Energy Meter

A recently published working paper by the Great Plains Institute (GPI), titled “Consumer Savings, Price, and Emissions Impacts of Increasing Demand Response in the Midcontinent Electricity Market,” explores the effects of increasing the use of demand response (DR) assets in the Midcontinent Independent System Operator’s (MISO) wholesale energy market.

Through the use of a supply and demand market model, we estimate the price and emissions impacts of a hypothetical case in which DR in the region is integrated into MISO’s energy supply stack and competes against other generation technologies to meet the energy demand of the region. We find significant price reductions and consumer savings can be achieved through increased DR dispatch during peak demand hours, especially during high price events.

We set out on this work because MISO has a large, but under-utilized group of DR assets in its markets, as well as a large potential for future growth. Additionally, the market has recently experienced high price spikes that could be mitigated by targeted DR dispatch.

DR has historically played a key role in maintaining the reliability of the bulk electric system. As such, the vast majority of DR in the MISO region is comprised of legacy load control programs registered under the Load Modifying Resource asset classification. Under this classification, utilities that own the programs can capture their capacity value to help meet resource adequacy requirements and, in exchange, agree to make these programs available for dispatch during grid emergency events. The problem is, since the launch of MISO’s markets, there have only been a handful of times these Load Modifying Resources have been called on.

As a result, there is a relatively large, low-cost, low-emission resource being underutilized which, as our analysis shows, could otherwise help lower cost to customers in the region.


Our analysis starts with a novel approach to developing energy offers for the DR programs in the MISO region. The Energy Information Administration publishes data on utility demand response programs, including estimates of the cost to run the programs and their total capacity. We use this data as a starting point to estimate the marginal cost to dispatch the program, which drives the programs would-be market behavior. This is an important step in our analysis because DR offers are a generally opaque topic of discussion in the industry. Apart from retail aggregators who have business models designed around offering DR programs into wholesale markets, many actors in the sector do not have a strong sense of how to expect these programs to participate in energy markets.

Energy offers for traditional generation are based on the marginal cost of providing the energy being offered into the market, typically driven largely by fuel costs. However, there is no fuel (typically) for DR and instead the costs of deploying and calling on DR programs are driven by a complex fabric of program costs including:

  • the cost of electricity interruption
  • outreach to and registration of customers
  • equipment installation and maintenance, and administrative costs

This means that the true marginal cost of dispatching a DR program may be very small, but the overall program cost may be much higher and therefore utilities and other DR providers have to balance long-term cost recovery with short-term market bidding strategies. The reality is, this balance will be different for each individual utility or aggregator. This is a continued area of work for us and future work in this analysis will include the development of sensitivities on these estimated offers.

Once we determined the DR programs and their respective offers, we identified the days and hours in 2015 which offer the highest value for dispatch of the DR via the calculation of a price-load product. We multiplied the locational marginal price are the major hubs in each sub-region in MISO (North, Central, and South) times the total load in each sub-region.

We then used a supply-demand model to predict which resources clear the energy market and market clearing prices, comparing our test case with DR embedded in the supply curve, and the observed market outcomes from 2015. We also ran additional sensitivities around rebound assumptions, which is the amount of demand expected to show back up on the system after the DR dispatch event. For example, if a DR program controls the use of a hot water heater, load would be reduced during the dispatch event, but we would still expect that load to show up to some degree later on so that the homeowner still has hot water. We also estimated avoided CO2 emissions based on the marginal emissions rate of the MISO grid at the time of the DR dispatch.


Our analysis shows that dispatching DR during high-value times in MISO’s energy market can generate consumer savings ranging from 0.5 percent to 3 percent of the daily market value in MISO under typical peak conditions. We also observe several outliers which demonstrate the large potential value of DR. We see that a relatively small amount of DR deployed in MISO South could have dramatic impact on prices if deployed when and where the market is clearing in the vertical portion of the supply curve. The carbon dioxide emissions reduction we estimated show environmental benefits from DR dispatch, which are somewhat offset depending on the amount of energy shifting that occurs. Furthermore, DR enhances flexibility for grid operators, helping to increase levels of renewable generation and provide longer-term environmental benefits.

For much more detail on our methodology and results, check out the full working paper here.

Share this: