Published September 2018 | Version v1
Journal article

A structural model to evaluate the transition from self-commitment to centralized unit commitment

  • 1. Quantil (Colombia)
  • 2. Stanford University (United States)
  • 3. Université catholique de Louvain (Belgium)
  • 4. CEII - Banco de la República (Colombia)
  • 5. University of los Andes (Colombia)
  • 6. University of California Berkeley (United States)

Description

Highlights: • We study the relative merits of self-commitment vs. centralized unit commitment in liberalized electricity market. • We use data from the transition that took place in Colombia in 2009 from self-unit commitment to centralized unit commitment. • The data is from the years 2006–2012 and built a dispatch model for the Colombian electricity system. • The model replicates the unit commitment optimization employed by the Colombian independent system operator. • Using this model, we simulate a competitive a benchmark using estimated marginal costs, startup costs and opportunity costs. • For thermal plants we estimate marginal costs using fuel prices, transportation costs and the currency exchange rate. • Hydro's opportunity costs are estimated using the marginal costs of dispatch plants and the hydro plant energy bids. • The competitive benchmark for the period 2006 - 2009 is based on counterfactual simulations using the estimated costs and our dispatch model. • We estimate changes in deadweight losses due to misrepresentation of cost by bidders and dispatch inefficiency. • Results suggest that centralized unit commitment has improved economic efficiency, reducing the relative deadweight loss by at least 3.32%. • The result supports the claim that dispatch efficiency has improved after the transition. • Our results are in line with our previous econometric study: Riascos, A., Bernal, M. de Castro, L. and S. Oren. (2016). • Results also suggest that these welfare gains have been captured by the generators rather than being transferred to the consumers. - Abstract: We introduce a dispatch model of Colombia's independent system operator in order to study the relative merits of self-commitment vs. centralized unit commitment. We capitalize on the transition that took place in Colombia in 2009 from self-unit commitment to centralized unit commitment and use data for the period 2006–2012. In our analysis we simulate a competitive benchmark based on estimated marginal costs, startup costs and opportunity costs of thermal and hydro plants. We compare the differences between the self-commitment for the period 2006–2009 and the competitive benchmark to the differences between the bid-based centralized unit commitment and the competitive benchmark after the transition. Based on these comparisons we estimate changes in deadweight losses due to misrepresentation of cost by bidders and dispatch inefficiency. The results suggest that centralized unit commitment has improved economic efficiency, reducing the relative deadweight loss by at least 3.32%. This result could in part be explained by the observation that, before 2009, there was an underproduction of thermal energy relative to the competitive benchmark and it supports the claim that dispatch efficiency has improved after the transition.

Availability note (English)

Available from http://dx.doi.org/10.1016/j.eneco.2018.09.009

Additional details

Identifiers

DOI
10.1016/j.eneco.2018.09.009;
PII
S0140988318303803;

Publishing Information

Journal Title
Energy Economics
Journal Volume
75
Journal Page Range
p. 560-572
ISSN
0140-9883
CODEN
EECODR

INIS

Country of Publication
United Kingdom
Country of Input or Organization
International Atomic Energy Agency (IAEA)
INIS RN
50035021
Subject category
S29: ENERGY PLANNING, POLICY AND ECONOMY;
Descriptors DEI
BENCHMARKS; COLOMBIA; COST; ECONOMETRICS; ELECTRICITY; FOREIGN EXCHANGE RATE; FUELS; MARKET; OPTIMIZATION; POWER TRANSMISSION; PRICES; SIMULATION; STRUCTURAL MODELS; THERMAL POWER PLANTS
Descriptors DEC
DEVELOPING COUNTRIES; ECONOMICS; LATIN AMERICA; POWER PLANTS; SOUTH AMERICA

Optional Information

Copyright
Copyright (c) 2017 Elsevier Science B.V., Amsterdam, The Netherlands, All rights reserved.