughly three times the total annual revenue of the government. The
only way to even attempt to meet the foreigners’ demands was to
reorient all government revenues to that one goal.
It is little wonder that China eighteenth century focus on prosperity
based in an agrarian rural economy lost pertinence. After 1911 and the
fall of the Qing dynasty, such a political economy became completely
irrelevant as the Chinese mainland was politically fragmented for most
of the years preceding the founding of the People’s Republic in 1949.
Even in the decade between 1927 and 1937 when the Nationalists claimed
to rule China, they could collect agricultural taxes from only five
provinces. Their rule over many areas depended on understandings with
military warlords, while they could claim in-name-only sovereignty
over other parts of the former Qing empire, such as Tibet. Taiwan,
which had been settled by Chinese immigrants centuries earlier and had
been incorporated administratively by the eighteenth-century state,
was no longer a part of China, but a formal colony of the Japanese.
More ominously, the Japanese established a puppet state in the
northeastern area of Manchuria, taking away from Chinese rule the
Manchu homeland to which millions of Han Chinese had migrated in the
nineteenth century. Together these changes meant that China faced a
most uncertain future in the 1930s.
Political competition and military conflict were chronic features of
Republican-era China. The kinds of conditions fostering economic
growth in the empire that we have examined in earlier chapters of this
book were largely lacking. Instead, political conditions in China
resembled more closely the war-making competition and chronic fiscal
shortfalls of early modern Europe. If European history had supplied
all the lessons, we could have expected the Chinese mainland to become
a set of states in competition with each other. Chinese political
development could be presented as a late copy of dynamics that had
worked themselves out in Europe centuries before. In this light the
Chinese phase of political competition would have been set off by
imperialism more than any other factor. However, in the nineteenth
century foreigners did not pursue strong colonial initiatives, thus we
might expect dynamics similar to those in early modern Europe to have
played a decisive role. From a European perspective China might have
remained politically fragmented since so many empires broke apart.
Yet, if generalize from European history or put the Chinese empire
into a common category with other landed empires, we simply fail to
explain what did in fact happen. Fragmentation did not endure and the
People’s Republic of China comprises almost all the population and
territory once ruled by the Qing dynasty.
Historians depict nineteenth-century Chinese history as a narrative of
decline qualified by some signs of adaptive abilities to develop new
institutions to accommodate the increased presence of Western
merchants, missionaries and diplomats. Yet while it was obvious to
Chinese leaders that they faced multiple challenges, none could
anticipate in the 1850s and 1860s or even the 1880s and 1890s that
their system of government would fail in 1911. It may well be that
leaders in other large nineteenth-century polities also lacked the
foresight to recognize growing signs of failure, but in the Chinese
case leaders would confront the end of their imperial system by
developing new strategies and institutions to create a government that
replaced empire. While their success was certainly in doubt for
several decades of war in the first half of the twentieth century,
defeat of the Japanese and the conclusion of a civil war did not lead
to a divided country. It culminated in the establishment of a regime
claiming in large measure to rule all the territories and peoples once
ruled by Qing dynasty. With political stability reestablished the
Chinese could once again enjoy many advantages of a large political
unit. Obviously, the leaders of the PRC did not take effective
advantage of these possibilities until three decades of rule had
passed. In many ways China’s experience since Mao’s death reminds us
that the region had a real economy before the Qing empire’s demise and
that foreign political demands clearly constrained how that economy
could evolve.
The failure of the Qing dynasty manage the transition to a modern
economy turns out to be distinct from the possibility of a large-scale
polity re-emerging after a period of disunion. In one sense, such an
event lends credence to those who suggest that the Communist Party is
simply the most recent “dynasty” in a long line of rulers who have
controlled the Chinese mainland. But in another sense, the Chinese
state that emerges after 1949 is one that can take advantage of
practices begun by earlier generations who had managed to adapt and
adopt a variety of foreign economic, social and political ideas and
institutions. The fact that little advantage was taken of this legacy
until thirty years after the founding of the People’s Republic does
not make those earlier experiences any less relevant to understanding
how and why China has grown so rapidly since the early 1980s.
Understanding how the spatial scale of a polity matters to economic
growth today is, however, a question quite different from the one that
has occupied our attention in this book: how differences in spatial
scale impacted China and Europe before the nineteenth-century
divergence. We have offered an abbreviated sketch of some features of
Chinese political change in the nineteenth and twentieth centuries to
remind the reader of the durability of a spatially large polity on the
Chinese mainland. Empire, as we have used the term, has survived in
China, at the same time as Europe has (since the 1950s) been moving
more explicitly toward political and economic unification. We have
allowed the narratives of European national state making to supply the
norms for political development for too long. If it is disorienting to
realize that Europe has been moving toward a Chinese norm of political
scale rather than China has been moving toward becoming like any
particular European state that is only a measure of the bias of our
long standing approaches. The historical perspective we have gained
here at least begins to correct that bias.
Conclusion: Political Competition and Economic Growth
Although the two ends of Eurasia achieved radically different
political equilibria, the dominant underlying political economy
analysis used to explain both is remarkably similar. For Europe,
scholars have emphasized the importance of institutions of
parliamentary representation and inter-state competition for growth.
Conversely, for China, and despotic governments more generally,
scholars have found only economic stagnation. In Europe the advent of
good institutions was thought to be responsible for the onset of
sustained growth, while in China the stifling oppression of the
omnipotent emperor led to a population living near the Malthusian
minimum. As the reader has discovered, our thesis is rather different.
At the aggregate level, inter-state competition was quite costly and
it certainly had a negative impact on the size of the market, while we
see emperors surviving in part because they cared about their
subjects’ welfare. Nevertheless, the superiority of a particular form
of governance should not be overstated because well into the
nineteenth century massive variation in political structure remained
within Europe and massive variation in levels of well-being
characterized life both within China and within Europe.
In China, scholars have been recently uncovering mounting evidence of
regional differences in income prior to the twentieth century that is
not consistent with an empire whose subjects were barely eking out
more than subsistence. Moreover, imperial policies do not seem to have
been so extortionary that they led to low levels of investment or to
massive poverty. On the contrary it seems that these policies aimed at
expanding the regions of prosperity across the realm. The evidence for
Europe is even more at odds with the old assumptions, given that
representative governance was not consolidated until the last quarter
of the twentieth century (even in the parts of Europe that were not
behind the Iron Curtain). To be sure one could argue (and we have done
so elsewhere) that rulers were unwilling to adopt the more efficient
structures of governance because that would have reduced their power;
but that amendment would be insufficient. The level of economic growth
in Wilhelmine Germany was remarkably robust even though by English or
French standards it was an incomplete democracy. Equally problematic,
the levels of economic achievement of England had few echoes in
Ireland (though it was formally part of the same polity) during the
120 years in which the union between the two countries prevailed. And
these examples are small matters relative to examining either
Austria-Hungary or the Iberia peninsula. In short, the logic whereby
the competitive state system provides great rewards simply does not
hold through Europe. Economically more efficient states, like the
Netherlands in the seventeenth century or England in the eighteenth,
did not gain territory in Europe. States that transformed themselves
may have garnered a higher rate of economic growth, but their
territorial expansion in Europe was nil—to the extent that there was a
reward it came in the form of colonial empires.
We do not mean to suggest that efficient forms of governance neither
exist nor prevail in the long run, but rather that the pressures adopt
representative institutions was weak. Moreover, the impact of reform
was dramatically different across space. Political structures affected
economic growth historically and continue to do so today. But the
putative virtues of European state formation for economic growth have
been mis-specified and contemporary political changes in Europe
suggest that China is at least as much a political norm for effective
state policies toward the economy as any individual European state or
the EU can claim to be. We emphasize that in historical terms
political regimes were adopted largely for fiscal reasons, not because
of a love of liberty or an unwillingness to put up with a corrupt
monarch. Furthermore, the conflict over representation was, more than
anything else, a struggle over the control of expenditures and the
level of taxes. Hence, one cannot argue that representation was
somehow promoted by individuals who wanted to reduce the distortions
inherent to despotic taxation. Rather, these individuals wanted to
strip the power of choosing the level of taxes and the distribution of
expenditures from the sovereign. The European dynamics of political
transformation did matter for economic growth because, as Chapter 3
argued, the competitive state system was directly (though
unforeseeably) responsible for Europe’s adoption of capital intensive
methods of production, while China’s peaceful empire privileged
recourse to labor intensive methods. In this chapter we have seen how
the political structures that were in place in at the time of Charles
V in Europe and the Ming dynasty in China have continued to influence
the process of institutional change. To be sure Europeans are no
longer quite enamored of their parochial privileges but national, and
to a surprising extent provincial, identities already in place in 1500
continue to hamper European unification. China’s growth by contrast is
occurring under the guidance of a very strong center that must
sometime compose with provincial priorities. How the spatial scale of
polities continues to matter to economic growth today is a topic to
which we now turn as part of a more general conclusion about the ways
China and Europe have been changing in recent times.
Conclusion
Historical Perspectives on Politics and Economic Change in China and
Europe:
Findings, Methods and Implications
This book has considered a classic question in economic history: why
did sustained economic growth arise in Europe rather than in China?
The preceding seven chapters argue that political processes drove the
economic divergence between the two world regions. This divergence
became increasingly visible in the nineteenth century but its causes
are located in far earlier times. For centuries, China’s peaceable
empire was more prosperous and more stable than Europe’s warring
polities. Yet, war, which offered to those who lived through it little
more than misery (and even less to those who perished), also produced
a series of distortions that pushed Europe towards urbanization and
capital using technologies several centuries before 1700. To stress
the political contexts of these two world regions does not mean we
wish to overturn the economic arguments. On the contrary, for
pre-industrial economies, the theories of the school of economists
epitomized by Smith and Ricardo are extraordinarily insightful.
The problem with earlier attempts to compare the significance of
political differences for economic development rests upon the
inference that the competition so useful for economic development is
also salutary among polities. That view has relevance for modern times
because, if political actors are themselves subject to the rule of
law, their political campaigns may well impose a far lesser economic
burden than the follies of rapacious dictators. By implication
scholars have concluded that a competitive and innovative Europe
outperformed an imperial and traditionalist China. This volume has
argued against such easy inferences from the contemporary world to the
past. We suggest that the historical costs of political competition
were very high. Although political competition has been overwhelmingly
prevalent through human history and throughout the world, it has
rarely created prosperity. In the past rather than gentlemanly
electoral jousting, political competition involved real internal and
international violence. The need to secure the resources for political
action drives a political actor to intrude into his own economy and
destroy those of his rivals. In historical environments in which
rulers faced few constraints, the economic consequences of such
competition were dire—as Hobbes famously put it, life is nasty,
brutish and short.
The roots of the economic divergence between China and Europe did
indeed lie with their political differences. Yet we view European
political competition less as the source of economic virtue and more
as a vice that reduced the possibility for economic growth. Europe’s
persistent poverty before the late eighteenth century resulted from
the limited domestic realms of rulers and the resulting restrictions
on markets. The rise of capital intensive methods of production that
characterize the modern economy was an unintended consequence of
Europe’s political anarchy, not a carefully crafted result of
government efforts. In contrast, China’s vast and stable empire was
the source of its millennium-long prosperity, a linkage presented in
Chinese historical texts in terms of the state promoting prosperity in
order to sustain a vast and stable empire. Given this, it is no longer
so easy to presume that China failed either because its economic
system was incapable of development or because it was hobbled by some
overarching cultural, environmental or political factors.
**
It turns out that European institutions were not so obviously superior
in the ways that are conventionally believed. We cannot therefore
accept the still common narratives of a European march forward towards
technical breakthrough contrasted with Chinese stagnation. Because we
have evened out the playing field, it becomes worthwhile to study
these economies jointly. We believe the intellectual payoffs from such
a focus are demonstrated in the previous chapters. On one level we
argued that other often invoked economic or cultural factors (e.g.
demography, informality, capital markets) have their roots in the
political processes we highlight or else they fail to stand up to
evidence. On a second level we have traced out the implications of
differences in international relations for technological change,
credit markets and government spending. This has allowed us to show
that the chronic threat of war in Europe produced unanticipated
positive conditions for economic change and its absence allowed the
Qing dynasty to implement policies favorable to Smithian growth but
unlikely to produce industrialization.
Our analysis has been less concerned with explaining precisely when
and why Europe overtook China’s economic leadership than in tracing
the consequences of two political structures (empire and
fragmentation) on economic change. We have built our argument in terms
of increasing likelihoods that new forms of economic production would
emerge in part of Europe than any part of China and have demonstrated
that what drives those different probabilities can all be brought back
to differences in political structures. The more typical comparative
analyses seeking to explain when and why Europe overtook China in the
early modern era, faces two dangers we can more easily avoid. First,
given the state of quantitative information a precise dating is likely
to be inaccurate. In fact, any statement more precise than, “sometime
between 1450 and 1800 per capita income came to be higher in Europe
than China” is unlikely to be very meaningful. This may be a measure
of the dismal precision of social sciences but we should not presume
more. Second, analyses that seek to pinpoint a moment of major shifts
tend also to search for all the factors present in that historical
moment. Such accounts of change are usually quite thick on
description. They thus invoke many causal factors whose relative
importance or significance are difficult to discern. By arguing for
the growing probability that Europe more than China would be the world
region where modern economic development would begin, we offer a kind
of explanation similar to those more common in the social sciences: a
thesis about expected likelihoods of certain events or effects taking
place given the presence of certain other conditions or factors.
Our approach to comparative economic history differs significantly
from those currently on offer. Rather than one big theory, our
explanation relies on a number of small sharp theories. Each theory or
model has clear implications both for differences in the structure of
economic activity between Europe and China and within each region. For
instance, in Chapter Two we consider the effect of household structure
on the labor market by positing a model of how household structure
affects the size of the labor market and then we formulate a series of
propositions about the average skill of wage earners in economies with
different household structures. We seek to be explicit in creating
specific causal chains—because such chains can be fruitful, in
particular when large amounts of data are unavailable. In Chapter 4,
to briefly offer a second example, we use first a Leontieff production
function and then a Cobb-Douglas production function to work through
the effects of war on relative factor prices, thus revealing
differences in urban versus rural locations of production as a
function of the fear of military disturbances. We move from a static
model to a dynamic argument that considers war’s influences on
relative factor prices and the direction of technical change. At all
stages, the links in our reasoning are explicitly identified and
evaluated.
Our comparative economic history is economic because it consciously
applies economic theories to the questions we face. It is explicitly
comparative and historical because we attend to various elements of
context, in particular seeking to explain how specific sets of
institutions operate in different settings, be these household
structures and kinship systems, financial markets and credit
practices, or commercial dispute resolution by government officials
and merchants themselves. Our scales of comparison take China and
Europe as large and different world regions within each of which there
is all manner of variation. We argue that variation in some phenomena,
such as intensities of commercial production, should arise both within
each region and between China and Europe for simple economic reasons.
Among the differences that emerged between China and Europe we
distinguish those for which political factors were most crucial.
Our strategy of analysis applies a number of general principles to
specific regions over long periods of time. We are by no means modest
in our ambition yet our claims are certainly bounded—they do exist
within certain contexts. We do not offer any universal explanation of
economic change or general theory about the impact of politics on
economic change. Indeed, we are somewhat skeptical that much universal
explanation is plausible in the social sciences, historical or
otherwise. More specifically, we have offered an explanation for why
modern economic growth began in Europe rather than China. Many of our
explanations are specific to major aspects of this large problem. A
few are more general, such as the argument in Chapter Six about the
composition of public goods and levels of taxation in China and
Europe; we explain both why China had lower taxation and higher public
good provision that Europe did before the nineteenth century and how
China’s levels of taxation subsequently rose and public good provision
fell for reasons similar to those at work in an earlier period of
European history. While circumstances have changed, in particular
military budgets have shrunk relative to other government spending,
tensions over fiscal policy remain at the core of politics.
Furthermore, and as we discuss below, the institutions that distribute
power between the center (Beijing or Brussels) and the provinces or
countries (e.g. Sihuan or Guangdong, Spain or Sweden) have tremendous
persistence.
**
Overall, however, our explanation of why modern economic growth began
in Europe rather than China has stopped around 1800. In this book we
seek to understand the factors that caused the great divergence in
technical change and that process was completed by 1800. Thus we have
not discussed much nineteenth or twentieth-century material nor
evaluated other world regions outside of China and Europe. Certainly,
the process of economic growth has changed during the nineteenth and
twentieth centuries and thus includes new problems and possibilities
we have not had reason to consider. Nevertheless our historical
perspective on institutional change has implications for how we view
twentieth-century transformations in China and Europe, as well as what
we might anticipate in the future. We argue that institutional change
is always, at least in part, an extension and elaboration upon
previous practices, whether consciously conceived as such or not.
Moreover, contrasts between China and Europe help to highlight the
challenges these regions face and the opportunities they can seize.
The global twentieth-century economic environment is, of course,
fundamentally different from the settings in China and Europe with
which we have been principally concerned in this book. Technical
progress and political change have altered both the kinds of
institutions people can construct and the choices they are likely to
make. For instance, the importance of relative factor costs for
production choices so basic to our Chapter Four account of
manufacturing locations in early modern China and Europe matters far
less in the twentieth century. Nowadays, war is not so important and
entrepreneurs and policy makers throughout the world pursue capital
deepening. Local variations in relative factor prices may affect the
process but even the most labor intensive outsourcing involves capital
deepening in poor economies. Similarly, changes in labor markets and
demography render the arguments we analyzed in Chapter Two quite
irrelevant after 1900 or so: firm size is now so large that household
structure is pretty much irrelevant for labor markets. We are neither
surprised nor dismayed to confirm that some of our substantive
analyses work for particular times and places and do not readily
extend to other cases. On the contrary, these limitations are what we
expect of many explanations in the social sciences.
Many scholars accept a periodization of history in which the end of
World War II marks a significant rupture with the preceding decades
and previous centuries and thus would be naturally skeptical that much
of what we have considered in this book could matter to the past
half-century. As a result readers may not be especially disturbed by
the reminder that some of our empirical analyses are not directly
relevant to the study of the present-day world. If this temporal
divide created a consistent division of labor between those working on
earlier periods and those working on the recent past, scholars might
comfortably continue along their separate ways to develop their
distinct literatures. Yet, this is hardly possible since so much of
the social sciences and humanities makes claims to levels of
generality that depend upon propositions persuasive for the present
being plausible for earlier eras. Many of us are quite ready to look
upon the past with our eyes fixed largely on the present. Similarly,
many scholars are comfortable making the ideas and institutions of
Europe and neo-Europes the norms by which we generalize about the
world, an ease demonstrated by Violence and Social Orders by North,
Wallis and Weingast, already discussed in our introduction. Much of
our book has aimed to counter this latter convention. We have put
China and Europe on the same analytical platform and, guided by some
basic principles of economic theory and knowledge of Chinese and
European history, we have evaluated factors of possible significance
for the economic performance in the past. Now we turn to the post
World War II era to suggest that much about the recent pasts and
possible futures of the Chinese and European economies can be better
understood by including an understanding of history.
Because institutional changes take place in particular contexts with
important historical dimensions, politics can always influence
economic practices. We have made much of the recurring capacity of
states to create ideas and institutions of empire across the Chinese
mainland and the absence of a comparable capacity in Europe. For both
China and Europe, the years from 1914 to 1947 were a succession of
catastrophes, most of which had political origins and an international
scope. Although economic growth might have been rapid in some places
during the 1920s, the longer period bracketed by the two world wars
was a very dark period at both ends of Eurasia which were beginning to
converge toward polities of more similar size. In particular, war-torn
China fragmented and began to resemble a more familiar fragmented
Europe. Thus, if we were to focus on the first half of the twentieth
century, it would be easy to conclude that our contrast is no longer
relevant: the empire was vanishing.
Fast forward to the late twentieth century and we witness a reunified
China and a Europe moving in fits and starts toward reduced
competition, more coordination and even integration. Europeans began
to achieve a spatial level of political coordination and economic
integration that China repeatedly achieved in earlier periods of
history and continued to pursue after the founding of the People’s
Republic in 1949. Europeans were encouraged to achieve greater
political cooperation among themselves because the globe had been
divided by the Cold War. The fault lines created by the Cold War made
intra-European political competition less plausible. Although economic
recovery from World War II was pursued at the national level, it was
framed within a new international competition between the capitalist
West and socialist East. The communist threat made the battle lines of
the world wars obsolete and enabled the seeds of European economic
integration to be sown. Yet economic coordination remained limited and
integration slowed since regional policy making continued to be
hostage to nationalist visions of economic growth. We can look back
from the present and see the precursors to the European Union in such
institutions as the European Coal Commission and later the Common
Market, but these were hardly key components of political policies
that framed economic activities. The more visible flowers of
unification were to bloom decades later.
In China a different rupture with earlier practices took place in the
1950s and 1960s. By the mid-1950s central planning replaced markets
that had spanned urban and rural areas and that had induced many
people to adopt technologies and institutions first formulated in the
West. Although many of the economic practices and their institutional
settings we have analyzed in previous chapters were demolished by the
Communist regime, certain key political and economic elements remained
or resurfaced at various points after 1949. The People’s Republic
formed a unitary centralized state governing virtually all the
territory amassed by the Qing Empire at its height. The advantages of
centralized bureaucratic rule as well as the institutional limitations
of such rule were rediscovered by the Communists, even as they forged
a political ideology and institutions consciously owing more to Soviet
influences than to earlier Chinese ones. The ideological and
institutional ruptures between the late imperial past and the
socialist present of the 1950s and 1960s obscures just how much these
two unitary and centralized states shared.
To appreciate the significance of China’s late imperial past to its
present and future practices, we must consider some persistent
differences in Chinese and European political economies. If we turn to
contemporary public finance and recall the argument presented in
Chapter 6—that the eighteenth-century Chinese state made all taxation
decisions at the central level—and contrast this to the absence of an
EU level of government in this same era, we can uncover some of the
bases upon which we hold very different expectations of public finance
in Beijing and Brussels. For EU administrators to acquire a budget
equal to 10% of EU GDP would require a far greater transfer of
sovereignty than most Europhiles contemplate. At the same time it is
difficult to imagine the central government in China managing such a
small percentage of Chinese GDP in the future. Similarly, Beijing
produced a fiscal stimulus response to the 2008 financial crisis that
far exceeded what Brussels could even imagine. The Chinese approach
combined funds from the center with directions for provincial-level
stimulus targets. It also left many details for provincial authorities
to decide. In its structure it is more than reminiscent of the ways in
which mid-eighteenth century Chinese officials mounted famine relief
campaigns that involved central government authorities making plans
and the coordination of the efforts of multiple provincial-level
administrations. The EU level response to the 2008 financial crisis
was simply palid in contrast. Bailouts and fiscal stimulus packages
were left to the national governments. Beijing, by contrast, put up
half the funds for its stimulus package and dictated the kinds of
infrastructure projects that would be supported. Even more recently,
Brussels has no funds to response to the Greek financial crisis, at
best it can coordinate the different national governments. In the end
each member state decides whether to help out or not. The EU simply
does not have the money to do much, and of course, Europe had no early
modern parallel to China’s famine relief campaigns.
Over the past three decades, China has embarked on processes of
economic transformation that promise a great deal of improvement for
extraordinarily large numbers of people. The number of potential
consumers in these countries has made the heads of global firms giddy
with anticipation. But exactly what are the bases for such stellar
performance? Among contemporary China specialists it is generally well
understood that growth in the 1980s largely occurred without the
benefit of the formal institutions deemed so important both to
Europe’s economic history and the prescriptions made for development
in the contemporary world. Much of China’s industrial growth in the
1980s and early 1990s was produced by township and village enterprises
(TVE), firms outside the state plan that lacked clear property rights
structures and engaged in exchanges without the benefits of a court
system to enforce contracts. One hears some China scholars and
observers remark that this was a natural way to begin growth. It did
not pay for Chinese officials to develop formal institutions to manage
production and exchange early on. Instead, informal institutions could
shoulder the burden until China became rich enough that it could
afford to improve its legal infrastructure. That scholarship leaves in
the dark why such “natural” growth experiences don’t occur more
generally throughout the world and conversely why post communist China
was able to rely on informal institutions during the explosive TVE
growth period. Our account in Chapter 3 suggests that informal
institutions had long been important historically in China not to
palliate failed formal institutions but as complements that enabled
market exchange. Chinese policies after 1949 took away many informal
institutions and put in their place formal institutions quite
different from those of a market economy. Mao’s radical rule was brief
enough that the earlier history was not forgotten. When in the mid
1970s leaders decided to allow and accept growth outside the formal
state sectors and plans, people depended greatly on earlier informal
institutions as the basis upon which they began to pursue development.
Chinese economic growth in the 1980s was thus built upon the past.
Since the early1990s policies and economic conditions have changed and
so have Chinese enterprises. Industrial production increasingly takes
place in sophisticated factories whose owners and managers require
more clearly stipulated property rights than those of the TVE era. Yet
contract enforcement remains uneven at best and not all property
rights deemed necessary and appropriate in Euro-American contexts have
been specified clearly in Chinese situations. We believe such
differences can often be explained at least in part by preferences and
practices of earlier eras. The extension of our argument in Chapter
Three to more recent conditions counsels us to avoid simple
projections about institutional convergence that accompanied the once
popular “end of history” kinds of arguments.
Similarly, our analysis does not support the view that China’s
development can serve as a refutation of American economic practices
specifically and Western ones more generally. It also does not support
any arguments about a “clash of civilizations.” It has become popular
to see different economic practices as evidence of persistent
differences that make foes of people in different world regions. As we
have repeatedly indicated, economic principles at work in one world
region apply equally well in another. The differences we have found
depend on history’s influence over institutions which are, to some
degree, always embedded in broader social contexts that have features
distinguishing them from others.
This perspective figures prominently in our account of credit
institutions and financial markets in Chapter 5. In spatial terms we
developed and applied an argument to explain variation among early
modern European situations as well as between them and the far less
well documented range of situations in late imperial China. Different
types of debt were developed in Europe and financial markets in many
countries responded to new demands for credit. European governments
did not collect large amounts of data on private financial markets,
but their policies did greatly influence the institutional
particularities of financial markets. The diversity of financial
practices did not however produce clear and important economic
differences among different parts of Europe. Europe was able to
tolerate financial diversity and variation with no sharp impact on
economic efficiency. Scholarship on Chinese economic history has yet
to discover and analyze data that would allow us to assess the nature
of variation across the empire. However, we have been able to show
both that the absence of European-style financial institutions does
not mean Chinese were bereft of credit mechanisms, and that it is
unlikely that the limitations of capital availability were a crucial
constraint on economic growth in the era preceding the Industrial
Revolution. Moreover, when we turn to the recent past we can see that
politicians from different Europeans countries have continued to be
willing to pay a high price for financial diversity—most notably by
the regulatory failures that led to the Icelandic financial collapse.
In contrast, a strong centralized government in Beijing after 1949 was
able to redefine the institutional bases for Chinese credit
institutions and financial markets, thereby asserting its capacities
to define formal institutions. Its ability to coordinate banking
policy has avoided the tensions and inconsistencies that have plagued
the EU. This contrast is plausible despite the repeated rounds of
banking and financial reforms that have changed the formal system and
permitted some local and less formal forms of financing to thrive.
Irrespective of the relative virtues of the Chinese and European
financial systems it is at best premature to anticipate that Chinese
practices should converge toward any European or American practices
that can be confidently assessed as superior. Some of the formal
reforms undertaken by China have made some of its banking practices
conform more closely to international standards defined by some mix of
European and American practices. Yet, the Chinese financial system
remains very distinct from those present in Europe and the United
States. Both are subject to reforms, some of which make them more
similar. Other features, less commonly remarked upon, reflect their
persistent differences.
**
The politics of contemporary economic differences have historical
dimensions that almost all observers of the contemporary world ignore.
We suspect that such ignorance handicaps our abilities to anticipate
the likely range of future changes in any of these situations. This
book is certainly not intended primarily to proclaim the virtues of
historical social sciences for confronting present problems and
imagining future possibilities. It has taken on a more modest
challenge of exploring and explaining the relative performance of
China and Europe over many centuries. We chose to do so by combining
our different expertises; that in turn forced us to reconsider and
reject some approaches to comparative economic history.
In understanding persistent differences in economic institutions,
social scientists have become fond of frameworks that emphasize the
long shadow of history. A variety of cognitive, cultural or political
factors conspire to make that shadow so powerful that societies become
locked into specific institutions. Whether these are informal
institutions, religious constraints, or family practices, these modes
of behavior lie outside the standard policy domain and largely doom
these societies to poverty. Even when they don’t suggest that
institutional change is virtually impossible because of path-dependent
constraints, social scientists have come to recognize the tremendous
difficulty some societies experience in making institutional
innovations. (Laporta et al. 1997, 1998, Acemoglu et al. 2001,
Engerman and Sokoloff 1997). Even though a set of technologies and
institutions that massively raised individual welfare have been
developed, only a fraction of the world has yet to take full advantage
of these innovations. Explaining the persistent differences between
societies that have embarked upon modern economic growth and those
that have not done so attracts scholarly attention to institutional
differences. With respect to such arguments about different kinds of
path-dependent institutional lock in, we suggest that such ideas be
pursued with extreme caution: appearances can be very deceiving. What
seem to be path-dependent institutions can change rapidly, if the
economic or political contexts change. To the extent that we seek a
revolution, ours is more narrowly political. Rather than consign some
societies to poverty in the absence of radical cultural change, one
should seek to alter the political structures that shackle growth. Our
research indicates, that this would require more than the simple
transcription of western models (e.g. democracy). Rather, the history
of China and Europe favor a gradual evolution, one where either
indigenous elements are transformed to serve new purposes or where
external institutions are inserted into a local structure.
This economic history also argues that if social science is to
contribute to the process of change, it must ally local historical
expertise with the abstract concerns of economic and political theory.
This is not so easy as it seems because of disciplinary conflicts. Too
often economists consign the knowledge of historians to the bin of
irrelevant details while historians and area studies scholars treat
economic theory as a construct with little relevance to the real
world. It is also difficult because scholars working on Europe or
North America have more often than not thought to evaluate specific
institutions with a home country bias. From their point of view
because Europe experienced economic growth earlier than elsewhere, it
must have had better institutions. This has led them to a line of
inquiry that rationalizes the inferiority of alternative structures.
Economists are particularly tempted by such analyses because they fit
neatly in the discipline’s focus of optimal decision making.
Unfortunately, as the preceding chapters show, such approach can lead
one seriously astray.
This set of observations on our method of comparative economic history
and its virtues for analysis both historical and contemporary
concludes our book. We have sought in the preceding seven chapters to
provide a combination of Chinese and European historical narratives
and economic analysis adequate to persuade you that understanding the
politics of economic change in China and Europe before the Industrial
Revolution is both possible and useful. In particular we suggest these
efforts make us better able to identify key reasons for the economic
divergence between these two world regions than those previously
proffered. If we have achieved any success at reaching our objective,
we hope you might also consider the book’s approach for subjects far
beyond our particular historical subject that lie in the recent past
and will confront us all in the future.
Bibliography
Cambridge History of China vol 6, chs 4-9
Acemoglu, Daron, James Robinson, and Simon Johnson. 2001 “The Colonial
Origins of Comparative Development: An Empirical Investigation”,
American Economic Review, 91, 1369-1401.
_____, 2005. “The Rise of Europe: Atlantic Trade, Institutional
Change, and Economic Growth,” The American Economic Review 95, no. 3:
546-79.
Adshead, S.A.M. 2004. T’ang China: The rise of the east in world
history. Palgrave.
Alesina. Alberto, and Enrico Spolaore. 2003. The Size of Nations,
Cambridge, MA: MIT Press,
Allen, Robert C. 2001. “The Great Divergence in European Wages and
Prices from the Middle Ages to the First World War.”Explorations in
Economic History. 38: 411-47
_____, 2004. “Mr. Lockyer meets the Index Number Problem: the standard
of living in Canton and London in 1704.” Mimeo Oxford.
______, 2007. “The Industrial revolution in Miniature: the Spining
Jenny in Britain France and India.” Mimeo Oxford.
______, 2009. The British Industrial Revolution in Global Perspective.
Cambridge: Cambridge University Press.
Allen, Robert C., Jean-Pascal Bassino, Debin Ma, Christine
Moll-Murata, and Jan Luiten van Zanden, 2007. “Wages, Prices, and
Living Standards in China, 1738-1925: in comparison with Europe,
Japan, and India,” Mimeo Oxford.
Altorfer, Stefan, 2004. “The Canton of Berne as an Investor on the
London Capital Market in the18th Century.” LSE working paper
Anderson, B. L. 1969a. "Provincial Aspects of the Financial Revolution
of the Eighteenth century," Business History, n° 1, p. 11-22.
___________. 1969b."The Attorney and the Early Capital Market in
Lancashire," in R. R. Harris, ed., Liverpool and Merseyside: Essays in
the Economic and Social History of the Port and its Hinterland.
London. 50-77. (reprinted in F. Crouzet ed., Capital Formation in the
Industrial Revolution, London, Methuen 1972).
Arrighi, Giovanni. 2007. Adam Smith in Beijing: Lineages of the
Twenty-First Century. New York: Verso.
Bairoch, Paul, Jean Batou, and Pierre Chèvre. 1998. The Population of
European Cities From 800 to 1850. Geneva: Droz.
Bates, Robert, Avner Greif, Margaret Levi, Jean-Laurent Rosenthal and
Barry Weingast. 1998. Analytic Narratives. Princeton NJ: Princeton
University Press.
Béaur. Gérard 2000. Histoire agraire de la France au XVIIIe siècle
: inerties et changements dans les campagnes françaises entre 1715 et
1815 Paris : SEDES.
Becker, Gary, 1981. A Treatise on the Family. Cambridge, Mass.:
Harvard University Press.
Beik, William. 1989. Absolutism and Society in Seventeenth-century
France: State Power and Provincial Aristocracy in Languedoc. Cambridge
and New York: Cambridge University Press
Bielenstein, Hans. 1999. “Wang Mang, the restoration of the Han
dynasty and Later Han,” in Michael Loewe, ed. The Cambridge History of
China, vol. 1, PAGE Cambridge University Press
Berg, Maxine, 1994. "Factories, Workshops, and Industrial
Organization," in Floud R and D.N. McCloskey eds. The Economic History
of Britain Since 1700 : 1700-1860. 2nd edition Vol. 1 Cambridge
University Press.
_________, 1986 The Age of Manufactures; Industry, Innovation and Work
in Britain 1700-1820. Oxford: Osfrod UP.
Bernhardt, Kathryn. 1992. Rent, Taxes and Peasant Resistance: The
Lower Yangzi Region, 1840-1950. Stanford University Press.
Beveridge William,1965. Prices and wages in England from the twelfth
to the nineteenth century. London : F. Cass, 1965.
Bisson, Thomas N. 2009. The Crisis of the Twelth Century; Power,
Lordship, and the Origins of European Government.
Bodde, Derk. 1999. “The state and empire of the Ch’in,” in Michael
Loewe, ed. The Cambridge History of China, vol. 1, PAGE Cambridge
University Press
Bonney, Richard, ed. 1995. Economic Systems and State Finance. New
York: Oxford University Press.
Bonney, Richard, ed. 1999. The Rise of the Fiscal State in Europe, c.
1200 1815 New York: Oxford University Press, 1999.
Bonney, Richard. 2007. Vindication of the Fronde? The cost of Louis
XIV’s Versailles Building Program.” French History. 21(2):205-225
Bourgon, Jerome. 2004. “Rights, Freedoms, and Customs in the Making of
Chinese Civil Law, 1900-1936” in William Kirby, ed., Realms of Freedom
in Modern China. Stanford University Press. 84-112.
Botticini, Maristella, 2000. "A Tale of 'Benevolent' Governments:
Private Credit Markets, Public Finance, and the Role of Jewish Lenders
in Medieval and Renaissance Italy." Journal of Economic History 60
(March 2000): 164-89.
Boyer-Xambeu, M.T., G. Deleplace, and L. Gillard. 1995. Bimetalisme,
taux de change et prix de l’or et de l’argent; 1717-1873. Cahier de
L’I.S.M.E.A. Serie A.F. #19-20, Paris.
Brandt, Loren. 1989. Commercialization and Agricultural Development:
Central and Eastern China, 1870-1937. Cambridge and New York:
Cambridge University Press.
Brandt, Loren,Scott Rozelle and Matthew Turner. 2004. Local Government
Behavior and Property Right Formation in Rural China Journal of
Institutional and Theoretical Economics (JITE), 2004, 160, (4), 627-
Braudel, Fernand, 1967. Civilisation matérielle et capitalisme,
XVe-XVIIIe siècle Paris: A. Colin. English translation. Capitalism
and material life, 1400-1800; translated from the French by Miriam
Kochan: New York : Harper Colophon.
Braudel, Fernand. 1966. The Mediterranean and the Mediterranean World
in the Age of Philip II. Second Revised Edition ed. Glasgow: William
Colins & Sons.
Brennan, Thomas, 1997. From Burgundy to Champagne; the Wine Trade in
Early Modern France. Baltimore Johns Hopkins University Press.
Brenner, Robert. 1976. “Agrarian Class Structure and Economic
Development in Pre-Industrial Europe” Past and Present, No. 70 (Feb.,
1976), pp. 30-75.
Brewer John, 1989. The Sinews of Power: War, Money, and the English
State, 1688-1783. New York: Alfred A. Knopf.
Brown, Henri Phelps, and Sheila V. Hopkins. 1981. A perspective of
wages and prices London ; New York : Methuen, 1981
Bury, J.B.1928. The Invasion of Europe by the Barbarians; a series of
Lectures. London McMillan and Co.
Caferro, William. 1998. Mercenary Companies and the Decline of Siena.
Johns Hopkins Press.
Calomiris, C.W., 1995. The costs of rejecting universal banking:
American finance in the German mirror, 1870– 1914. In: Lamoreaux,
N.R., Raff, D.M.G. (Eds.), Coordination and Information: Historical
Perspectives on the Organization of Enterprise. University of Chicago
Press, Chicago, pp. 257–315.
Campbell, Bruce.2006, ‘The land’, 179-237 in M. Ormrod and R. Horrox,
eds., Social history of England, 1200-1500, Cambridge, Cambridge,
Cambridge University Press.
Campbell, J.B. War and Society in Imperial Rome, 31 BC-AD 284: 31 BC -
AD 284. London: Routledge, 2002
Chaianov, Aleksandr Vasilevich. 1966. The Theory of Peasant Economy.
Homewood, Ill., R.D. Irwin, 1966.
Ch’iu Pengsheng. 2008. Dangfalu oushang jingji: Ming Qing Zhongguo de
shangye falu (When law meets economics: commercial law in Ming Qing
China). Wunan tushu chuban gongsi.
Clark, Gregory, 2007. A Farewell to Alms: A Brief Economic History of
the World. Princeton NJ: Princeton University Press)
Coase, Ronald, 1937. “The Nature of the Firm” Economica, New Series,
Vol. 4, No. 16, pp. 386-405.
Collins Michael. 1991. Banks and Industrial Finance in Britain
1800-1939. Cambridge: Cambridge University Press.
Courdurie, Marcel. 1974. La dette des collectivités publiques de
Marseille au XVIIIe siècle: du debat sur le pret a interet au
financement par l'emprunt. Marseille, Institut historique de Provence.
Davis, Lance E. And Robert A Huttenback. 1986. Mamon and the Pursuit
of Empire; The Political Economy of British Imperialism 1860-1912.
Cambridge: Cambridge University Press.
Delille Gérard. 2000. Le maire et le prieur: Pouvoir central et
pouvoir local en Méditerranée occidentale (XVe–XVIIIe siècle).
Civilisations et sociétés, volume 112. Bibliothèque des écoles
françaises d’Athènes et de Rome, volume 259/2. Paris: Éditions de
l’École des Hautes Études en Sciences Sociales; Rome: École française
de Rome, 2003.
De Moor, Tine and Jan-Luiten Van Zanden. 2008. “Girl Power”,
Forthcoming Economic History Review 2008.
De Roover, Raymond 1953. L’e̲volution de la lettre de change,
xive-xviiie sie̲cles. Paris : Armand Colin.
de Vries, Jan, 1984. European Urbanization, 1500-1800. Harvard
University Press.
de Vries, Jan and Ad van der Woude.1997. The First Modern Economy.
Cambridge University Press.
Di Cosmo, Nicola. 2004. Ancient China and its Enemies: The Rise of
Nomadic Power in East Asian History. Cambridge University Press.
Dincecco, Mark. 2008 “Fragmented Sovereignties: A Quantitative
Analysis of Divided Authority on the European Continent, 1700-1815”
IMT working paper.
________, 2009. “Fiscal Centralization, Limited Government, and Public
Revenues in Europe, 1650-1913” Forthcoming, Journal of Economic
History, 2009
Deng Gang. 1993. Development versus Stagnation: Technological
continuity and agricultural progress in pre-modern China. Greenwood
Press.
Deyon, Pierre. 1996.”Proto-Industrialization in France,” in Ogilvie
and Cernan eds. European Proto-Industrialization. Cambridge: Cambridge
University Press. 38-48.
Di Cosmo, Nicola. 2002. Ancient China and its enemies: the rise of
nomadic power in East Asian history. Cambridge University Press.
Diamond, Jared. 1997. Guns, Germs, and Steel : The Fates of Human
Societies. W W Norton.
Dickson, P.G.M. 1967. The Financial Revolution in England. London.
Ding, Changxian [1987], " Zhongguo Jindai Jiqi Mianfang Gongye Shebei,
Ziben, Chanliang, Chanzhi de Tongji he Guliang (Statistics and
Estimates of Industrial Equipment, Capital, Production and Value of
Chinese Modern Cotton Industry)"(C), Zhongguo Jindai Jingji Shi Yanjiu
Ziliao (Materials on the Study of the Economic History of Modern
China) vol. 6, Shanghai Academy of Social Science Publishing.
Domar, Evsey. 1970. "The Causes of Slavery or Serfdom: A Hypothesis,"
Journal of Economic History. (March).
Drelichman, Mauricio. 2005. "The Curse of Moctezuma: American Silver
and the Dutch Disease." Explorations in Economic History 42 (3):
349-380. July 2005.
Delichman, Mauricio and Hans-Joachim Voth. 2009. Lending to the
Borrower from Hell: Debt and Default in the Age of Philip II,
1556-1598. Mimeo UBC
Ehrenberg, Richard. 1922. Das Zeitalter der Fugger: Geldkapital und
Creditverkehr im 16. Jahrhundert. 3rd ed. Jena.
Lüthy, Herbert. 1959-61. La banque protestante en France de la
révocation de l'Edit de Nantes à la Révolution. Paris, SEVPEN.
Hayhoe,Jeremy. 1988. Enlightened feudalism: seigneurial justice and
village society in eighteenth centnry northern Burgundy. Rochester NY
University of Rochester Press.
Brennan, Thomas, 1997. From Burgundy to Champagne; the Wine Trade in
Early Modern France. Baltimore Johns Hopkins University Press.
Dreze, J and Amartya Sen. 1989. Hunger and public Action. New York:
Oxford University Press.
Duby Georges, 1974. The Early Growth of the European Economy ;
Warriors and Peasants from the Seventh to the Twelfth Century. Ithaca:
Cornel Universty Press
____________, 1979. L'économie rurale et la vie des campagnes dans
l'Occident médiéval, France Angleterre Empire IX-XVE siècles essai de
synthèse et perspectives de recherches. Paris: Flammarion Champs.
Dykstra, Maura. n.d. “Commercial Dispute Mediation in Chongqing: 1875
– 1949.” UCLA Mimeo
Dyson, Henry 2003. French Property and Inheritance Law: Principles and
Practice. Oxford University Press, 2003
Elliott, John H. The Count-Duke of Olivares: The Statesman in an Age
of Decline New-Haven: Yale University Press, 1986.
Elman, Benjamin. 2005. On Their Own Terms: Science in China, 1550-1900.
Harvard University Press.
Elvin, Mark. 1973. The Pattern of the Chinese Past. Stanford
University Press.
________. 1972. “The High Level Equilibrium Trap: the Causes of the
Decline of Invention in the Traditional Textile Industries. In E.
Willmott (ed.) Economic Organization in Chinese Society. Stanford:
Stanford University Press.
Emigh, Rebecca, 2003. “Property Devolution in Tuscany” Rebecca Jean
Source: Journal of Interdisciplinary History, Vol. 33, No. 3 (Winter,
2003), pp. 385-420
Engerman, Stanley. and Kenneth Sokoloff. 1997. “Factor Endowments,
Institutions, and Differential Growth Paths Among New World Economies:
A View from Economic Historians of the United States.” in Stephen
Haber, ed. How Latin America Fell Behind: Essays on the Economic
Histories of Brazil and Mexico, 1800-1914. Stanford University Press.
260-304.
Epstein, Stephan R. 2000. Freedom and Growth: The Rise of States and
Markets in Europe,1300-1750, Routledge, London.
Fan Jinmin 1998 Ming Qing Jiangnan shangye de fazhan. (The development
of commerce in Ming Qing Jiangnan). Nanjing University Press. 1998
Fan Jinmin 2007. --天津商会— 在内容上大量模仿了外国商法，而脱离了本国伤情，亲戚上任的不满，
“政府颁布商事法令，每不与商人协议，致多拂逆商情之处，是非徒不足以资保护，而且转多室碍.”)
Faure, David. 2007. Emperor and ancestor: State and lineage in South
China. Stanford University Press.
Ferrie, Joseph P. 1999. Yankeys Now’: European Immigrants in the
Antebellum U.S., 1840-60 (New York: Oxford University Press, 1999).
Frank, Andre Gunder. 1998. ReOrient : global economy in the Asian Age.
Berkeley: U of California Press.
Fu Yiling. 1956. Ming Qing shidai shangren ji shangye ziben (Merchants
and merchant capital in the Ming Qing period). Renmin chubanshe.
Fuji Hiroshi. 1953-1954. “Shin’an shōnin no kenkyū” (A study of the
xin’an merchants) Tōyō Gakuhō36.1:1-44;2:32-60; 3:65-118; 4:115-45.
Fujita, Masahisa, Paul R. Krugman, and Anthony Venables. 2001. The
Spatial Economy: Cities, Regions and International Trade. Cambridge
Mass: MIT Press.
Galenson, David W., and Clayne L. Pope,1989. ”Economic and Geographic
Mobility on the Farming Frontier: Evidence from Appanoose Country,
Iowa, 1850-1870 The Journal of Economic History, Vol. 49, No. 3 (Sep.,
1989), pp. 635-655
Geary, Patrick. J. The Myth of Nations: The Medieval Origins of
Europe. Princeton: Princeton University Press, 2003
Gelderblom Oscar, 2000. Zuid-Nederlandse kooplieden en de opkomst van
de Amsterdamse stapelmarkt (1578-1630). Hilversum 2000.
___________, 2008. Confronting Violence and Opportunism. The
Organization of Long-Distance Trade in Bruges, Antwerp and Amsterdam,
c.1250-c.1650.
Gelderblom Oscar and Joost Jonker. 2004. “Completing a Financial
Revolution. The Finance of The Dutch East India Trade and the Rise of
the Amsterdam Capital Market, 1595-1612”, The Journal of Economic
History, 64-3, September 2004, 641-672.
______________, 2006. Exploring the market for government bonds in the
Dutch Republic (1600-1800). Mimeo Utrecht.
_______________, 2008. Collective Spirit or Aggregate Wealth?
Understanding the Structure and Growth of Holland’s Public Debt,
1514-1713. Mimeo Utrecht.
Gerschenkron Alexander 1962. Economic Backwardness in Historical
Perspective: A Book of Essays. Cambrdige Mass. Belknap Press.
Graff, David.2002. Medieval Chinese Warfare, 300-900 Routledge 2002\
Grant. Michael. 1978. History of Rome. New-York: Scribners and Son
Grantham, George. 1993. :Divisions of Labour: Agricultural
Productivity and Occupational Specialization in Pre-Industrial
France.” The Economic History Review, New Series, Vol. 46, No.3, pp.
478-502.
Greif, Avner. 1989. "Reputation and Coalitions in Medieval Trade:
Evidence on the Maghribi Traders." Journal of Economic History, Vol.
XLIX, No. 4 (Dec. 1989): 857-882.
________. 2006. Institutions and the Path to the Modern Economy;
Lessons from Medieval Trade. Cambridge University Press.
Goetzmann, William and Elisabeth Koll. 2006. "The History of Corporate
Ownership in China: State Patronage, Company Legislation, and the
Issue of Control." Chap. 2 in A History of Corporate Governance around
the World: Family Business Groups to Professional Managers, edited by
Randall K. Morck. University of Chicago Press,
Goodman, Martin. 1997. The Roman World; 44BC-AD180. London: Routledge.
Guinnane, Timothy W., 2002. Delegated monitors, large and small:
Germany’s banking system, 1800–1914. Journal of Economic Literature
40, 73–124.
Guinnane, Timothy W, Ron Harris, Naomi R. Lamoreaux and Jean-Laurent
Rosenthal. 2007. “Putting the Corporation in its Place,” Enterprise
and Society. Vol 8, (3) 2007, 687-729
Gutmann, Myron. 1980, War and Rural Life in the Early Modern Low
Countries. Princeton, Princeton University press.
Haber, Stephen, 1991. Industrial concentration and the capital
markets: a comparative study of Brazil, Mexico, and the United States,
1830-1930. The journal of economic history 51 (3):559-580
Hajnal, John. 1965. “The European marriage patterns in perspective” in
D. Glass and D. Eversely Population in History. London: Edward Arnold
Ltd.
Hale, J. R. 1985. War and Society in Renaissance Europe. Johns Hopkins
University Press.
Hall, John A. 1985. Powers and Liberties: The Causes and Consequences
of the Rise of the West. Basil Blackwell.
Hamashita Takeshi. 1989.Chūgoku kindai keizaishi kenkyū (Studies on
modern Chinese economic history). University of Tokyo Institute of
Oriental Culture.
Hanley, Anne G. 200-5. Native Capital: Financial Institutions and
Economic Development in São Paulo, Brazil, 1850–1920. Stanford,
Calif.: Stanford University Press, 2005.
Harris, Ron. 2005. “The English East India Company and the History of
Company Law” in VOC 1602- 2002 - 400 Years of Company Law , Series Law
of Business and Finance, Volume 6, Eds. Ella Gepken-Jager, Gerard van
Solinge, Levinus Timmerman, Kluwer Legal Publishers p. 219-247.
Harris, Ron, The Institutional Dynamics of Early Modern Eurasian
Trade: The Commenda and the Corporation (November 3, 2008). Available
at SSRN: http://ssrn.com/abstract=1294095
Harrison, Lawrence, and Samuel Huntington, eds., 2000. Culture
Matters: How Values Shape Human Progress. New York : Basic Books.
Heather, Peter. 2006. The Fall of the Roman Empire; A New History of
Rome and the Barbarians. Oxford University Press.
Herlihy, David and Klapisch-Zuber Christiane, 1985 Tuscans and Their
Families: A Study of the Florentine Catasto of 1427. (New Haven,
I985).
Hirshman, Albert O 1970. Exit, Voice, and Loyalty: Responses to
Decline in Firms, Organizations, and States. Cambridge Mass. Harvard
University Press, 1970
Hoff, Karla and and Joseph E. Stiglitz, 2004. “After the Big Bang?
Obstacles to the Emergence of the Rule of Law in Post-Communist
Societies” The American Economic Review, Vol. 94, No. 3 (Jun., 2004),
pp. 753-763
Hoffman, Philip T. 1996. Growth in Traditional Society: The French
Countryside, 1450-1815. Princeton NJ, Princeton University Press.
Hoffman, Philip T.,and Kathryn Norberg eds. 1994. Fiscal Crises,
Liberty, and Representative Government, 1450-1789, Stanford.
Hoffman, Philip T., Gilles Postel-Vinay, and Jean-Laurent Rosenthal.
2000. Priceless Markets: The Political Economy of Credit in Paris,
1662-1869.
______________, 2008a. “The Old Economics of Information and the
Remarkable Persistence of Traditional Credit Markets in France
1740-1899”
______________, 2008b “History, Geography, and the Markets for
Mortgage Loans in 19th Century France.” Mimeo.
Hoffman, Philip T., and Jean-Laurent Rosenthal. 1997. “The Political
Economy of Warfare and Taxation in Early Modern Europe: Historical
Lessons for Economic Development.” In J. Droback and J. Nye eds.
Frontiers of Institutional Economics. Academic Press, 31-55.
Huang Jianhui 1994. Zhongguo yinhang ye shi (A history of the Chinese
banking industry). Taiyuan Shi : Shanxi jing ji chu ban she.
Huang Jianhui 2002. Shanxi piaohao shi (A history of the Shanxi native
banks). Taiyuan Shi : Shanxi jing ji chu ban she, 2002.
Huang, Philip. 1985. The Peasant Economy and Social Change in North
China, Stanford U.P.
Huang, Philip. 1990. The Peasant Family and Rural Development in the
Yangzi Delta, 1350-1988, Stanford U.P.
Huang, Philip. 1996. Civil justice in China : representation and
practice in the Qing. Stanford University Press.
Ihara, Hiroshi and Umemura Hiroshi. 1997. Sō to Chūo ūrashia (The Song
and central Eurasia) sekai no rekishi (world history) vol. 7.
Chuokoron-sha.
Jiangsu sheng changgongzhu di se ke. 1919. Jiangsu sheng shiye shicha
baogaoshu zongmuci (The complete Jiangsu province industrial survey
report). Commercial Press.
Jones, Eric Lionel. 1981. The European Miracle: Environments,
Economies, and Geopolitics in the History of Europe and Asia.
Cambridge University Press.
________. 1988. Growth Recurring : Economic Change in World History.
Oxford University Press.
Jones, James R, 1972. The Revolution of 1688 in England. New-York:
Norton.
Kertzer, David, and Marzio Barbagly; 2001-2. The History of the
European Family. New Haven: Yale University Press.
Kim, Sukkoo. 1995. "Expansion of Market and the Geographis
distribution of Activities: The Trends in U.S. Regional manufacturing
Structure, 1860-1987." Quarterly Journal of Economics 110(4)
1995:883-908.
Kirby, William. 1995. “China Unincorporated: Company Law and Business
Enterprise in Twentieth- Century China” Journal of Asian Studies 54.1
(Feb. 1995): 43-63.
Kishimoto Mio. 2007. “Land markets and land conflicts in late imperial
China,” Utrecht Conference on Law and Economic Development in
Historical Perspective.
Kohl, Benjamin. 1998. Padua Under the Carrara, 1318-1405. Johns
Hopkins University Press.
Kuran, Timur. 2004. “Why the Middle East Is Economically
Underdeveloped: Historical Mechanisms of Institutional Stagnation” The
Journal of Economic Perspectives, Vol. 18, No. 3 (Summer, 2004), pp.
71-90
Kuran, Timur. 2004. “The Islamic Commercial Crisis: Institutional
Roots of Economic Underdevelopment in the Middle East” The Journal of
Economic History, Vol. 63, No. 2 (Jun., 2003), pp. 414-446.
La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and
Robert W. Vishny. “Legal Determinants of External Finance.” Journal of
Finance, 52 (July 1997), pp. 1131-50.
_____. “Law and Finance.” Journal of Political Economy, 106 (Dec.
1998), pp. 1113-55.
Landa, Janet.1994. Trust, Ethnicity, and Identity: Beyond the New
Institutional Economics of Ethnic Trading Networks, Contract Law, and
Gift-exchange Ann Arbor: University of Michigan Press.
Landes, David S. 1979. “Watchmaking: A Case Study in Enterprise and
Change,” Business History Review LIII (1) 1-39.
____________. 1983. Revolution in Time; Clocks and the Making of the
Modern World. Cambridge Mass: Belknap.
_____________ 1998. The Wealth and Poverty of Nations: Why Some Are So
Rich and Some So Poor. W.W. Norton.
Lamoreaux, Naom R. and Jean-Laurent Rosenthal. 2005. “Legal Regime and
Business’s Organizational Choice: A Comparison of France and the
United States during the Mid-Nineteenth Century.” With American Law
and Economic Review, Vol. 7 (1), 28-61.
Lavely, William, and R. Bin Wong. 1998. “Revising the Malthusian
Narrative: The Comparative Study of Population Dynamics in Late
Imperial China.” The Journal of Asian Studies 57,3 (August):714-748.
Lavely, William and R. Bin Wong. 1992. "Family Division and Mobility
in North China", Comparative Studies in Society and History.
Levi, Margaret. 1989. Of Rule and Revenue. Berkeley: University of
California Press, 1989
Le Roy Ladurie, Emmanuel, 1966. Les Paysans du Languedoc. 2 vols.
Paris SEVEPEN.
Lee, James, Cameron Campbell and Tan Guofu. 1992. "Infanticide and
family planning in rural Liaoning, 1774-1873." in Li, Lillian and
Thomas Rawski, eds. Chinese History in Economic Perspective. Berkeley:
University of California Press, 149-176.
Lee, James and Cameron Campbell. 1997. Fate and Fortune in Rural
China: Social Organization and Population Behavior in Liaoning,
1774-1873. Cambridge University Press.
Lee, James and Wang Feng, 2001. One Quarter of Humanity: Malthusian
Mythology and Chinese Realities, 1700-2000. Harvard University Press.
Lee James and R. Bin Wong. 1991. ‘Population movements in Qing China
and their linguistic legacy.’ In Languages and Dialects of China.
William Wong ed., Journal of Chinese Linguistics Monograph Series No.
3. Berkeley 1991: 52-77.
Li, Bozhong. 1998. Agricultural Development in the Yangzi Delta,
1600-1850. MacMillan.
__________, 2003. Duo shijiao kan Jiangnan jingji shi (Different
perspectives on the economic history of Jiangnan) Sanlian. 2003. Pp
137-76; 224-40.
Liang Zhiping. 1996. Qingdai xiguanfa: shehui yu guojia (Customary law
in the Qing dynasty: society and the state). Zhongguo zhengfa daxue
chubanshe.
Lindert. Peter H. 2004. Growing Public; Social Spending and Economic
Growth Since the Eighteenth Century. Cambridge: Cambridge University
Press.
Luckett, T. 1992. “Credit and Commercial Society in France,
1740-1789.” Ph.D. dissertation, Princeton University.
Lufrano Richard.1997. Honorable Merchants: Commerce and
Self-cultivation in Late Imperial China. University of Hawaii Press.
Lynch, John. 1989 Spain 1516-1598. From Nation State to World Empire.
London Blackwell.
Ma Debin 2006. “Law and Commerce in Traditional China, an
Institutional Perspective on the ‘Great Divergence’” Keizai-Shirin,
vol. 73, No. 4, March 2006.
Macauley, Melissa Ann 1998. Social Power and Legal Culture: Litigation
Masters in Late Imperial China Stanford: Stanford University Press.
Magnac, Thierry, and Gilles Postel-Vinay, 1997. "Wage Competition
between Agriculture and Industry in Mid-Nineteenth Century France,"
Explorations in Economic History, vol. 34(1), pages 1-26.
Malthus, Thomas R. 1992 [1806]. An Essay on the Principle of
Population; or a View of its Past and Present Effects on Human
Happiness. Cambridge: Cambridge University Press.
Mamoru Tonami and Takeda Yukio. Eds. 1997. Zuitō teikoku to kodai
chōsen (Sui Tang empires and ancient Korea) sekai no rekishi (world
history) vol 6. Chuokoron-sha
Mathias, Peter and Patrick O'Brien, ‘Taxation in Britain and France,
1715 - 1810: A Comparison of the Social and Economic Incidence of
Taxes Collected for the Central Governments,’ The Journal of European
Economic History, 5 (Winter 1976), 601-50.
McCloskey, Donald. 1975. “The Persistence of English Common Fields.”
in W. Parker and E. Jones eds. European Peasants and Their Markets.
Princeton University Press, pp. 73-114.
________________ 1976. "English Open Fields as Behavior Towards Risk,"
Research in Economic History. 1:124-170.
McGuire, Martin, and Olson Mancur. 1996. “The Economics of Autocracy
and Majority Rule: The Invisible Hand and the Use of Force.” Journal
of Economic Literature 34(1) pages 72-96.
McNeill, William H. 1976. Plagues and peoples Garden City, N.Y. :
Anchor Press.
________. 1982. The pursuit of power: technology, armed force, and
society since A.D. 1000. Chicago : University of Chicago Press.
Macauley, Melissa. 1998. Social Power and Legal Culture: Litigation
Masters in Late Imperial China. Stanford University Press.
Mann, Susan. 1987. Local Merchants and the Chinese Bureaucracy,
1750-1950. Stanford: Stanford University Press.
_________, 1992. “Household Handicrafts and State Policy in Qing
Times.” In Jane Leonard, and John Watt, eds, To Achieve Security and
Wealth. Cornell University Press.
Meskill, Joanna. 1979. A Chinese Pioneer Family: The Lins of Wu-Feng
Taiwan, 1729—1895. Princeton, NJ: Princeton University Press.
Michaud, Claude. 1991. L’Eglise et l'argent sous l'Ancien Régime: les
receveurs généraux du clergé de France aux XVIe et XVIIe siècles.
Paris Fayard.
Michie, R.C., 1999. The London Stock Exchange: A History. Oxford
University Press, Oxford
Mokyr, Joel. 1990. The lever of riches: technological creativity and
economic progress. New York : Oxford University Press.
________. 2002. The gifts of Athena : historical origins of the
knowledge economy. Princeton, N.J. : Princeton University Press.
Moky, Joel (ed.) The Economics of the Industrial Revolution. Totowa,
N.J.: Rowman and Allanheld, 1985. Reprinted 1989 by Rowman and
Littlefield.
Mori Masao. 1969. “Jūroku-jūhachi seiki ni okeru kōsei to jinushi
denko kankei” (Famine relief and landlord-tenant relations from the
sixteenth to eighteenth centuries). Tōyōshi kenkyū 27.4: 69-111.
Morris-Suzuki, Tessa. 1994. The Technological Transformation of Japan,
Cambridge University Press.
Muller, Reinhold C,.and Frederic Chapin Lane. 1997. The Venetian Money
Market: Banks, Panics, and the Public Debt, 1200-1500. Baltimore:
Johns Hopkins University Press.
Munro John, 2005. “Spanish Merino Wools and the Nouvelles Draperies:
An Industrial Transformation in the Late Medieval Low Countries.”
Economic History Review, Vol. 58, No. 3, pp. 431-484, August 2005
Needham, Joseph, ed. 1954. Science and Civilization in China.
Cambridge: Cambridge University Press.
Ng Chin keong. 1983. Trade and Society: The Amoy Network on the China
Coast 1683-1735 Singapore: Singapore University Press.
Nishijima Sadao. 1984. "The Formation of the Early Chinese Cotton
Industry," in Linda Grove and Christian Daniels, eds., State and
Society in China: Japanese Perspectives on Ming-Qing Social and
Economic History. University of Tokyo Press.
Nishijima Sadao. 1999. “The economic and social history of the former
Han,” in Michael Loewe, ed. The Cambridge History of China, vol. 1,
PAGE Cambridge University Press
Niu Guanjie. 2008. 17-19 shiji Zhongguo de shichang yu jingji fazhan
(17th-19th century Chinese markets and economic development).
Huangshan shushe.
Neal, Larry, 1993. The Rise of Financial Capitalism: International
Capital Markets in the Age of Reason. Cambridge: Cambridge University
Press.
Neal Larry, "The Finance of Business during the Industrial
Revolution," in Roderick Floud and Donald N. McCloskey, eds., The
Economic History of Great Britain, vol. 1, "1700-1860," 2nd ed.,
Cambridge: Cambridge University Press, 1994.
Neal, Larry and Stephen Quinn. 2003, “Markets and Institutions in the
Rise of London as a Financial center in the Seventeenth Century” in S.
Engerman et al eds Finance, Intermediaries, and Economic Development.
Cambridge: Cambridge University Press.
Nef, John. 1950. War and human progress: an essay on the rise of
industrial civilization. Cambridge Mass: Harvard University Press,
North, Douglass C. 1981. Structure and Change in Economic History. NY:
Norton.
________. 1990. Institutions, Institutional Change, and Economic
Performance. Cambridge: Cambridge University Press.
North, Douglass C, and Robert P Thomas.1971. :The Rise and Fall of the
Manorial System: A Theoretical Model” Journal of Economic History,
1971
North, Douglass C, and Barry R. Weingast 1989. "Constitutions and
Commitment: Evolution of the Institutions Governing Public Choice in
Seventeenth-Century England." Journal of Economic History 49: 803-32.
North, Douglass C., John Joseph Wallis, and Barry R. Weingast. 2009.
Violence and Social Orders: A Conceptual Framework for Interpreting
Recorded Human History. NY Cambridge University Press, 2009
Ogilvie, Sheilagh, and Markus Cernan. 1996. European
Proto-Industrialization. Cambridge: Cambridge University Press.
Ostrogorsky, George. 2002. History of the Byzantine State. Rutgers
University Press.
Pan, Ming-te. 1996. "Rural credit and the concept of 'peasant petty
commodity production' in Ming-Qing Jiangnan" The Journal of Asian
Studies. 55:1.
Patlagean, Evelyne. 2007. Un Moyen Age grec : Byzance, 9e-15e siècle.
in L'evolution de l'humanité Paris: Albin Michel.
Pezzolo, Lucianno. 2005. “bond and Government debt in Italian City
States, 1250-1650. In William N. Goetzmann, K. Geert Rouwenhorst eds.
Origins of Value: A Document History of Finance. Oxford: Oxford
University Press US, 2005
Pomeranz, Kenneth. 1997. “Traditional Chinese Business Forms
Revisited: Family, Firm, and Financing in the History of the Yutang
Company of Jining, 1756-1956. Late Imperial China. Vol. 18 (1) 1-38.
_________, 2000. The Great Divergence: Europe, China, and the Making
of the Modern World Economy. Princeton, N.J. : Princeton University
Press.
Postel-Vinay, Gilles. 1994. "The Disintegration of Labour Markets in
Nineteenth Century France." In G. Grantham and M. MacKinnon, eds.
Labour Market Evolution. Routledge.
Potter, David. 2004. The Roman Empire at Bay AD. London: Routledge.
Potter, Mark, and Jean-Laurent Rosenthal. 1997. "Politics and Public
Finance in France: The Estates of Burgundy, 1660-1790." Journal of
Interdisciplinary History. 27(4): 577-612.
Pfister, Ulrich. 1994. ”Le petit crédit rural en Suisse au XVIe-XVIIIe
siècles”, Annales: Histoire Sciences Sociales, (6): 1339-58.
Quinn, Stephen. 2004. Money, Finance and Capital Markets,” in
Cambridge Economic History of Britain since1700 Volume I, edited by
Floud and Johnson, Cambridge University Press (2004): 147-74.
Rankin, Mary. 1986. Elite Activism and Political Transformation in
China. Stanford University Press
Rawski, Thomas. 1989. Economic Growth in Prewar China. Berkeley:
University of California Press.
Reddy, William M. 1984 The Eise of market culture: the textile trade
and French society, 1750-1900. New York : Cambridge University Press.
Ricardo, David. 1973. [1793] The Principle of Political Economy and
Taxation. London: J.M. Dent and Sons ltd.
Rossabi, Morris. 1988, Khubulai Khan, his Life and time. Berkeley,
University of California Press.
Rosenthal, Jean-Laurent. 1998. “The Political Economy of Absolutism
Reconsidered.” In R. Bates et al eds. Analytic Narratives. Princeton
University Press, 1998, 63-108.
Ruskola, Teemu. 2000. “Conceptualizing Corporations and Kinship:
Comparative Law and Development Theory in a Chinese Perspective,” 52
STANFORD LAW REVIEW 1599 (2000).
Sabean, David. 1998. Kinship in Neckarhausen, 1700-1780. Cambridge:
Cambridge University Press.
Scheidel, Walter, ed. 2009.Rome and China; Comparative Perspectives on
Ancient World Empires. Oxford: Oxford University Press.
Schnapper, Bernard. 1957. Les rentes au XVIe siècle: Histoire d'un
instrument de crédit. Paris: SEVPEN.
Sen, Amartya. 1981. Poverty and Famines: an essay on Entitlement and
Deprivation. Oxford : Clarendon Press,
Servais, Paul. 1982. La rente constituée dans le Ban de Herve au
XVIIIe siècle, Bruxelles, Crédit Communal de Belgique, Collection
Histoire, n° 62.
Shiba Yoshinobu. 1970. Commerce and Society in Sung China. Ann Arbor:
University of Michigan Press.
_________, 1977. “Ningpo and its Hinterland” in G William Skinner,
ed., The City in Late Imperial China. Stanford University Press.
Shiga Shuzo. 2002. Shindai Chūgoku ho to saiban (Law and Adjudication
in Qing China). 2nd edition. Soubunsha.
Skinner, G. William 1977a. “Regional Urbanization in
Nineteenth-Century China,” in Skinner, ed. 1977. The City in Late
Imperial China.
________. 1977b. “Cities and the Hierarchy of Local Systems,” in
Skinner, ed. 1977. The City in Late Imperial China.
Smith, Adam. 1776 (1976), An Inquiry Into the Nature and Causes of the
Wealth of Nations. Chicago, U of Chicago Press.
Scheidel, Walter. Ed.2009. Rome and China; Comparative Perspectives on
Ancient World Empires. Oxford University Press.
Scoville, Warren, 1960. The persecution of Huguenots and French
economic development, 1680-1720. Published/distributed: Berkeley :
University of California Press, 1960
Sokoloff, Kenneth, and David Dollar. 1997. "Agricultural Seasonality
and the Organization of Manufacturing in Early Industrial Economies:
The Contrast Between England and the United States." Journal of
Economic History 57.2: 288-321.
Summerhill, William. Inglorious Revolution: Political Institutions,
Sovereign Debt, and Financial Underdevelopment in Brazil. Forthcoming
Yale University Press
Tabellini. Guido 2008. 2009. “The Scope of Cooperation: Values and
Incentives,” forthcoming Quarterly Journal of Economics.
Tanaka Masatoshi. 1984. "Rural Handicraft in Jiangnan in the Sixteenth
and Seventeenth Centuries," in Linda Grove and Christian Daniels,
eds., State and Society in China: Japanese Perspectives on Ming-Qing
Social and Economic History. University of Tokyo Press.
Tang Xianglong. 1987. Zhongguo jindai caizheng jingjishi lunwen xuan
(Selected essays on modern Chinese fiscal and economic history). Xinan
caijing daxue chubanshe.
Tawney, R. H. 1966. Land and Labor in China. Beacon Press.
Terada Takanobu. 1972. Sansei shōnin no kenkyū (An analysis of Shanxi
merchants) Dohosha.
Tilly Charles, 1984. Big Structures, Large Processes, Huge Comparisons
New York: Russell Sage Foundation, 1984.
________, 1990. Coercion, capital, and European states, AD 990-1990Cambridge,
Mass.: B. Blackwell, 1990.
Tracy, James D. 1985. A Financial Revolution in the Habsburg
Netherlands. Berkeley: California UP.
Twitchett, Denis and Michael Loewe. Eds. 1987. The Cambridge History
of China. vol. 1 The Ch’in and Han Empires. Cambridge University
Press.
Twitchett, Denis. Ed. 1979. The Cambridge History of China. vol. 3.
Sui and T'ang China, 589–906 AD, Part One. Cambridge University Press.
Van Dyke, Paul, 2005. The Canton Trade: Life and Enterprise on the
China Coast, 1700–1845. Hong Kong: Hong Kong University Press, 2005.
Van der Wee, Herman. 1988. “Industry Dynamics and the Process of
Urbanization De-Urbanization in the Low Countries from the Late Middle
Ages to Eighteenth Century, A Century. in Van der Wee, ed. The Rise
And Decline of Cities in Italy and the Netherlands. University of
Leuwen Press.
Van der Wee, Herman, and Monique Verbreyt. 1997. The General Bank,
1822-1997: A continuing Challenge. Tielt.
Van Zanden, Jan Luiten. 2007. The skill premium and the ‘Great
Divergence. Unpublished Manuscript Utrecht.
Van Zanden, Jan Luiten, and Arthur van Riel. 2004. The Strictures of
Inheritance: The Dutch Economy in the Nineteenth Century. Princeton:
Princeton University Press.
Vardi, Liana. 1993. The Land and the Loom; Peasants and Profif in
Northern France 1680-1800. Duke University Press.
Velde, François, and David Weir. 1992. "The Financial Market and
Government Debt Policy in France, 1746-1793." Journal of Economic
History 52: 1-40.
Veyne, Paul. 1976. Le pain et le cirque: sociologie historique d'un
pluralisme politique. Paris : Seuil.
Vogel, Ezra. 1979. Japan as Number One. Harvard University Press.
Vogt. Joseph, The Decline of Rome. New York New American Library 1965.
Wallis, John J. 2000. “American Government Finance in the Long Run:
1790 to 1990,” Journal of Economic Perspectives.
Wallerstein, Immanuel, 1974-1989. The Modern World-System. 3 volumes
New York : Academic Press.
Wang Yeh-chien. 1973. Land Taxation in Imperial China, 1750-1911.
Harvard University Press.
Weber, Max, 1930. The Protestant ethic and the spirit of capitalism,
translated by Talcott Parsons, London, G. Allen & Unwin.
Wei Guanqi 1986. “Qingdai houqi zhongyang jiquan caizheng tizhi de
wajie” (The collapse of the fiscal system of central authority in the
late Qing). Jindaishi yanjiu 1:207-30.
Weir, David R. 1984. “Life Under Pressure: France and England,
1670-1870,” The Journal of Economic History, Vol. 44, No. 1 (Mar.,
1984), pp. 27-47.
Veyne, Paul. 1976. Le Pain et le Cirque; Sociologie Historique d’un
Pluralisme Politique. Paris: Seuil.
Will, Pierre-Etienne and R. Bin Wong. 1991. Nourish the People: The
State Civilian Granary System in China, 1650-1850. University of
Michigan Center for Chinese Studies.
Wittfogel, Karl. 1957. Oriental despotism; a comparative study of
total power. New Haven : Yale University Press.
Wong, R. Bin. 1997. China Transformed: Historical Change and the
Limits of European Experience. Cornell University Press.
_____. 1999. “The Political Economy of Agrarian Empire and Its Modern
Legacy,” in Timothy Brooks and Gregory Blue, eds., China and Imperial
Capitalism: Genealogies of Sinological Knowledge Cambridge: Cambridge
Univ. Press, pp. 210-245.
_____. 2004. “Relationships between the Political Economies of
Maritime and Agrarian China, 1750-1850,” in Wang Gungwu and Ng
Chin-Keong eds., Maritime China and the Overseas Chinese Communities,
Wiesbaden (Germany): Harrassowitz Verlag.
_____. 2007. “Les politiques de dépenses sociales avant or sans démocratie,”
Annales,HSS (novembre-décembre 2007) 62.6: 1405-1416.
_____. 2009. “Taxation and Good Governance in China,” in Bartolome Yun
Casalilla et al (eds.), Fiscal States in Europe and Asia (Cambridge,
2009)
Wrigley E. Anthony. 1967. “A Simple Model of London's Importance in
Changing English Society and Economy 1650-1750,” Past and Present, No.
37, pp. 44-70
_______________, 1985. "Urban Growth and Agricultural Change: England
and the Continent in the Early Modern Period," Journal of
Interdisciplinary History
_______________, 1988. Continuity Chance and Change; the Character of
the Industrial Revolution in England. Cambridge: Cambridge University
Press.
_______________,, and Roger Schofield, 1981. The Population History of
England. Cambridge Mass. Harvard University Press.
Xu Dixin and Wu Chengming, eds. 2000. Chinese Capitalism, 1522 1840.
New York: St. Martin's Press
Yan Se, 2008. Essays on Real Wages and Wage Inequality in China, 1858
to 1936. PhD Dissertation UCLA.
Yang, Li Mu. 2002. Essays on Public Finance and Economic Development
in a Historical Institutional Perspective. Ph.D. Dissertation Stanford
University.
Ye Shichang and Pan Liangui. 2004.Zhongguo gujindai jinrongshi (A
history of ancient and modern Chinese banking). Fudan University
Press.
Zelin, Madeleine 2005. The Merchants of Zigong: Industrial
Entrepreneurship in Early Modern China. New York: Columbia University
Press.
Zelin Madeleine, Jonathan K. Ocko and Robert Gardella, eds. 2004.
Contract and Property in Early Modern China. Stanford: Stanford
University Press.
Zelin, Madeleine. 2004. “Economic Freedom in Late Imperial China,” in
William Kirby, ed., Realms of Freedom in Modern China, Stanford
University Press, 57-83.
Zelin 1990. “The Rise and Fall of the Furong Well-Salt Elite,” in
Joseph Esherick and Mary B. Rankin, eds., Chinese Local Elites and
Patterns of Dominance. University of California Press.
Zheng Changgan. 1989. Ming Qing nongcun shangpin jingji (Ming Qing
agrarian commercial economy). Zhongguo renmin daxue chubanshe.
Zhou Yumin. 2000. Wan Qing caizheng yu shehui bianqian (Late Qing
fiscal administration and social change). Shanghai renmin chubanshe
Zhou Zhichu. 2002. Wan Qing caizheng jingji yanjiu (Research on late
Qing fiscal administration). Jilu shushe.
1 These assumptions make about as much sense as one that would assume
that marginal productivities are equal for all households at all
times.
2 The analogy come from flipping coins an even number of time and
examining the share of heads. A the number of flips increases, the
share of heads gets concentrated around ½ while at the same time the
likelihood of a sequence that has exactly half heads goes to zero. The
first effect drives the shrinking labor market while the second drives
the increasing share of households in the labor market.
3 (with n adults The number of different combinations of E,W is 2n,,
consider households that have me E adults and mw W adults (n=me + mw).let
m= min of (me , mw ) ; then the share of households that have me E
adults and mw W adults is (n!/ 2n m!).
4 Technically we are looking at a stationary repeated equilibrium
(where no one ever cheats). The condition under which network members
exclude cheaters have been explored at length see Greif (2006)
5 A factor share is the ratio of expenditure on one factor to total
expenditure. Thus the factor share for labor is wL/(wL+rK) while the
factor proportion is simply L/K.
6 Obviously war raises costs everywhere, and cities’ walls can protect
both capital and labor. The specification above that only considers
the effects of war on the relative price of capital thus understates
the extent of the bias towards urban manufacturing.
7 The one mysterious period is the long hiatus to the empire under the
Sung. For some three centuries, the Chinese mainland was divided into
competing regimes. A European at least would have surmised that either
the Sung or its rivals would have developed credit institutions in a
gambit to reunify the empire.
8 After that time, some states launched into road building efforts to
move their troops; the same roads were also useful in speeding up
transport. See Arbellot (19xx)
9 There are exceptions of course such as the French State investment
in royal roads in the 18th century. It is worthwhile to note, however,
that these roads had important strategic value and that this
investment came late in the pre-industrial period.
10 Richard Bonney (2007) puts the cost of Versailles at 92 million
livres, That was less than 2% of tax revenues over Louis’s half
century reign.
11 Crucial Manchu innovations took place in terms of the dynasty’s
relations to other groups along the empire’s northern frontiers—with
different Mongol, Uighur and Tibetan groups especially. But these
important changes, the subject of much recent and current research in
Qing history, concern areas that are not the sites of economic
practices with which we have been concerned in this book.
12 We concentrate on the European empire because it is most relevant
to the issues in this book. We return briefly to the colonial empires
in the last part of the Chapter.
13 As in all things in Europe there are exceptions. Notably the
Austro-Hungarian Empire was immune to the reform epidemic. Its various
nineteenth-century guises emphasized the institutional distinctiveness
of its different components (down to the emperor of Austria separately
ruling as the king of Hungary).
0