Navigant Research Blog

PJM’s Latest Capacity Auction Shows Drop in Demand Response, but Not Catastrophic

— May 25, 2017

The holding of breath for PJM’s annual capacity auction results ended on May 23, with the results indicating mixed feelings. The price for most of the market was down from $100/MW/day for the 2019-2020 auction last year to $76.53/MW/day for 2020-2021. However, certain subzones cleared at nearly twice that price or more, so bidders in Chicago, Philadelphia, New Jersey, and Cincinnati came out smiling.

For demand response (DR), there was a lot of speculation going into the auction about the effect that the first 100% Capacity Performance procurement would have. Some analysts predicted 50% or greater reductions in DR participation, assuming most DR providers and customers would not want to take on annual performance risk. In my Market Data: Demand Response report for Navigant Research last year, I estimated a 25%-30% reduction, feeling that large commercial and industrial (C&I) customers would continue to participate; DR providers would continue to aggregate midsize C&I customers with more conservative megawatt values; and residential DR would take the biggest hit since it is almost all summer based.

Pricing, Aggregation Rules Influence Auction

The actual reduction was 24% from the last auction, dropping from 10,348 MW to 7,820 MW. Nothing to sneeze at, but far from a total market abandonment. Last year, only 614 MW of DR cleared as an annual product, so there was a large portion that was willing to convert. Pricing may have influenced DR quantities as well. While all zones decreased year-over-year, the zones with the lowest prices showed the biggest drops and those with higher than expected prices shed fewer megawatts.

This was also the first auction in which PJM instituted new aggregation rules, where summer and winter resources could match up with each other to meet the annual obligation. While 2,000 MW of summer resources (mostly DR, energy efficiency, and solar PV) submitted aggregation bids, only 485 MW of winter resources bid (mostly wind), limiting the effects of the new mechanism.

Silver Linings

Historically, EnerNOC has happily proclaimed its percent procurement of PJM DR in the auctions, but has been quiet the last couple of years. However, this year EnerNOC tweeted: “@EnerNOC captures 34% of the DR market in #PJM BRA.”

On the residential DR side, it appears that the Exelon utilities—which have been the biggest bidders in that sector—largely pulled out of the auction from the supply side. The utilities had put out an RFP in March looking for 700 MW of winter resources with which to aggregate, but apparently did not find enough partners. However, this does not mean that they exited the capacity market entirely. PJM reported that, for the first time, price-responsive demand resources cleared in the auction to the tune of 558 MW, mostly in the Baltimore Gas and Electric and Pepco regions—likely from those host utilities. If those megawatts get added to the DR megawatts that cleared in the auction, the drop is only 19% from last year.

All in all, I’d consider this a positive outcome for DR compared to some of the draconian forecasts. Now we’ll have to see how well the market performs once the annual requirement kicks in.


Deploying Energy Efficiency to Lower CO2 Emissions and Comply with the Clean Power Plan

— May 17, 2016

Cloud ComputingThis post originally appeared on the Association of Energy Services Professionals (AESP) website.

This article was co-authored by Frank Stern and Rob Neumann. Amanvir Chahal and David Purcell also contributed.

There has been a great deal of discussion on compliance with the Clean Power Plan (CPP). Surprisingly, there is little discussion of specific costs and benefits in leveraging energy efficiency (EE) to reduce CO2 and move toward complying with the CPP. Navigant investigated the effects of deploying additional EE resources to decrease CO2 emissions in two regions—California and PJM [1]. Our analysis shows that deploying additional EE for CPP compliance results in reduced CO2, as would be expected, but it also reduces costs and system congestion. Additional EE can reduce cost to serve load by 3% to 5% in California and PJM, which reduces costs annually up to $825 million in California and $1.5 billion in PJM. Another benefit of deployed EE is system congestion relief, which reduces the cost to serve load—this is important since large, urban utilities are focused on reducing congestion points, and EE can be used as a solution.

CPP and CO2 Reduction Timeline

The CPP has been stayed by the U.S. Supreme Court until final resolution of the case through the federal courts. The U.S. Supreme Court may not have final resolution of the case until 2018, although it could be sooner. Regardless, many states and regions continue to move toward the CPP goals to reduce carbon emissions, plan for an advanced energy economy, and meet cleaner generation goals. It is not known at this time if the deadlines in the CPP will be modified.

Modeling EE for CO2 Reduction

Navigant has been modeling supply resources for many years and has been including EE as a modeled resource. For this analysis, we focused on modeling PJM Transmission Interconnection and the state of California. To establish our EE base case across California and PJM, we included levels of EE modeled in each of Navigant’s most recent PROMOD and POM [2] transmission model runs. The data and assumptions in these runs are updated and verified with industry experts each quarter. Variables in the model include (i) rate of EE adoption over time, (ii) amount of EE compared to new generation, and (iii) varying amounts of EE deployed. EE was modeled across CA and PJM for the three cases (high/medium/low)—each case was run for 2025 and 2030. These years are important since 2025 is the middle of the CPP implementation period and 2030 is the first year of full compliance with the rule (final goal). The low case included a 50% reduction in EE from the base case, while the high case included a 50% increase in EE from the base case—the base case in 2030 is 33 million MWh for PJM and 24 million MWh for California.

Modeled Results

Deployed EE can provide up to 8.8% of California’s and 3.6% of PJM’s overall CPP Compliance goal in 2030. There is also a reduction in the cost to serve generation load based upon deployed EE. In PJM, the cost savings from the low EE case to the high EE case results in over $1.5 billion in savings annually in 2030 (3.6% of total cost to serve load), while in California, the same metric results in up to $825 million in savings annually in 2030 (4.7% of the cost to serve load). To state it in different terms, the cost to increase EE in 2030 to assist meeting CPP requirements is approximately $900 million in PJM and $550 million in California, which results in an EE return on investment of $600 million in PJM and $300 million in California. This lowers 2030 system capacity requirements by 5.6% in PJM and 10.7% in California. The lower savings and returns in California are due to aggressive renewable and EE policies already underway today in advance of CPP compliance.

Another benefit of deployed EE is reduced system congestion, which reduces the cost to serve load. EE will lower the need for new thermal generation on the system and put downward pressure on capacity and resource prices. Our model shows that system congestion is reduced by approximately 1.5% and is seen systemwide. This amounts to cost reductions of more than $765 million a year in PJM and $270 million a year in California. This system congestion finding is important, since there are various efforts underway across the nation to improve congestion (e.g., Con Edison Brooklyn/Queens Demand Management Initiative).


CPP initiatives would benefit greatly by incorporating additional EE into the planning process. EE reduces emissions and systems costs and pushes out the need for large, costly new generation projects. Specifically, we showed that CO2 emissions would be significantly lowered in PJM and California in both 2025 and 2030, while system costs are lowered in PJM and CA by at least 3% and 5%, respectively. This all adds up to longer glide paths for meeting regulatory requirements or when state goals have to be implemented. By including EE as a resource into the resource mix, system planners and environmental offices gain significant benefits in the form of decreasing costs, flattening demand and a zero-emitting resource.

[1] PROMOD IV is a detailed hourly chronological market model that simulates the dispatch and operation of the wholesale electricity market. It replicates the least cost optimization decision criteria used by system operators and utilities in the market while observing generating operational limitations and transmission constraints. The Proprietary Portfolio Optimization Model (POM) is leveraged for regional analysis of regulatory impacts.

[2] PJM coordinates movement of electricity through all or parts of Delaware, Illinois, Indiana, Kentucky, Maryland, Michigan, New Jersey, North Carolina, Ohio, Pennsylvania, Tennessee, Virginia, West Virginia and the District of Columbia – numerous states and diverse regions.


PJM Capacity Auction Livens Up the Dog Days of Summer

— August 24, 2015

A lot of people normally take vacations and start to think about the back-to-school rush in August, but nothing productive gets done. The same cannot be said for 2015, as PJM’s capacity auction, normally held in May, was moved to August this year due to regulatory proceedings. This change has kept people checking their messages from the beach to make sure they don’t miss any important news while working on the perfect tan.

PJM’s 2018-19 Base Residual Auction (BRA) for its Reliability Pricing Model (RPM) capacity market was held last week and it released results late last Friday. This was the first auction to include the new Capacity Performance (CP) requirements, which increase risk to suppliers but also potentially increase revenue. The auction prices for CP fell within expected ranges, elevated over the last auction. Importantly, PJM only procured 80% of its supply need with CP, with the other 20% coming from Base Capacity (BC) resources, which have lower performance requirements and lower risk. The main analyst sentiment going into the auction was that BC would clear at a much lower price than CP due to the risk premium. This did not turn out to be the case, however, as CP only cleared 7%–9% higher in most zones.

What does all this mean for demand response (DR), which was seen as a wild card in the auction outcome? All signs point to a positive prognosis—well above most expectations—with 11,000 MW clearing, about 100 MW more than the year prior. This increase is probably due to the higher prices rather than any DR industry trends. Over 90% of DR cleared in the BC product. Had the BC price ended up much lower, as was widely expected, it would have been interesting to see how much DR would have stayed in the market.

One big question was how much DR would clear in the CP product given the higher risk of penalties. The answer was about 1,500 MW, less than 10% of total DR. There are many ways to interpret this result. First, it rebuffs the notion that little to no DR would take the CP plunge. So some level of DR is here to stay once PJM starts procuring 100% CP in a couple of years. On the other hand, a very small percentage of DR cleared in CP, so it does not look like a mass-market opportunity. However, a third perspective is that because the CP premium over BC was so small, most DR suppliers chose BC for the lower risk; had the premium been much larger, perhaps more DR would have jumped to CP. A lot of those details are hidden in the bidding strategies of the suppliers and are not made public unless willingly volunteered. EnerNOC normally releases a statement soon after the auction announcing its results, but probably not that level of detail.

PJM has stolen the headlines once again, but I’m sure there will be time to discuss other energy developments once I put my surfboard away and school commences. In the meantime, you can read about EnerNOC and other DR providers in Navigant’s recently published Demand Response Leaderboard Report.


Blog Articles

Most Recent

By Date


Clean Transportation, Digital Utility Strategies, Electric Vehicles, Energy Technologies, Policy & Regulation, Renewable Energy, Smart Energy Practice, Smart Energy Program, Transportation Efficiencies, Utility Transformations

By Author