|
|
|
|
Index | Editor's note | Interviews | Discussion Platform | Project analysis | Downloads |
- go back -
CFL Projects under the CDM ?
Compact Fluorescent Lamps (CFLs) have a number of advantages over regular incandescents light bulbs. First, they only need about 20% of the electricity that incandescent lamps need and hence provide an enormous scope for energy savings. Second, their use can significantly reduce the electricity demand during peak hours, which is beneficial for the suppliers of electricity. Third, because of the lower electricity consumption, CFLs contribute to GHG emission reduction.
For those reasons many projects have been initiated since 1990 to promote the use of CFLs, both in OECD and non-OECD countries. DGIS - the Netherlands Directorate General for International Development of the Ministry of Foreign Affairs (DGIS) - was an early promoter and supporter of such energy saving projects. It has supported projects to promote the residential use of CFLs both in Honduras and in The Gambia. In addition, DGIS has also commissioned baseline and monitoring studies for both projects. The main purpose for developing a baseline and monitoring methodology for both projects was not to obtain actual credits, but to gain experience in the development of such methodologies and to share that experience with international partners and with the UNFCCC.
Preparations for the CFL project in Honduras started in 1997 at a time when the price for a CFL was still over US $15. Hence, early CFL projects such as the one in Honduras - which started effectively in 2000 - included a component to provide CFLs on credit and a mechanism for repayment in 12 instalments through additions to the monthly electricity bill. By the start of the CFL promotion project in The Gambia in 2003, the price had gone down to US $5 and hence, no credit facility had been used. By 2006 CFLs were available in The Gambia at a price of US $2 and even less.
The Dutch energy agency Senter has supported the development of specific methodologies for baseline and monitoring studies. Summit Blue company (USA) had been contracted by Senter to develop a methodology for small scale promotion of residential energy efficient lamps. This methodology had been accepted by the CDM Executive Board in its first batch of approved methodologies. The methodology included a stratified sampling approach and the generation of data on hours of use of lamps both from meters - the recorded hours of use - and from the respondents - the reported hours of use. Combining these data would allow for a ratio approach and a cost-effective increase in the accuracy of the results and their levels of confidence. On the other hand, such an approach complicates the design of the studies and increases the demand for statistical expertise. The present UNFCCC guidance for the concerned approved methodology for small scale demand-side energy efficiency programmes for specific technologies - Type II.C/Version 7, 28 November 2005 - does not prescribe the use of the ratio approach and hence its use is optional.
The 'success rate' of obtaining proper data sets for the baseline and two monitoring studies in Honduras was 85% in all 3 studies, i.e. for 85% of the lamps metered, both the recorded and the reported hours of use were acceptable and properly registered in the data base. This included some adjustments to about 5% of the recorded data for those cases were the number of 'on' and 'off' switches during the period of one week of recording exceeded the capacity of the internal memory of the meters used. Both the decisions to accept or reject and/or to adjust parts of the data involved some degree of arbitrariness as well as a significant demand on expert time. Both elements were also involved in the subsequent statistical analysis and interpretation because of some 'deviant' results in the first monitoring study.
The same 'success rate' in The Gambia was much worse; for the baseline study it was 40% and in the only monitoring study done it was 60%, despite significant efforts to reach the target of 60%. The reasons for this rather poor performance were both the irregularity in the power supply and the virtual absence of armatures or lampshades (see picture). Without power it was impossible to check the appropriateness of the installation and sensitivity of the meter and without lampshade it was difficult to fix the meters in a stable manner at a proper distance and to avoid the interference of sunlight and/or doors being opened and closed. For the monitoring study a generator was carried along to allow for testing as well as a set of 'iron wire lampshades' to install the meter. Because of the irregularity in the power supply it was not considered appropriate to gain the '5-20%' as done for Honduras by adjusting for data overflows. These overflows were also much more frequent - up to 20% in The Gambia - which suggests continued interference of sunlight and/or doors.
The light sensitive meters as used were all produced in 2000. The same company now offers light sensitive meters with an internal memory which can record 43,000 'on' and 'off' switches as opposed to just over a 1000 cycles in the earlier 2000 version. Moreover, the data analysis and conversion software has been much improved and simplified and will require less additional expert time and programming, e.g. to analyse the data in terms of use during peak hours. Hence, with further improvements in the 'iron wire lampshades' and the use of a generator for testing a 'success rate' of 75% under the 'ratio approach' should be achievable even under the worst possible circumstances.
However, the irregularity in the power supply poses major problems in interpretation, both for the respondent "how many hours a day do you normally use that lamp?" and for the interpretation of the recorded hours of use; e.g. "these 2 hours on average per day this week with an average but fluctuating power supply of 40%", what does that mean, especially if the next week the supply can be 90% or 10%. More data from a bigger sample and from additional sources is needed to deal with this problem, such as from the utility on their load shedding and how it affects the different areas.
The considerations so far were of a methodological and statistical nature with respect to the hours of use. Hours of use as found under the very different circumstances in Honduras and The Gambia are remarkably similar; around 3 hours per CFL per day in a range from 2.4 to 3.5 hours. Similarly, the wattage saved per CFL was 36W in both countries and their 'rebound effects' where identical as well at 26% (cheaper light increases total electricity consumption for light by 26% by adding more lights and/or using lights more hours per day). With some variations between baseline and monitoring results, the average CFL thus saves 0.11 kWh per day. Total kWh savings per CFL will depend upon the number of hours the CFL can be expected to last. Based on a nominal use of 10,000 hours per CFL and on standard conversions into tonnes of CO2 per MWh for generation and transmission as applied in Honduras, one such CFL can save about 0.2 tCO2 (conservative estimate). At a price of US $10 per CER, each CFL with 10.000 hours of use would carry a potential credit value of US $2, which is more than the current CFL market price in The Gambia.
In addition, CFLs reduce the need for investments in meeting peak demand and the electricity bill of consumers. The pay-back-period for CFL users in The Gambia has been calculated a 2 months, i.e. under those systems of sharing of the electricity bill for a compound as has been observed in the baseline survey.
With these tremendous potential benefits for all parties concerned, the development of CFL projects as CDM projects would certainly be able to further spread the dissemination of the CFLs at a much quicker pace. If the CER value by itself exceeds the cost per CFL, it makes sense for a CDM project developer to give CFLs away for free and reclaim the costs from the CER revenues and even make a profit on them with increasing CER prices. In fact, this is how a recent CFL project for Ghana has been designed by a CFL manufacturer, who will also gain more experience and market access. However, the methodology as proposed for this 'Ghana efficient lighting retrofit project' has as of yet not been accepted by the CDM Meth Panel and the CDM Executive Board (CDM EB) (a final decision on the acceptance of this methodology is expected early 2007).
A general requirement of a CDM project is that it must result in additional emission reductions. In order to streamline the operationalisation of the CDM additionality assessment, the CDM EB, at its 16th and 17th meeting (2004), decided to adopt a 'tool for the demonstration of additionality'. In order to provide more guidance on this aspect of the project design document (PDD), the CDM EB consolidated the additionality methods used by project developers. However, the studies for Honduras and The Gambia have been designed before the development of this tool. In Honduras the approach to demonstrate the additionality of the CFL project has been to ask the households subject to the project if they would also have purchased CFLs without the incentives from the project. The surprising answer by virtually all respondents was "yes, they did buy / would have bought the CFLs anyway." After the baseline survey the same question has been rephrased twice for both monitoring studies, but to no avail. The answer against all evidence of CFLs on credit from the project remained an almost 100% "yes." It could not be ascertained whether this was the result of culture or individual pride, but the survey could not demonstrate 'additionality'.
The baseline and monitoring methodology as developed for Ghana will rely strongly on surveys. The project participants which will be given the CFLs free of cost and they as well the control group will be surveyed every three months for a period of 10 years. This approach is required to deal with issues such as fall-out, leakage and free-riders. To demonstrate the additionality of the project, the methodology refers to the CDM EB approved 'tool' as referred to above. The Meth Panel has made observations both on the proposed application of that tool and on the treatment of the issues of leakage and free-riders. It has awarded the proposed methodology the B status, which means that the project can qualify as a CDM project if the given observations have been properly taken care of and the concerned documents have been resubmitted and approved. The EB has accepted the recommendation of the Meth Panel. The project's PDD provides details on the scale and on the costs. A total of 45,000 CFLs will be distributed, which are able to withstand voltage fluctuations between 190 and 260V and are expected to last 15,000 hours each and to generate in total 11,307 CERs, i.e. approx. 0.25 per CFL. The costs of these CFLs are €122,000 at a unit cost of €2.71 and the undiscounted costs of applying the methodology are €147,000 or €3.27 Euro per CFL. Undiscounted, the break-even rate would be €25 per CER, i.e. without the costs of preparing and adjusting the methodology and the PDD, for which the costs have not been indicated.
The CFL project in Ghana follows a retrofit approach; i.e. incandescents will be replaced by CFLs. In The Gambia a follow-up of the previous CFL project is under consideration in which the project will concentrate its promotion in certain areas in order to have other areas which can serve as a control group. Some discount will be offered, but the project will follow a market approach and not a retrofit approach. Key variables to demonstrate the success and additionality of the project will be differences in CFL market penetration rates between the targeted areas and their control areas, both to be stratified on the basis of income levels. Summit Blue has developed a rough outline for the statistical approach to be followed. However, the combination of the costs involved to further develop the methodology and the PDD plus the risks that the methodology will not be accepted and the possibility that the demonstrated success will not generate sufficient CERs to make up for the extra costs resulted in the decision not to pursue this project as a CDM project. An additional and more hidden cost is the need to design the project in a suboptimal way in order to create control areas which should be exposed as little as possible to publicity on CFLs in order to increase the difference between targeted and non-targeted areas.
The targets of the CFL projects in Honduras and The Gambia were 300,000 and 40,000 CFLs and the achievement rates at the end of the project periods were about 50 and 25% respectively. These projects are very small in comparison with e.g. Indonesia, where so far 12 million CFLs have been installed under promotional programmes and where promotional activities are still being continued. If the additional promotional activities can be defined as projects and can qualify as additional under the CDM, the scale involved may well warrant the extra costs to elaborate and implement these projects as CDM projects. If this increases the replacement of incandescents by CFLs, it will benefit the project developers, the consumers, the utilities and - most importantly - the environment.
Information on the proposed project in Ghana and its methodology is available at http://cdm.unfccc.int/methodologies/process under NM0150-rev: 'Ghana efficient lighting retrofit project'.
For more information on the practical aspects and on the results of the CFL projects in Honduras and The Gambia, please contact:
Mr Martin Zwanenburg
energy@etcnl.nl
For more information on the statistical and methodological aspects, please contact:
Ms Rachel Freeman
rfreeman@summitblue.com
Discussion Platform in JIQ Issue October 2006
Dutch Advisory Council Favours Security Council for Climate
Discussion Platform in JIQ Issue April 2006
Mr Esteve Corbera on AIJ Forestry Evaluation and Implications for CDM A/R Projects
Discussion Platform in JIQ Issue December 2005
Mr Axel Michaelowa and Mrs Katharina Michaelowa on Sustainable Development and the CDM
Discussion Platform in JIQ Issue October 2005
Mr Jusen Asuka and Mr Tadashi Okimura on the Quality and Price of Carbon Credits
Discussion Platform in JIQ Issue July 2005
Mr John O. Niles on the CCB Standards
Discussion Platform in JIQ Issue April 2005
Mr Jason Anderson on Efficiency Standards
Discussion Platform in JIQ Issue December 2004
Mr Wytze van der Gaast and Mr Joris Laseur on Unilateral CDM
Discussion Platform in JIQ Issue October 2004
Mr Donald Goldberg and Mr Kevin Baumert on Action Targets
Mr Andreas Oberheitmann and Mr Manuel Frondel on CDM in China
Discussion Platform in previous JIQ Issues
...not yet available...
|
|
|
