Like Supply-Side Resources, Demand Response Must be Managed through efficient processing of business transactions, including registration, bidding, scheduling, measurement, verification, settlement and dispute resolution. With an increasing number of participants, rising volume of transactions, and frequent changes in tariffs and business rules, management of demand-response (DR) programs is becoming more difficult. To meet this challenge, PJM, LLC (Valley Forge, Pennsylvania, U.S.) is implementing the next generation of DR management solutions.


DR programs have the potential to act as shock absorbers in wholesale electricity markets, dampening price spikes during peak demand periods and significantly reducing price volatility while enhancing grid reliability. When effectively integrated in organized markets, DR programs can produce impressive benefits for grid operators and customers alike.

Recognizing the potential of DR, the Federal Energy Regulatory Commission issued an order in October 2008 that effectively puts DR on par with supply-side options. Specifically, FERC's new rules require regional transmission organizations (RTOs), including PJM, to:

  • Accept bids for ancillary services from technically capable DR resources, as RTOs currently do for supply-side resources

  • Eliminate certain charges to buyers in the energy market for voluntarily reducing demand during a system emergency

  • Assess and report any barriers to comparable treatment of DR

  • Allow the market price to more accurately reflect the value of energy during periods of operating reserve shortages.

Permit an aggregator of retail DR to bid the combined negative load directly into organized markets.

The question is: Why have DR programs not been more widely applied and routinely used to date? The simple answer is that implementing DR programs is not as simple as it may appear. And, integrating their effective operation is challenging due to the growing volume of transactions and frequent changes in market protocols, rules and tariffs.


The challenge of meeting customer demand during infrequent system peaks is well known. With the evolution of organized markets, competing generators offer to meet customer demand requirements day-ahead, hourly and in real time. For most hours of the year, ample generation capacity is available to serve load, which forces generators to compete with one another to be dispatched, keeping prices in check.

However, during heavy demand periods or system emergencies, when peak load approaches or exceeds installed capacity, generators could effectively offer very high prices. And, such offers could be accepted to ensure that the grid operator has sufficient resources to meet load, plus a reserve margin to maintain reliability. This, of course, assumes there is no price elasticity, meaning that customers will take what they need regardless of the price.

However, it is known that many customers — especially price-sensitive, energy-intensive industries — have discretionary loads they can curtail or on-site generation units they can deploy, if given sufficient incentives to do so. The ability of customers to voluntarily reduce load in response to financial incentives is the very definition of DR programs. PJM and other grid operators have developed specific markets for customers who can declare how much load they are willing to reduce at different price levels when there is a heavy demand. To enable the participation of customers with small individual loads, PJM allows curtailment service providers to aggregate such loads and act as intermediaries.

The graph above illustrates the extent to which load curtailment, attributed to DR programs, affected the load-duration curve for the PJM system in August 2007. The graph also distinguishes the contribution of economic and emergency programs.

In such circumstances, the grid operator can select the negative resource, the DR, as opposed to paying higher prices to dispatch more expensive peaking units. The sidebar "How DR Programs Can Deliver Big Savings" describes PJM's savings during a summer heat wave. There is considerable difference between incentives paid in the DR market for participants to reduce load and the higher prices required by generators to operate peaking units or provide expensive ancillary services.

In effect, the implementation of DR introduces an element of price elasticity into what is otherwise virtually inelastic demand. Numerous studies, theoretical and empirical, have documented the dramatic benefits of demand elasticity in electricity markets.


Many challenges exist when integrating DR in wholesale markets. For example, managing the number of participating customers required to make a noticeable difference in a market the size of PJM requires substantial aggregation and automation. To make DR work, in practice, PJM provides market opportunities for demand resources in energy (day ahead and real time), capacity and ancillary service markets. With approximately 7000 sites and 5700 MW of available DR registered across 11 states, the estimated annual revenue for DR providers was more than US$180 million in 2008. Each DR transaction may have up to five different entities involved before it is completed. In addition, the market rules and procedures change frequently and are different for each wholesale service.

As the scale of DR programs has grown in PJM's market, the volume of transactions among market participants has grown correspondingly. Processing these transactions in real time has become more complex, resulting in additional requirements for market participants. Based on PJM's experience, three key issues have to be addressed in this complex environment to maintain the integrity of DR programs:

  1. Measurement and verification

    The actual quantity of DR "bid and delivered" must be measured and verified beyond doubt. If participants in the DR program do not have confidence in the amount of DR delivered, then that will result in protracted arguments and retroactive adjustments, adversely affecting future participation and tarnishing the overall perception of the program.

  2. Automated and flexible processes

    Participants need automated processes they can easily execute. Certain business processes may be fluid and need to be designed with future change in mind.

  3. Transparency

    Information access and transparency will minimize communication problems in a complex environment, which requires trust and accountability.

PJM, of course, is not alone in facing these challenges. It is merely among the first to encounter them. Other grid operators and many utilities are or will be facing similar challenges.


As the PJM experience demonstrates, for DR programs to work effectively, they must be integrated with existing computer applications of organized markets while complying with the changing market rules and protocols. Such computer systems are complex, and often inflexible and expensive to change.

Historically, grid operators have managed DR programs with semi-manual processes powered by a patchwork of homegrown computer applications that are not well integrated with existing systems. Such processes are typically inefficient and error prone, and have become bottlenecks in the growth of DR. They are also costly to maintain. A more comprehensive approach is needed that provides an integrated, automated and flexible DR management solution that also can automate the DR business processes end to end. Integration requirements mandate that the approach not entail substantial changes to existing applications and make incorporating changes to business rules easy.

PJM has been working to implement a more comprehensive solution for DR management. The solution is based on DRBizNet, which stands for demand-response business network software. This software was successfully demonstrated in 2006 in California, with funding from the California Institute for Energy & Environment. The solution leverages state-of-the-art systems integration and business process management technologies to manage DR in a flexible, scalable, reliable, secure and automated way.

Utility Integration Solutions Inc. (Lafayette, California, U.S.), the developers of DRBizNet, is currently working with PJM to implement the next generation of DR management systems in the PJM market. The solution is expected to be implemented in June 2009 to address the major challenges facing today's grid operators in managing DR. Expected benefits include:

  • Quicker participation of DR resources in markets

  • Improved transparency for all participants

  • Reduced time and administrative costs

  • Scalability for higher volume of participation

  • Simplified processes to provide more opportunity to participate for small resources.

PJM expects the new system to significantly reduce participation barriers related to data management, rule validation and response verification. The increased level of automation should significantly reduce participation costs for DR providers and enable substantial growth in DR participation.

Current Organized U.S. Wholesale Markets
RTO or ISO Installed capacity Transmission Population served (approximate)
CAISO 54,000 MW 25,526 miles 30 million
ERCOT 71,812 MW 38,000 miles 20 million
ISO-NE 32,000 MW 8000 miles 14 million
MISO 156,000 MW 93,600 miles 40 million
NYISO 44,851 MW 12,000 miles 19 million
PJM 164,634 MW 56,250 miles 51 million
SPP 50,392 MW 40,364 miles 4.5 million


There has been significant discussion about the cost effectiveness of demand-response (DR) programs in wholesale markets when capacity is in short supply. Major savings accrue when DR programs can substitute for expensive peaking power — or rolling blackouts, which are even more expensive. But, is there empirical evidence to support this?

The most impressive empirical evidence was provided during an extensive heat wave in the United States that affected major portions of the Midwest and the East Coast during the summer of 2006, breaking historical peak loads and setting record prices for peak power.

During the August 2006 heat wave, PJM Interconnection reported cost savings totaling US$650 million attributed to its DR programs. On just one day alone, Aug. 2, 2006, when PJM set a new peak load record of 144,796 MW, it reported DR savings of $230 million. These savings were based on incentives paid to DR program participants versus the cost of acquiring peaking generation, as determined by the market-clearing prices on that day. These (DR) voluntary curtailments reduced wholesale energy prices by more than $300/MWh during the highest usage hours.

In the case of PJM, where there are a number of markets for DR, participating customers or curtailment service providers can effectively bid load reduction into the wholesale market to be used during a peak demand episodes. These resources can be invoked when their price is less than buying peaking capacity or ancillary services during a heat wave or some other system emergency. The savings reported by PJM and others for reliance on DR are real and substantial.


The significance of price elasticity can be seen in the graph shown here. In most markets, demand exhibits some elasticity, represented by the downward sloping line D1, on the left side of the graph. If demand shifts from D1 to D2, for example because of a heat wave, the market-clearing price increases from P1 to P2.

Contrast this with the steeply rising supply curve on the right side of the graph, and the inelastic demand curve represented by vertical line D1. In this case, a slight shift in demand, from D1 to D2, would result in a significant price spike. When there is little or no demand elasticity, relatively small changes in demand can cause a very large spike in prices.

In practice, the aggregate supply offer curve in an organized market during a peak demand episode resembles the shape of the curve on the right side, with a near vertical tail. Making matters worse, the aggregate demand curve, in the absence of DR, tends to be highly inelastic, represented by the vertical lines. This is why demand-response programs are so effective in reducing price volatility when supply is constrained.

Andy Ott ( is senior vice president of markets for PJM, LLC. Following 13 years in transmission planning and operations at GPU Inc., Ott joined PJM and is responsible for executive oversight of its Market Operations, Market Strategy, Member Training, State Relations, Customer Relations and Performance Compliance divisions. He was responsible for implementation of PJM's wholesale electricity markets, including the PJM locational marginal pricing, financial transmission rights day-ahead energy market and capacity market systems. He has extensive experience in energy market restructuring, including electricity market design and implementation issues, and in power-system engineering applications. Ott is the U.S. representative and working group chair for CIGRÉ Study Committee C5 on Electricity Markets and Regulation.