Energy Policy, Vol.35, No.2, 899-906, 2007
Modeling market power in Korea's emerging power market
The Korean power market is being formed from the unbundled generation, transmission and distribution assets of Korea Electric Power Corporation. The KEPCO generation has been allocated to six independent gencos with a combined generating capacity of 46,629 MW in 2002. This gave an 11% margin over the peak load that year (41,921 MW). One of the concerns for any power market is whether individual participants can increase profits (and prices) by withholding generation from the market. To address this concern, a Cournot-based model of Korean power system was created and applied to a set of loads representing the load duration curve for Korea's system loads in 2002. Our simulation results show a strong possibility for exercise of market power to increase market price in Korean market. Under tight market conditions, even I GW of withholding can cause a large increase in market price. If loads unexpectedly grow faster than the 5% recent experience, the gencos will have the collective ability and incentive to spike prices further. Vesting contracts can reduce the incentive to act strategically. Requiring that the gencos offer 50% of their capacity in long-term forward contracts greatly reduces the payoff to act strategically, and requiring vesting for 75% of their capacity results in prices that are essentially the same as the competitive equilibrium. Depending on the price for the vesting contracts, this policy can reduce the incentives to add new generation by gencos or the competitive fringe. Another approach to reducing the effects of market power is establishing demand-response programs, simulated here by increasing the elasticity of overall demand. These programs can reduce the incentives to withhold capacity, but to a lesser degree than vesting contracts. The genco with the greatest ability to influence prices through withholding is the largest, KNHP. However, acting on its own, without the support of the other gencos, its ability to raise prices is limited. This applies even more for the other gencos; alone their ability to run up prices by withholding is limited. Only when they reinforce each other's withholding by further withholding do we see the extreme prices of the Cournot equilibrium. In the long run, the high prices from the Cournot equilibrium create a strong incentive for entry of new generating capacity. While the gencos that do the most withholding in the Cournot solution do not gain by adding new generation, several of the gencos that do relatively less withholding would gain net revenues from adding capacity that is likely to exceed the cost of new capacity. For participants on the competitive fringe, the incentives to add new generating capacity are much stronger. They do not have a stake in maintaining the high Cournot solution prices and benefit from increased withholding by the incumbent gencos. The projected net revenues from new baseload and peaking capacity greatly exceed the thresholds needed to recover investment costs. With freedom for new entrants to add capacity, the Cournot equilibrium prices cannot be sustained. (c) 2006 Elsevier Ltd. All rights reserved.