World Bank Experience with Power Sector Baselines Workshop

World Bank Experience with Power Sector Baselines Workshop on CDM Methodologies for Grid-Connected Power Projects Buenos Aires, 8 December 2004 Fernando Cubillos – World Bank, Carbon Finance Business

Financial Impact of CO 2 Revenues Carbon Revenues make a difference, but not a big difference

Power Sector Projects System capacity Thermal Dispatch margin Dispatch Hydro Base load Coal Renewable Base load capacity replacement Wind Major capacity additional? Micro capacity addition

WB-CFB Experience • 2001, Chile - Chacabuquito, hydro r-o-r, 26 MW • 2001, Costa Rica – Umbrella, wind and hydro, 10 MW • 2002, Guatemala – El Canada, hydor r-o-r, 53 MW • 2002, Colombia – Jepirachi, wind, 19 MW • 2003, Mexico – El Gallo, hydro r-o-r, 30 MW All of them well before CDM rules and methodologies were developed

Guessing CDM rules development • How we developed the ERPAs/MP to buy VERs • Baseline Studies to asses – if the project is part of the most likely course of action (additionality) – ERs projections using different scenarios – the best method to calculate verifiable ERs

Least cost baseline methods for power projects 1. Run expansion planning model with project: is project not being picked up by the model? (Chile, Colombia) 2. Use expansion planning model to determine long-run marginal cost (LRMC): (Costa Rica, Chile, Colombia) is project cost (per k. Wh) higher than LRMC? 3. Determine low/least cost project to represent expansion option (Guatemala, Chile): is project cost higher than cost of comparable expansion option? 4. Barrier analysis / additionality tools (Mexico): is project impeded by barriers other than costs If yes: the baseline is the power system without the proposed project => project additionality

Chile – Chacabuquito Experience • Comparison of the project with the Least Cost Option for the Expansion – Possible alternatives:

Chile – Chacabuquito (cont. ) • ERs Projection – Will the project postpone new capacity? , or – Will the project impact the actual operation? • Size comparison of the project with the Least cost option for the expansion: – 110 GWh of firm energy vs. 3, 000 GWh – Theorical/modelling postponing less than 1 months! • Will that change in the grid affect the investment decision for the next capacity addition?

Chile: Central Grid Generation and Dispatch to Meet Future Demand (to scale) GWh Year New 300 MW CC plants

Calculating the impact in the actual operation • Ex ante Emission Factor – Coal is the most likely fuel of displacement (60% - 80%) – CCGT Natural Gas fired (15% - 35%) – Only other renewable Hydro (<5%) • How to weight them for an ex-ante EF? – 100/00/00; 80/20/0; 60/35/5; etc. • Proposal: Ex-post monitoring based on Dispatch Analysis No need to weight!

Some characteristics of Chile Power Sector • Efficient regulation for more than 20 years • Grid operation run by an independent agency (Dispatch Center) • Detailed and public data available to calculate in a very accurate way the marginal plant and the emission factor associated

Monitoring ERs with Dispatch Analysis

Step 1: Hourly Net Generation from CDM Power Plant

Step 2 – Hourly Generation from all power plants in the Grid

Step 3 – Dispatch Priority List

Step 4 – Determination of the marginal generation displaced

Dispatch Analysis for the Chacabuquito Project: Emission Reductions Factor

Dispatch Analysis for the Chacabuquito Project: EF and Generation

Issues with dispatch marginal calculations • Double Counting: – Central coordination may be needed to prevent double counting of displaced marginal generation – Calculate ERs for all CDM projects jointly or sequentially on a seniority basis? • Capacity replacement: – How does CDM system expansion modify baseline system expansion? – What is the impact of one CDM project vs. the accumulated impact of many CDM projects? – Include emission factor of delayed capacity for time of delay • Storage hydro: – Can operate at the margin to follow load. – Is not being replaced, but shifted in time. – Impact on CO 2 emissions difficult to analyze. • System bottlenecks and “political” decisions setting system boundaries

Mexico – El Gallo 30 MW hydro • Mexico’s grid mainly stated owned and operated centralized • Thermal based energy matrix • There are no public data available to calculate dispatch margin

El Gallo: Methodology for additionality test Step 1: Analyze barriers to proposed project Sub-step 1: • Analyze barriers to the proposed project: • The following barriers identified: – (i) access to financing; – (ii) perceived technology risk; – (iii) transaction costs.

El Gallo: Methodology for additionality test Step 2: Discussion of similar projects It is demonstrated that the project differs in important dimensions from similar projects: 1. Comparable projects are less capital intensive 2. Most comparable projects had access to sufficient equity capital 3. Comparable projects are structured as auto-generators

El Gallo (cont’d…) Baseline emission rate The emission rate is calculated as the sum of the emission rates of (1) the operating margin and (2) the build margin divided by 2: ER (t. CO 2/MWh) = WOM x OM + WBM x BM Default weights are set equal to 0. 5.

El Gallo (cont’d…) where Gi = the generation (MWh) from unit i, ei = the emission rate (in tons CO 2/MWh) for unit i, and i = individual units, excluding low running cost or mustrun plants (i. e. hydro, geothermal, wind, low-cost biomass, nuclear and solar generation)

El Gallo (cont’d…) where Gj = the generation (MWh) from unit j, ej = the emission rate (in tons CO 2/MWh) for unit j, and j = individual plants included among the lesser (in MW) of the most recent 20% or the 5 most recent (over 15 MW) units built

Lessons learned • Complexity: – Baselines / ER calculation for projects in integrated technical or economic systems (e. g. power grids) can be highly complex – May require regional/country standardization as the number of projects grows. • Meth Panel: – Prefers a combination of additionality tools and simple to calculate, predetermined, and conservative baseline emission factors. – Simplification / generalization is at the expense of accuracy and leads to more conservatively calculated ERs. • Policies: – The problems of national and sectoral policies, supply shortage situations, transmission bottlenecks and “perverse” incentives needs.