The Energy Policy Act of 2005 requires FERC to publish an annual account of advancements in smart metering and demand response programs in the US, the latest of which was published in Dec 2012. It is a comprehensive survey of 3,349 “entities,” all but a handful considered “utilities” of one form, shape, or size. Over 1,900, roughly 60%, responded – not bad as survey participation rates go. The sheer number of active entities in this space is simply mind-boggling and may explain why it is difficult to get things done, whether it is demand response or anything else.
There are over 1,800 municipally owned utilities (Munis), over 800 cooperatively owned utilities (Coops), nearly 200 investor-owned utilities (IOUs), plus a sizeable number of federal and state power agencies, over 100 retail power marketers operating in jurisdictions where there is competitive retail such as in Texas, 11 curtailment service providers (CSPs) – entities that deliver negawatt savings during peak demand periods to whoever is willing to pay the going price – plus 7 organized wholesale market operators (ISOs and RTOs) spread across 50 states and the District of Columbia (DC). Each “entity” has its own motives and agenda and operates under different regulations, in some cases virtually no regulations, with vastly different priorities.
The survey results are summarized by region, utility type and customer class. In the case of demand response (hereafter, DR) results are further broken down by type of program implemented. It is a useful and timely survey.
Already, all but one of the top 10 states are above 50% in terms of penetration of smart meters by mid 2012, when the survey was conducted, 87% in the case of District of Columbia, followed by California at 70% – where virtual total penetration is mandated for the 3 large IOUs by the end of 2012, the deadline likely to spill into 2013 if not beyond (Table below).
While smart meters are far from universal across all states and various types of “utilities” in the US, recent trends suggest reaching 50% milestone by 2015-16 and much higher rates in many jurisdictions by the end of the decade.
Considerable scope is devoted to how much DR potential there is in the US and how it may be tapped. After all, the business case for investing big money in AMR is usually not compelling without the ability to manage, modify, shift or clip peak demand. The starting point in FERC’s survey is a definition for DR, which FERC defines as “changes in electric use by demand-side resources from their normal consumption patterns in response to (a) changes in the price of electricity, or (b) to incentive payments designed to induce lower electricity use at times of high wholesale market prices or when system reliability is jeopardized” (“a” and “b” added for emphasis to quote).
Following the above definition, FERC classifies DR programs into 2 broad categories, incentive based and time-based, each with many sub-categories, even though the distinction may be artificial – all DR programs are incentive-based – i.e., customers respond to incentives, one way or another, which may be in the form of variable prices.
The big surprise, to the extent that there is any, is that FERC’s latest survey puts total reported DR potential at 72 GW, a 25% rise from the 2010 survey – that is more than 9% of US peak demand – an astonishing figure. The number, of course, is a fictitious one since demand peaks on different parts of the network at different times. No matter, it is a big number and it suggests that DR resource can save the day and keep the network from collapsing if you can indeed get it and count on it when you actually need it. Actual DR – what utilities or organized market operators have actually been able to deliver when needed – typically falls far short of such targets.
Among the US organized wholesale market operators, PJM and Midwest ISO (MISO) appear to be more bullish in estimating how much DR potential there is, while other markets, notably California Independent System Operator (CAISO) put the figures at negligible levels. It is a puzzle, which FERC report does not adequately cover. Some of these discrepancies may be definitional while some may be explained by lack of operating DR auctions within the existing wholesale markets. As is always the case, one has to read the fine print, and even that is not always enough.
In case of PJM and MISO, for example, a big potential for DR is identified under “load as a capacity resource” and “demand bidding & buyback” – in other words negawatts that can be shed by or bought from participating customers in response to given price signals and other schemes that have been incorporated into wholesale auctions managed by these market operators. Clearly, organized wholesale markets in the US operate differently, some have incorporated functional DR auctions with significant volume of DR bidding while others have not or are in the process of establishing such markets.
A chapter of the report is focused on what FERC has attempted to do since the passage of EPAct 2005 to foster the development of DR, most notably Order 719 in October 2008 and Order 745 in March 2011, the latter previously covered in this newsletter. While progress has been made following these and other FERC directives, much more can and needs to be done.
According to the 2012 FERC survey, the following 4 programs account for 80% of the total reported potential peak reductions in the US – graph below shows the relative contribution from each progam:
• Load as capacity resource;
• Interruptible load;
• Direct load control (DLC); and
• Time of use (TOU).
Clearly, if you want to start a serious DR program, these would be obvious places to explore. As for what has actually been achieved to date, FERC reports a little over 20 GW of peak demand reductions from “demand response resources” in 2012, “representing use of 31% of the total reported peak load potential.”
As for the future, FERC survey identifies key DR programs planned for the period to 2017, mostly focused on the same areas that appear to have been successful in the past, such as DLC and TOU.
When it comes to time variable pricing, FERC survey exposes the near total lack of progress to date. Residential time-of-use (TOU) programs, for example, were offered by 151 among the roughly 1,900 “entities” who responded to the survey. Only 28 entities reported offering real-time pricing (RTP) in 2012 suggesting substantial room for improvements, a generous way to put it.
Implementation of TOU and RTP schemes for residential customers is limited to a handful of active utilities in jurisdictions with favorable regulatory support. The two large utilities in Arizona, Arizona Public Service (APS) and Salt River Project (SRP), for example, report that a third of their residential consumers “have voluntarily chosen to participate in” TOU programs.
California may become the proving ground for testing the popularity of dynamic tariffs with offerings by the 3 large investor-owned utilities by the end of 2012. Critical peak pricing (CPP) has been the default option for large commercial and industrial (C&I) customers in California for some time. The debate has shifted to applying the CPP default tariff to all residential consumers – pending the resolution of a number of remaining regulatory and legal challenges currently before the California Public Utilities Commission.
Regulators in California, as many other jurisdictions, are weary of consumer backlash that may follow switching large numbers of residential consumers to dynamic pricing because the volatility of these rates may prove unpopular. In addition to California, FERC lists Maryland, Arkansas, Oklahoma, Illinois, Idaho, Colorado and Connecticut as potential candidates for widespread use of RTP, CPP or similar time-variable rates.
Despite the progress achieved since 2005, FERC identifies several key barriers to future implementation of DR including:
• Limited number of retail customers on time-based rates;
• Measurement and cost-effectiveness of (load) reductions;
• Lack of uniform standards for communicating DR price signals and usage information;
• Lack of customer engagement; and
• Lack of DR forecasting and estimating tools.
In short, despite its great potential, DR remains broadly under-rated, and not just in America. The power industry is still dominated by those who view load as a “given” and generation as what has to be adjusted to meet the variable load. These same people are greatly troubled by intermittent generation from renewable resources. Ironically, no one seems bothered by intermittent load, variable or unpredictable demand, or spikes in peak demand due to weather or cyclic customer behavior.
The chairman of FERC, Jon Wellinghoff, is among a handful of people who have discovered the absurdity of the outdated perspective of treating demand to be whatever it wants to be and ramping generation up and down to match it, no matter the costs. Why shouldn’t we adjust demand when and if it is cheaper to do rather than generation?
Time variable pricing is one way to go, as are multiple other ways of managing load. It may be easier, more environmentally benign, and potentially a lot cheaper.
This post is extracted from EEnergy Informer, February 2013 issue.