The Potential for Continued Growth in Demand Response and Energy Efficiency Participation in OrganizedCapacity Markets
Matthew Tanner, Navigant, (202) 973-2439,
Brett Cullen, Navigant, (202) 481-8626,
John Higgins, Navigant, (781) 270-8323,
Timothy McClive, Navigant, (202) 973-4555,
ABSTRACT
OverviewIn recent years, demand-side options have provided an increasingly large proportion of capacity procured in the capacity markets of PJM, NYISO, and IS-ONE. There are several reasons for this. First, demand-side resources are often lower cost than generation capacity. Second, market designers have emphasized demand side resources as an important alternative to traditional generators . Third, the ability of central aggregators to control large numbers of demand-side capacity providers has lowered barriers to enter the market. Fourth, DR has made substantial strides in the reliability rate such that it is on par with traditional generation resources and can thus be used nearly interchangeably in some markets.
For the purposes of this paper, demand-side options are categorized by demand response (DR) and energy efficiency (EE). DR is active demand-side management. It consists of electricity users lowering their load during peak times as required by the system operator. DR is often provided by aggregators, or curtailment service providers, which pool and manage reductions by retail customers. Most demand response contracts limit the number and length of load reductions that can be called by the system operator. EE is passive demand-side management and consists of electricity users lowering their load requirements across all hours, consistently and permanently,through use of more efficient devices.
One of the surprise results of the first few years of capacity markets has been the growth of the amount of DR/EE entering and clearing the markets. Over the last fourPJM Reliability Pricing Model (RPM) auctions, the amount of DR/EE resources offered and cleared in the capacity market has risen from 10,500 MW offered with 7,600 MW cleared in 2009to 20,900 MW offered with 15,800 MW cleared in 2012. The latter is nearly 10% of all capacity procured in the auction. The other ISOs have had similar, if not quite as steep, growth. One consequence of this growth has been rising concern by system operators that system reliability is threatened by DR capacity not being as reliable as traditional generation capacity – will it be there when needed? PJM's response to these concerns was to modify the eligibility rules and divide qualified demand response into more categories depending on the reliability of the offered service. PJM also has rigorous rules for measuring and verifying reductions[1]. The other ISOs may need to consider a similar step to address concerns and performance.
Capacity market revenue is providing an increasing proportion of total revenues for demand response providers providing the economic driver for growth in demand-side participation in the capacity markets. Until very recently, the energy market comprised a substantial majority of DR revenues. As capacity market clearing prices have grown, capacity market payments expanded to account for over 95% of all DR revenues in PJM.This trend has been combined with a large increase in total revenue going to demand response resources. On the energy efficiency side, EE resources cannot participate in power markets other than the capacity markets and so have no market incentives other than this revenue.
This project presents an analysis of the future potential of demand-side resource participation in the organized capacity markets. Clearly, the future growth in demand-side resources is highly dependent on future capacity market clearing prices. Clearing prices are expected to rise in future years but concerns over system reliability and continued growth in EE/DR resources offered in the markets could limit capacity revenue received by these resouces. In PJM, the mechanism would be more forcing of demand response into unlimited contracts rather than the traditional limited ones. In the other ISOs, a concern might be the potential for rule changes similar to PJM's. A further risk to demand-side resource growth and revenue could come from a change in thedispatch frequencyof those resources. As demand-side resources constitutea larger proportion of reserve margins, peak events large enough to force curtailment will increase in likelihood from historical rates. This will increase the cost to demand side providers and may limit growth in DR.
MethodsTo forecast capacity market clearing prices and DR/EE penetration in the organizedcapacitymarkets, Navigant developed a proprietary capacity market price forecasting model. This model is integrated with Navigant’s energy price forecast and fuel price forecast models. The model estimates each plant’s bid as the amount of money required to make the plant whole after accounting for energy and ancillary services market revenue and subtracting fuel costs, fixed costs, and environmental costs. The model determines the market clearing price by equilibrating the demand for capacity by the ISOs with the aggregate supply curve created by each plant’s bid. It produces a 20-year capacity price forecast and considers transmission limitations between pricing regions. The model is also used in scenario analysis to study the impacts of market rule changes and the evolution of DR/EE role in the capacity markets.
Results
In PJM, coal retirements due to environmental regulations and low gas prices will provide support for capacity prices in the short- to medium-term. Over the long-term, rising peak loads will require more capacity to be procured, leading to increasing capacity prices. In ISO-NE, the removal of the price floor over the next few years will cause a temporary dip in price that will reduce incentives for new capacity. Over time, as in PJM, rising peak loads will raise prices back up to levels that will provide strong incentives for new capacity. In NYISO, the seasonal trend of high capacity prices in the summer and relatively lower prices in the winter can be expected to continue. NYC will continue to have much higher prices than the rest of the state due to high barriers to new plant construction and transmission constraints.
The results of the capacity market model show that DR/EE resources will continue to play a large role in providing capacity in the organized capacity markets. In PJM, eligibility rules were recently strengthened and this will dampen the growth rate of limited demand response resources. It is possible that DR will continue to grow at the same pace but it is likely that power users will demand higher premiums for participating if they expect to be called upon to interrupt their load with higher frequency. In ISO-NE, there are currently no rules limiting the penetration of DR resources. There is a risk, as future capacity prices rise and provide ever greater incentive for developing DR resources, that ISO-NE will be forced to strengthen their rules for measuring and verifying reductions due to DR in order to preserve system reliability.
Conclusions
Navigant’s capacity market pricing model forecasts capacity prices and the impact that current trends such as coal retirements will have on the clearing price. This provides evidence of the revenue that new DR/EE providers can anticipate from entering the capacity market. However, while energy efficiency and demand response resources have had impressive growth in providing capacity in the organized capacity markets, there are several factors that will work to limit continued growth. The first is changing market rules that limit the amounts of certain types of DR/EE resources that can clear. Second, the cost to develop new DR/EE resources is likely to increase in future years as the initial blocks of lower-cost resources (the “low-hanging fruit”) is exhausted. Third, the proportion of capacity resources that are active demand response is increasing,leading to higher dispatch rates of these resources and increasing the cost to convince power users to participate in demand response programs.
[1] PJM Manual 19: Load Forecasting and Analysis. Attachment B: Direct Load Control Load Research Guidelines. Revision 20, 2012.