Application Of Multiproject Baseline Methods In Practice By

application of multi-project baseline methods in practice --------------------------------------------------------- by wytze van der gaast

Application of Multi-project Baseline Methods in Practice
---------------------------------------------------------
by Wytze van der Gaast, Foundation JIN
Contact: tel. 0031 50 3096815; e-mail: [email protected]
=====================================================
Introduction
A JI or CDM project is a GHG-neutral investment. A country with an
assigned amount of GHG emissions under the Kyoto Protocol could
increase this amount by reducing GHG emissions through a project on
the territory of another country. If transaction costs are assumed to
be zero and if it is assumed that baselines at all times correctly
represent the situation in absence of the project, the allocation of
projects takes place on the basis of an international comparison of
marginal abatement costs only. However, the theoretically optimal
allocation of projects could be disturbed in the real world where
transaction costs exist and the baseline is surrounded by
uncertainties.
With transaction costs, the marginal cost curves for JI/CDM emission
reductions (serving as the supply curves) shift upwards as for each
quantity supplied the supply price (excluding transaction costs) is
increased by the transaction costs (e.g., project design costs).
Typical transaction costs for JI/CDM projects (the GHG emission
reduction component) are costs related to: identifying the project;
obtaining approval of the project idea by the host country government;
compiling the project design document; negotiating the emission
reduction purchase agreement (ERPA) with the host country government
and relevant institutes; validating the project design; and,
registering a project with the JI and CDM supervisory bodies.
In a recent overview based on actual CDM projects (excluding
small-scale projects), Ellis et al. (2004)1 found that the CDM-related
transaction costs could vary from US$ 50,000 per project to almost US$
270,000. According to Ellis et al. (2004), even the lower cost figure
implies that a project would need around 10,000 tonnes emission
reduction credits over a crediting lifetime to cover transaction costs
(assuming a credit price of US$ 5). Moreover, several programmes
established to generate CDM credits, such as the Rabobank Carbon Fund,
the KfW Bankengruppe Fund, and projects established under the CDM
programmes of the Netherlands Government (e.g., CERUPT), have, in
order to be able to cover transaction costs, set a minimum to the
amount of Certified Emission Reductions (CERs) that proposed CDM
projects must deliver during the crediting lifetime (in most cases
500,000 tonnes). Finally, the intermediate evaluation of the Dutch JI
policy in 2005 showed that the transaction costs related to the
purchase of JI credits by the Netherlands Government amounted to
almost 25% of the nominal prices paid for these credits.2
The inclusion of a large set of modalities and procedures for
baselines in Decision -/CMP.1 of the Conference of the Parties serving
as the meeting of the Kyoto Protocol Parties (COP/MOP-1), as well as
the specific additional rules set by the programmes of individual
governments and multilateral funds, has resulted in a rather large
share of baseline-related costs in the overall transaction costs of
the project design. For example, estimates in Krey (2003) show that,
depending on the complexity of the project and its baseline,
transaction costs related to baseline determination have thus far
amounted to between US$ 20,000 and US$ 25,000 per CDM project, which
is higher than search, monitoring and validation costs, but lower the
negotiation and project approval costs.
High transaction costs as a result of detailed baseline rules or the
overestimation of emission reductions due to inflated baselines in
case of too lenient rules results in an allocation of JI and CDM
projects which differs from an allocation based on an international
comparison of marginal abatement costs only. With high transaction
costs it becomes less attractive for investors to participate in JI
and CDM projects so that the Kyoto Protocol compliance costs for
industrialised countries increase. On the other hand, due to baseline
inflation JI/CDM projects are no longer GHG-neutral and do not
contribute – or at least to a smaller extent – to reducing GHG
emissions globally.
With a view to the above, several literature sources have recommended
standardising baseline determination for JI and CDM projects.3 First,
with multi-project baselines costs related to compiling the project
design document and the validation of the baseline could be
considerably reduced. Second, with a multi-project baseline
methodology, the scope for baseline inflation by project developers is
much reduced, especially when the baseline emission factors are
standardised.
A possible disadvantage of standardised, multi-project baselines is
that they, in their turn, could also affect the allocation of
projects. An example may clarify this. Suppose that in a particular
country energy is produced by both coal and gas-fired plants. In this
country, a JI project aims at installing a modern gas boiler which
emits fewer GHG per unit of output than both the existing gas and coal
plants. Should for the power sector in this country a multi-project
baseline be determined based on the average emissions of the existing
plants, its emissions would be higher than those of the gas-fired
plants and lower than those of the coal-fired plants. Consequently, a
JI project investor who replaces a coal plant by a modern gas boiler
only receives credits for the emission reductions below the
multi-project baseline and not for the full reductions achieved.
Should the investor instead replace an existing gas boiler, the amount
of credits would be larger than the emission reductions actually
achieved. This could reduce the incentives for investors to invest in
coal plants. In order to prevent such an adverse incentive,
multi-project baselines could be determined per fuel or fuel
technology; for example, one multi-project baseline covering the
existing coal plants in the country and one for the gas boilers.
Also the choice of the region covered by the multi-project baseline is
important for the extent to which the baseline can be considered
reasonable. In several JI countries differences exist between regions
within potential JI host countries in terms of industrialisation,
energy intensity and GHG intensity of production and consumption.
Determining national average multi-project baselines for these
countries could imply that these would not be representative for each
of the regions. Therefore, it is important that a multi-project
baseline is sufficiently representative for the region where the
project is implemented.
An important standardisation of baselines and baseline methodologies
has been the approval of baseline methodologies by the CDM Executive
Board (CDM EB). Once approved a methodology can be used for a multiple
of other projects of the same type. In the actual practice of the CDM,
this has led to a number of different methodologies approved for one
project category. A next step towards standardisation is that these
methodologies are somehow merged into one consolidated methodology per
project type. Each entity that intends to invest in a project of that
type could subsequently apply the consolidated methodology for the
project and ‘only’ needs to fill in the project or country-specific
data in order to determine the baseline, e.g., project activity level,
project-specific weighting factors (if applicable), and specific data
for the host country.4
A considerable step further in the direction of baseline
standardisation is the calculation of multi-project GHG emission
reduction factors (i.e. benchmarks) for a particular project type in a
particular host country. This type of standardisation is most
far-reaching as it reduces baseline determination to an effort of
multiplying the project activity level within the project boundary (e.g.,
number of kWh per year) with the benchmark emissions factor. An early
and, thus far, only example of such a benchmark application can be
found in the Dutch ERUPT programme for JI projects, which, as of its
second round in 2001, offers a table with benchmark CO2 emission
factors for electricity sector projects in each Central and Eastern
European country. The process of standardising baselines and baseline
procedures has thus far been primarily a bottom up exercise with
standards being developed based on a number of individual CDM project
baselines.5
Scope for multi-project baselines for JI projects in the ‘Marrakech
Accords’
Further to the Marrakech Accords, the COP/MOP-1 has provided clear
references to standardisation of baselines for JI projects. For
example, Decision -/CMP.1 on JI (‘Guidelines for the implementation of
Article 6 of the Kyoto Protocol’, Appendix B) states that a baseline
shall be established “[o]n a project-specific basis and/or using a
multi-project emission factor.” Next to a clear reference to
standardisation of baselines, it also indicates that applying a
project-specific baseline methodology does not exclude standardisation
of certain baseline parameters.
A strong reference to developing multi-project/standardised procedures
for CDM baseline methodologies can be found in Appendix C to the Draft
decision -/CMP.1 (Article 12). The Appendix states that the EB shall
provide guidance on the “appropriate level of standardization of
methodologies to allow a reasonable estimation of what would have
occurred in the absence of a project activity wherever possible and
appropriate,”6 which is a clear reference to multi-project baselines.
In addition, the EB shall develop and recommend to the COP-MOP
specific guidance on the “[d]efinition of project categories (e.g.,
based on sector, sub sector, project type, technology, geographic
area) that show common methodological characteristics for baseline
setting.”7 Finally, COP/MOP-1 adopted the decision that the CDM-EB
shall explore the possibility of using “[d]ecision trees and other
methodological tools, where appropriate, to guide choices in order to
ensure that the most appropriate methodologies are selected, taking
into account relevant circumstances.”8
Early experiences with baseline standardisation: the Dutch JI tenders
In 2000, the Netherlands Ministry of Economic Affairs launched a
tender programme for JI projects, called ERUPT (Emission Reduction
Units Procurement Tender), which is managed by the government agency
SenterNovem. Together with the World Bank’s Prototype Carbon Fund
(PCF), which started more or less at the same time and in which the
Netherlands Government is also an investor, ERUPT faced the challenge
of translating the Kyoto Protocol into detailed modalities and
procedures for the project cycle of JI (note that the Netherlands
initiated ERUPT about a year and a half before launching a similar
tender for the CDM, called CERUPT), without the guidance of the
Marrakech Accords, which became available only in November 2001. For
this purpose, a detailed Guidelines document was prepared, which
explained project developers how to determine a baseline and how to
deal with other JI accounting issues such as additionality and
leakage. After the first ERUPT round, these Guidelines were reviewed
based on the experiences of project developers and auditors who had to
validate the project design documents.9
The ERUPT Guidelines document and its revised version were an
important step in the direction of standardising procedures for
baseline determination because all project developers proposing
projects to ERUPT had to use the same procedure. Next to the
standardised key-factor approach, the revised ERUPT Guidelines also
standardised particular parameters of baseline determination, such as
the definition of the project boundary, i.e. which emission sources to
include in the baseline, and defining the additionality concept. In
addition, the revised Guidelines contained multi-project
baseline/benchmark values for the electricity sectors in potential JI
host countries in Central and Eastern Europe. The use of these
benchmarks under ERUPT has been optional; project developers who feel
that the benchmark value do not reasonably represent their specific
project baseline case, have been free to develop their own baseline
emission factors, provided that they use the key-factor approach.
When calculating these benchmarks, the following key issues had to be
dealt with:
*
How to deal with low variable cost power capacity? Since
power plants that operate at very low variable costs are
usually operated as many hours as possible, they are very
unlikely to be dispatched when new power production capacity
becomes available. Examples of such capacity are
run-of-river hydropower, co-generation and nuclear power
plants.
*
How to incorporate the EU pre-Accession process of Central and
Eastern European countries into the benchmark calculation?
Candidate members of the EU have to incorporate EU standards
(collected in the Acquis Communautaire) in their domestic laws,
but by the time of calculating the ERUPT benchmarks (2001),
negotiations between the European Commission and the countries
that were scheduled to become EU member state in May 2004 had not
yet clearly revealed to what extent these countries would be
allowed a transition period to postpone incorporating the
standards until after the accession date (countries such as
Bulgaria and Romania had not yet begun pre-accession negotiations
with EU).
Baseline standardisation under the CDM
As per February 2006, the CDM-EB has approved 27 baseline and
monitoring methodologies for CDM projects. With an increasing number
of approved methodologies, there could be a considerable variation in
how emission baselines are determined for similar projects. This trend
was observed in Ellis (2003) who categorised the baselines used in
‘operating margin’ (the effect of a project on the operation of power
plants on the grid), ‘built margin’ (the effect of a project in terms
of delaying or avoiding the construction of future power plants), and
‘combined margin’ types of baselines (a combination of operating and
build margin).10 However, even within these baseline categories
different methods had been used. For example, some methodologies
identify within a particular host country a currently operating plant
that would have been dispatched in case of newly added capacity. Other
methods, instead, apply an operating margin method by taking a
weighted average emission factor for the grid as a whole.
In order to harmonise methodologies per project category, the EB has
consolidated methodologies per category, eight in total. These
consolidated methodologies would function as standardised baseline
methodologies and could be used by all project developers proposing a
project in the category concerned.
Contribution of multi-project baselines to practicality and
environmental integrity
Baseline standardisation does not necessarily lead to a trade-off
between environmental integrity and scope for application. On the one
hand, standardisation, in order to reduce transaction costs related to
the project design, would be more aggregate and could be less precise
than individual project baselines, which could reduce the
environmental integrity of JI projects. On the other hand,
standardisation reduces the scope for overestimating emission
reductions from projects by project developers and enables projects in
sectors where only policy-based projects are feasible (e.g.,
transport, built environment) or where the capacity replaced by a
project cannot be clearly identified (e.g., greenfield, grid-connected
renewable energy projects). Application of multi-project baselines to
these sectors and project types would broaden the scope for JI and the
CDM, which would enhance environmental integrity. Finally,
multi-project baselines can provide a simplified alternative for host
countries for which data availability and/or data quality are
problematic for single-project baseline determination.
Organisation of multi-project baselines under JI
Baseline standardisation through templates (standardised procedures)
or multi-project emission factors (benchmarks) have thus far taken
place at the initiative of project developers submitting project
proposals to the EB and of governments through tender programmes (e.g.,
ERUPT, CERUPT, PCF). The role of the COP and its bodies has thus far
been mainly a passive one. The COP had set boundary conditions for
baseline determination, including the notion that multi-project
approaches are eligible under JI and CDM, provided that they are
sufficiently representative for the project-specific case.
Given the benefits from baseline standardisation in terms of
increasing environmental integrity, reducing project-specific
transaction costs and increasing transparency of procedures, it is
recommended that future procedures for standardisation of JI baselines
take place in a top-down context instead of the present bottom-up
practice of the CDM. The latter has proven to be rather time-consuming
and sub-optimal in the sense that the determination of a
standardised/multi-project baseline in itself may require more time
and data than one project-specific baseline, as it generally requires
a macro and/or sector analysis. The JI/CDM practice has shown that
project developers are generally reluctant to develop multi-project
baseline methodologies, because once approved, other investors can
freely use these methodologies without compensating the initial
developers.11
It is recommended that the determination of multi-project baselines
(benchmarks) is managed under the auspices of the JI-SC and carried
out by independent experts who determine benchmarks on behalf of JI-SC
and update the scenarios.
Finally, it is instructive to observe that multi-project baselines
meet the JI baseline criteria set by the COP/MOP-1 in that:
a.
They could both be standardised procedures with largely
project-specific data and/or contain multi-project emission
factors per unit of activity level; both these applications are
allowed under the COP/MOP-1 decision on JI baselines.
b.
Standardised/multi-project baseline methodologies would increase
transparency with regard to the choice of approaches, assumptions,
methodologies, parameters, data sources and key factors to a level
beyond the transparency of project-specific baselines;
c.
Multi-project baseline parameters generally cover a sector and/or
country/region as a whole, which implies that they have take into
account the host country context, relevant national and/or
sectoral policies and circumstances, such as sectoral reform
initiatives, local fuel availability, power sector expansion
plans, and the economic situation in the project sector;
d.
Since the multi-project baselines would be expressed in emission
factors per unit of activity level, no ERUs can be earned for
decreases in activity levels within the project project boundary
or due to force majeure. With respect to output changes outside
the project activity, it may be considered to explore indirect GHG
emission sources (which are outside the project boundary and
cannot be controled by the project participants). Any change in
activity and related GHG emissions change in these indirect
sources are considered leakage and must be incorporated in the
calculated emissions of the project itself, provided that these
changes are significant
e.
As any form of baseline determination multi-project baseline
determination is surrounded by uncertainties due to the
hypothetical character of the scenario. Multi-project baseline
determination could be accompanied by an uncertainty analysis and
in case uncertainties are relatively high, conservative
assumptions could be made.12
1 Jane Ellis, Jan Corfee-Morlot, and Harald Winkler, 2004. Taking
Stock of Progress with the Clean Development Mechanism (CDM), OECD.
2 CE, 2005. Interim Evaluation of the Dutch Joint Implementation
Programme, Delft, the Netherlands, p.10, table 5.
3 See for a literature overview, JIN et al. 2003. Procedures for
Accounting and Baselines of JI and CDM projects, Groningen, the
Netherlands.
4 See
http://cdm.unfccc.int/methodologies/Pamethodologies/approved.html
5 Naoki Matsuo, 2000. Proposal for step-by-step baseline
standardisation for CDM, IGES, Kanagawa, Japan.
6 Para. (b)(v) of Appendix C to Decision -/CPM.1 (Article 12).
7 Para. b(i) of Appendix C to Decision -/CPM.1 (Article 12).
8 Para. b(iv) of Appendix C to Decision -/CPM.1 (Article 12).
9 Ministry of Economic Affairs, 2003, Operational Guidelines for PDD’s
of JI Projects, the Hague, the Netherlands.
10 Jane Ellis, 2003. Evaluating Experience with Electricity-generating
GHG mitigation projects, COM/ENV/EPOC/IEA/SLT(2003)8.
11 Wytze van der Gaast, W.P., 2005. Baseline standardisation for JI
Track-I projects, presentation at JI Track I workshop, Prague, Czech
Republic, 7-8 September 2005.
12 See JIN, 2003. Procedures for Baselines and Accounting of JI and
CDM Projects, Groningen, the Netherlands, Chapter 5, www.jiqweb.org.
9