PROTOTYPE CARBON FUND, WORLD BANK AND THE MINISTRY OF ENVIRONMENT AND NATURAL RESOURCES, EL SALVADOR
Electric Power Sector Baseline Study
for El Salvador
November 2003
Lasse Ringius, UNEP Collaborating Centre for Energy and Environment
Ismael Sanchez, Universidad Centroamericana “J. S. Cañas”
Sivan Kartha, Michael Lazarus, SEI-Boston/Tellus Institute
Table of Contents
Executive Summary
The Electric Power Sector in El Salvador
1.1. Loads and Resources
1.2. Institutional structure
1.3. Structure of the Wholesale Electricity Market
1.4. The Retail Market
2. 2. Baseline Methodology......
2.1. Assessment of the current situation
2.2. A Framework for Baseline Standardization
2.3. Additionality evaluation
2.4. Baseline Emission Rate
2.5. Crediting periods and baseline updating
2.6. Scenario analysis and project-specific baselines for larger projects
2.7. Summary of the overall methodology and decision tree
2.8. Conclusion
Annex A: Alternative Electricity Baseline Methodologies
Annex B: Imports and SIEPAC
References:
El Salvador Power Sector Baseline Study
Executive Summary
An electric power sector baseline represents the most likely or plausible scenario for the future development of the power sector in the absence of Emission Reduction (ER) projects. The sector baseline is developed with one explicit aim, namely predicting to the extent possible the emissions of greenhouse gases (GHGs) that would occur as a result of the construction and operation of electricity generation facilities under business-as-usual conditions. This baseline can then be used on a routine basis to determine the ERs attributable to individual projects, thereby avoiding the need to conduct detailed baseline studies for each project activity.
This baseline study (BLS) takes into account data and information on historical and recent electricity sector developments. During the preparation of this BLS, government officials, regulators, system operators, energy companies, proponents of candidate PCF projects, and academic experts in El Salvador have been consulted. They have provided much of the data and information on which this baseline study is based.
This BLS suggests a two-pronged baseline methodology in which a combined margin approach would be applied to small-scale projects and project investments in continued grid operation and expansion, and a project-specific baseline emission rate would be determined for large-scale projects that displace discrete investments or actions. The combined margin approach provides a simple algorithm for estimating the effects of a new project on emissions from (1) the operation of current and future power plants (referred to as the operating margin) and (2) on whether and when new power plants would be built (referred to as the build margin). In order to assess whether projects would likely represent truly additional investments, all small-scale projects will be subjected to a simple test or screen based on penetration threshold. Small-scale projects using a technology that supply less than 5% of current generation are emerging or otherwise constrained and thus considered additional. Projects that exceed the 5% penetration rate would be considered additional if their generation cost exceeds the 3-year average annual spot market price of electricity in El Salvador, or if it could be demonstrated persuasively that significant barriers prevent projects from being undertaken.
Calculations of expected ERs produced by ER projects are presented in an accompanying Emission Reduction Simulation (ERS) analysis. Net emission reductions for ER projects are calculated on the basis of the emission rates of displaced power plants, their efficiencies, and type of fuels used. The principal factors that determine the ERs calculation are the baseline emission rate (as determined by the methodology described herein), the carbon intensity of ER projects, and the energy production from the ER projects. Determination of actual ERs will be based on actual plant operating data collected by SIGET[1], data on recent and ongoing plant construction, and data on actual energy production collected by the project operators. The procedures to be followed in project monitoring are presented in a separate document, the Monitoring Plan (MP). The BLS, the MP, and the ERS are integrated documents—the MP builds on and is fully consistent with the BLS—and should be understood and used as such.
The Electric Power Sector in El Salvador
1.1. Loads and Resources
Installed Capacity
By December 2001 El Salvador had 1,118 MW of installed electricity generation capacity. The generation mix consisted of 396 MW hydroelectric (35%), 161 MW geothermal (15%), and 561 MW of thermoelectric generation capacity (50%).
Historical Generation and Demand
Table 1 shows the pattern of electricity generation over the past 12 years. Total generation has nearly doubled, in response to an average annual growth rate in electricity demand of 6.1%.[2] Nearly half of this growth occurred during the three years following the end of the El Salvador civil war in 1992 in response to rapid economic growth. In recent years, growth in generation has slowed significantly due to earthquakes and slowed economic growth. As shown, in 2001, thermal generation accounted for 45%, hydro 31%, and geothermal 24% of the generation.
Table 1. Net Historical Generation per Resource in El Salvador (GWh).
Hydro / Geothermal / Thermal / Total Generation / Net Imports / Trans. Losses1990 / 1,642 / 384 / 139 / 2,164 / 1 / 149
1991 / 1,263 / 392 / 576 / 2,231 / 5 / 149
1992 / 1,410 / 359 / 547 / 2,317 / 53 / 121
1993 / 1,512 / 351 / 856 / 2,718 / 79 / 171
1994 / 1,442 / 373 / 1,260 / 3,075 / (11) / 159
1995 / 1,465 / 410 / 1,396 / 3,271 / (35) / 112
1996 / 1,877 / 400 / 1,064 / 3,341 / 21 / 153
1997 / 1,424 / 453 / 1,671 / 3,548 / 88 / 98
1998 / 1,561 / 418 / 1,758 / 3,737 / 38 / 89
1999 / 1,762 / 558 / 1,319 / 3,638 / 250 / 109
2000 / 1,170 / 739 / 1,416 / 3,377 / 696 / 142
2001 / 1,158 / 907 / 1,691 / 3,762 / 309 / 107
Source: SIGET (2002).
Electricity Prices
Since the privatization of the energy sector, El Salvador has experienced very high electricity prices (average of about 7c/kWh on the wholesale market) which may prompt the government to put pressure on generators, and has attracted investment and interest from generators from other countries.
1.2. Institutional structure
Beginning in 1998, in order to create a more competitive power sector, the government has significantly reduced the extent of its ownership, direct participation, and planning functions in the electric power sector in El Salvador. The sector has to various degrees unbundled its generation, transmission, and distribution activities. In the future most of the system expansion will be performed by the private sector. The government will only play a policy role and perform oversight and regulatory functions.
The Transactions Unit (UT) acts as the independent system operator, managing dispatch of the wholesale electricity market (MME) of El Salvador.[3] The UT is a private organization, owned exclusively by the operators and end users participating directly in the MME. Superintendencia General de Electricidad y Telecomunicaciones (SIGET) is the government regulatory body charged with overseeing UT operations. Dirección de Energia Eléctrica (“Department of Electric Energy”), established in 1999 under the Ministry of Economy, oversees generation, transmission, distribution and commercialization activities in the electric power and sets national policies.
Actors in the Generation Market
Five major generation companies with eleven power stations supply the Salvadorian wholesale market. Out of the eleven power stations, four are hydro, two geothermal, and the rest are thermoelectric power plants, as shown in Table 2 below. Private companies own the thermal capacity. Hydroelectric and geothermal facilities in El Salvador remain under state ownership.
Table 2. Main Generators Participating in the Wholesale Market (2001).
Central Generators / Resource / Available Capacity, MW / OwnershipGuajoyo / Hydro / 17 / CEL – state
Cerrón Grande / Hydro / 135 / CEL – state
5 de Noviembre / Hydro / 59 / CEL – state
15 de Septiembre / Hydro / 156 / CEL – state
Ahuachapán / Geoth. / 56 / GESAL – state
Berlín / Geoth. / 55 / GESAL – state
Acajutla / Therm. / 276 / Duke – private
Soyapango / Therm. / 54 / Duke – private
San Miguel / Therm. / 24 / Duke – private
El Paso / Therm. / 127 / Nejapa – private
El Ronco / Therm. / 21 / CESSA – private
Source: SIGET (2002).
Transmission
In accordance with the General Electricity Law of October 1996, the independent transmission company, ETESAL, which is owned by the state-owned CEL,[4] is responsible for maintenance of the national grid.
The national transmission network belongs to ETESAL. The network consists of 32 115 kV transmission lines plus two 230 kV interconnections with Guatemala (at Ahuachapán) and Honduras (at “15 de Septiembre”).
Distribution
The distribution system was privatized in January 1998. Today there are five distribution companies, covering five major regions in El Salvador: AES-CLESA; CAESS; DELSUR; EEO; and DEUSEM.[5]
1.3. Structure of the Wholesale Electricity Market
The wholesale electricity market in El Salvador is composed of all the generators, distributors, and major users that are directly connected to the 115kV transmission system. The players are known as the wholesale market participants (MPs). The Transactions Unit (UT) coordinates and undertakes the programmed dispatch for each defined time unit according to the contracted energy transactions. The wholesale market is composed of both a contracts market and a spot market.
Contracts Market
The contracts market is based on previously agreed contracts between generators and distributors/ brokers/users. The UT must be informed of all contracted transactions. But the UT receives no information on the prices agreed to in the contracts market; the prices negotiated between generation companies and users/distributors/users are considered proprietary information. It seems a reasonable assumption, however, that the contracts market prices are an agreed fraction or proportion of the spot market price—for instance, the contracts price would be the spot market price in the relevant one-hour time intervals less 5-10%. Contract lengths vary, and are most typically agreed for a one-year period at a time, with fixed delivery amounts and timing. At the end of the contracting period, the generator and the user usually negotiate a contract continuation.
Spot Market
Non-contracted energy is traded in the Mercado Regular del Sistema (MRS)—the spot market—and is based on a system of price and volume bids and offers for energy. Spot market prices are set for one-hour intervals. The UT sets the price one day in advance, based on bids and offers received from generators, distributors, and users. The market-clearing price—which is the price of the last and most expensive unit called upon to generate in the hour—sets the price received by all generators dispatched during that one-hour interval.[6] Table 3 illustrates the spot market prices, and the energy volumes dispatched, on a typical day at the end of the dry season. Spot market prices are publicly available free-of-charge on the UT’s web-page.[7]
Table 3. Volume of energy traded in the contract and spot
markets; hourly spot market prices on May 7, 2002.
Hourly interval / Contracts market(MWh) / Spot market
(MWh) / Spot market price (US$/MWh)
1 / 276.2 / 99.8 / 53.8
2 / 274.5 / 92.8 / 47.19
3 / 273.6 / 87.1 / 47.19
4 / 272.3 / 83.9 / 47.19
5 / 289.7 / 81.3 / 47.19
6 / 299.4 / 103.9 / 53.85
7 / 328.3 / 100.3 / 101.35
8 / 374.0 / 115.3 / 101.81
9 / 423.9 / 154.2 / 121.90
10 / 431.0 / 192.5 / 102.30
11 / 433.8 / 215.9 / 124.21
12 / 430.9 / 225.8 / 124.30
13 / 431.6 / 208.1 / 122.20
14 / 436.5 / 216.3 / 121.86
15 / 438.1 / 232.3 / 124.16
16 / 437.8 / 227.4 / 124.22
17 / 436.2 / 193.5 / 121.89
18 / 435.6 / 138.0 / 101.43
19 / 438.2 / 204.5 / 82.46
20 / 443.6 / 253.9 / 83.85
21 / 439.9 / 213.1 / 82.39
22 / 430.8 / 139.6 / 141.10
23 / 338.5 / 136.9 / 121.60
24 / 306.0 / 103.7 / 53.85
Total (%) / 70.5 / 29.5
Source: UT (2002).
The fraction of electricity delivered by the contracts market (vs. the spot market) has been steadily increasing, from an average of 69.2% in 1999 to 87.7% in December 2000.[8] As Table 3 and Figure 1 show, on May 7, 2002 around 70.5% of the total energy volume was traded in the contracts market; the remaining 29.5% was traded in the spot market. But, as mentioned, the spot market “drives” prices in the contracts market and, therefore, in the entire wholesale market for electricity in El Salvador.
Figure 1. Contracts and Spot Market, May 7, 2002.
1.4. The Retail Market
This market is made up of small hydro generation plants, sugar cane mills and self generators (diesel or fuel oil engines) which are connected to the distribution power lines system and sell their surplus electricity directly either to distributors, brokers, or final users.
2.2. Baseline Methodology
A primary goal of this study is to develop a robust, standardized methodology that can provide accurate, transparent, and conservative baselines for future CDM electricity generation and efficiency projects in El Salvador. At present, the Designated National Authority for the CDM in El Salvador has identified some renewable energy projects, such as small-scale biomass-based generation from sugar cane bagasse and sustainably produced wood as well as small- and large-scale[9] geothermal and hydroelectric projects, as potential CDM projects. It is anticipated that other renewable energy, energy efficiency, and fuel-switching projects will be developed as CDM projects in El Salvador. The cost of conducting project-specific baseline studies for each prospective CDM investment could easily outweigh the benefits of CDM registration, especially for smaller projects, thereby dissuading project development. A common, standardized baseline methodology for all projects offers not only reduced transaction costs, but predictability and a level playing field. A well-designed standardized baseline methodology could ultimately spur the development of CDM projects in El Salvador.
Key international precedents and a substantial research base on electricity baselines can guide the development of an appropriate standardization methodology. The Marrakech Accords of the UNFCCC establish basic guidelines that must be followed for baseline methodologies to be accepted.[10] The Accords also dictate that simplified procedures and modalities will be developed for small-scale CDM project activities and, in January 2003, the CDM Executive Board has established specific standardized baseline methodologies for small-scale grid-electricity projects.[11] The El Salvador methodology should ideally be consistent with these guidelines and small-scale methodologies, in order to maximize the prospects for prompt project approval.
2.1. Assessment of the current situation
Before discussing specific baseline methodologies, it is worth reviewing some general observations about the El Salvador power sector:
- Thermal generators exclusively use oil products, and are the highest operating cost resources in El Salvador, a situation unlikely to change in the near future. The load-following capability of some hydro resources and strategic behavior may sometimes put hydro in the spot market and near or perhaps above the market-clearing price. However, hydro remains a low-cost resource and if well managed, its generation potential will rarely be underutilized due to displacement by other resources. Therefore, it can be reasonably stated that thermal resources are and should remain on the operating margin for several years at a minimum, regardless of the operation of existing hydro and geothermal sources or the construction of new resources.
- For the next few years, anecdotal information suggests that nearly all new power plant capacity will be oil-fired, primarily using low-speed internal combustion engines. Construction of a natural gas pipeline from Mexico through Central America was considered a few years ago but is no longer under consideration. In fact, because of the high fuel costs and the increasing interconnection capacity, gas turbines in El Salvador are currently being physically moved to a more attractive market. Coal generation has for a number of reasons been given low priority (ash handling; SO2 emissions; high transportation costs; etc.). In this situation, the preferred options appear to be low-speed combustion engines running mainly on fuel oil #6 (bunker fuel) and low-capacity (30 MW) steam-power plants burning the same fuel. Thus, given current circumstances, this would imply that oil-fired plants are likely to be on the margin for new construction for much of this decade.
- There is considerable divergence of opinion on the extent of likely geothermal and hydro development in the next few years. Many point to how privatisation has transformed the electricity market. Private sector actors face higher costs of capital and discount rates and are thus biased toward lower capital cost technologies and projects with short payback periods. The large up-front investments and the long lead-time associated with geothermal and hydroelectric assets are currently major obstacles to their development. However, several hydro and geothermal projects are in the planning stages. This observation, combined with current concerns about the high electricity prices, could induce the El Salvador government to continue to promote further hydro and geothermal development, as it has done in the recent past.
If one assumes that hydro, geothermal, and major transmission investments will not occur in the near term, then one could safely infer that El Salvador emission reduction projects in the power sector would displace oil-based generation, either from existing or yet-to-be-built facilities. A simplified baseline corresponding to oil-fired generation would result. The main challenge would be in determining which of the oil-fired generation types would be displaced. Emission rates of most oil-fired plants vary from about 1.1 tCO2/MWh at small diesel-fired gas turbines to 0.7 tCO2/MWh at new internal combustion units powered by fuel oil.
However, uncertainties and other perspectives suggest that a simple oil-only baseline may not be sufficiently accurate:
- Investment conditions more favourable to hydro and geothermal resources. Even though the financial realities of the current privatised system may dictate against these resource investments, government policy could choose to play a stronger hand in power supply investments through regulation, direct ownership, or incentives.
- More rapid-than-expected interconnection among Central America countries (SIEPAC). High energy prices in El Salvador are attractive to generators from other countries in the region, particularly from Costa Rica, which has abundant hydro resources. Investors have also been considering a large combined cycle plant using liquefied natural gas on the coast of Honduras, and could aim to profit from high prices in the El Salvador market. Increased interconnection could lead to increased imports and regional grid operation, affecting which plants are effectively on the margin.
- Other major changes in market conditions. Given that El Salvador is heavily dependent on oil products for thermoelectric generation, any major international oil price hikes could have a major effect on fuel choice, plants operations, and future investment patterns.
A sectoral baseline methodology should thus be sufficiently flexible to account for such outcomes.
2.2. A Framework for Baseline Standardization
A wide range of standardized electricity baseline methods have been proposed and applied internationally in the context of existing project-based emissions credit trading programs.[12] Among others, these options include electricity baselines based on average or marginal generation, annual or time-of-use generation, and local or international data in regional or national grids. These baselines have been developed using both sophisticated models and simple off-line calculations. They have been tested and applied in a number of nearby countries, such as Mexico, Costa Rica, Nicaragua, and Colombia.[13] The choice among these and other options involves potential tradeoffs in data availability, accuracy, transaction cost, and likelihood of approval by the CDM EB that must be carefully weighed.[14]