Performance Contracting: Advice To Utilities

Gil Peach, formerly a program evaluator for Pacific Power and Light, is an independent consultant to several utilities.

A demand-side management program can unravel when expectations of utility and contractor don't mesh. A well-worded performance contract can keep all parties happy.

Editor's Note: Residential performance contracting is unfamiliar territory for most utilities. Agreeing upon what is to be delivered, how to evaluate it, and how to pay for it has been a matter of trial and error, often accompanied by disputes over differing expectations. Dialogue will increase the chance of negotiating contracts that ensure delivery of a quality product and a fair return to both parties. To promote such back and forth, this article and the one that follows (see p. 23) offer two perspectives on what performance-based contracts ought to include. One perspective comes from an experienced advisor to utilities, the other from the director of an energy services company.

We also recommend readers see these related Home Energy articles: "Bidding for Demand-side Management Performance Contracts" (Jan/Feb '91, p.33), "Moving Weatherization Agencies into the Utility Age," and "Contracting with WAP Agencies: the Utility Perspective" (Nov/Dec '91, p. 12 and p. 19, respectively).

Many utilities are using, or considering using, residential performance contracting to carry out energy conservation programs. Under this arrangement, an energy service company (ESCo) under contract to the utility agrees to deliver a stipulated amount of saved energy, for which the utility pays a set price per killowatt-hour (kWh).

Ideally, performance contracting should be good for the utility and ESCo alike. For the utility, it should decrease the financial risk of demand-side management (DSM) investment by, as its proponents claim, "turning conservation over to the free market." By transferring to the marketplace the responsibility for delivering savings, presumably the utility would avoid the need to invest in additional staff, materials, and management control systems and the need to build the expertise required to achieve savings cost-effectively.

Unfortunately, these benefits may fail to materialize because performance contracting also carries risks beyond those posed by the traditional utility-contractor relationship. Performance guarantees can be illusory. Payment plans that appear to put an ESCo at risk can actually ensure it will realize a profit. The need to monitor the retrofit process in order to protect against "cream skimming" and lost opportunities may oblige the utility to expand its staff and undertake new training programs. [Editor's note: Some conservation measures are cost-effective only when implemented in conjunction with other measures. When a contractor installs only the most cost-effective measures, a practice called "cream skimming," return visits by others to capture savings from more marginal conservation resources are less profitable.]

These risks stem from the fact that performance contracting is new territory. What sorts of payment and delivery terms end up being fair to both partners in the venture is a matter of experimentation, the results of which will become clearer as more utilities and ESCos strike mutually advantageous agreements.

Meanwhile, utilities can protect themselves by developing contract provisions that will reduce the risks besetting performance contracting at its present stage of development. The following contracting guidelines will help provide this protection.

Guidelines

Avoid front-loading. Don't pay for work until the savings have been measured. Many of the early performance contracting experiments were undertaken reluctantly due to orders from public utility commissions. Because the performance contracting market was not mature (and sometimes hardly existed), utilities thought they were obliged to help capitalize an ESCo infrastructure. Thus, a contract might call for the utility to pay in advance for the first year's planned savings. The ESCo would then reimburse the utility if measured savings fell below the targets set by the contract. A utility should only use this "front-loading" approach for a clear strategic reason, for example if the utility considers DSM a transient phenomenon. In this event, the contractor should pay back all interest earned on the money advanced by the utility.

Define baseline measurements cautiously. Be very careful in defining the baseline from which to measure savings, especially if the ESCo suggests calculating the baseline from several years of billing data on each house. Do not expect the ESCo to treat these data as a utility analyst or time-and-materials contractor would. Expect the ESCo to use the data to justify setting the most profitable-that is, the highest-baseline. A more prudent approach for a utility is to insist on a single year as the base year.

Avoid payments based on "treatment groups." Measure changes in the energy usage of treated buildings annually or in multi-year intervals. Be very careful if the ESCo suggests the creation of treatment groups based on months and quarters of work. Any complicated payment system based on treatment groups can easily cost the utility many person-months of effort to untangle if there is a dispute.

Provide for a waiting period between completion of installation and assessment of results. Let the project come to an end, and, if possible, let another year go by before finishing the assessment of results. ESCos often suggest measuring results as soon after the retrofit as possible, in order to maximize the value of the achieved savings. As time passes, some residents may decide to remove measures they don't like, and some of the effects of poor installation and materials have a greater chance of showing up. Unless the program is a low-cost/no-cost blitz in which public relations is as important as energy savings, the utility should wait to measure savings until a year or two after installation at the last site. From a utility's perspective, only reliable savings should be paid for.

Reserve control over inspections. By accepting "integrated packages," where the ESCo provides not only the audit and installation but also the inspection, financing, and verification, a utility loses control and risks prudence review by a public utilities commission. Do not accept conditions on when and where to inspect. Surprise inspections have a deterrent value. It may be necessary to concentrate inspections in locales where you suspect problems. Inspectors should always carry a camera and photograph every failed installation or broken measure. Full documentation with photos is powerful evidence and will make for quick, favorable legal judgments. (It's best to use a 35-mm camera with high-ASA color film. Polaroid instant photos are expensive to reproduce.)

Reserve control over reliability and verification studies. Send inspectors out three and six years after the measures, again with cameras. Many failures due to poor materials and improper installation will be visible by then. For example, "one size fits all" pipe wraps will have deteriorated significantly.

Do not agree to pay the full avoided cost per kWh saved. When a utility agrees to pay for all savings that can be acquired up to avoided cost, a certain proportion of those savings-the easiest ones to get-will actually cost less than full avoided cost. Paying the full cost for them is tantamount to handing over windfall profits to the ESCo.

Incorporate the use of control groups. If possible, measure net results against a control group to prevent the ESCo from being paid for savings that would have occurred anyway, say from "free riders," and to avoid penalizing the ESCo if electricity use is generally increasing. Realize that the ESCo and the utility have different interests in the measurement of results. An ESCo seeks to save the most kWh for the least investment. Moreover, the savings need only last through the final measurement. For this reason, the optimal saving for the ESCo is less than the utility would seek to produce in the same building, using a time-and-materials approach. The ESCo minimizes its performance risk at the expense of the utility's long- term interest in acquiring all cost-effective conservation resources at reasonable prices.

Preclude evaluation gaming. Do not accept any ESCo-proposed measurement plan that excludes cases where usage increases following treatment, or "negative savings." Dropping negative savings distorts the analysis and artificially raises average savings, providing a windfall for the ESCo (see Tables 1 and 2).

Beware of proposals to target high-use customers. If the ESCo requests that the utility supply a list of high-use customers to choose from, be careful how to interpret savings achieved. It's easy to "mine" high-use customers and bang out jobs that show, on the average, what one may be accustomed to seeing as reasonable savings for a utility-administered, time-and-materials contract program. But these are not reasonable savings for the high tail of the distribution. The ESCo can substantially underserve these customers and still be paid for substantial savings.

Prohibit the addition of units to the sample covered by the contract. If the ESCo proposes to "maintain" the reliability of savings by creating a contractual right to add units at their own initiative, consider finding another ESCo. Adding units creates the risks of unforeseen financial liability and erosion of conservation resources. Instead, insist on a guarantee of savings in the units assigned. Do not allow any language that opens the window for an ESCo to treat any residences except those reviewed and authorized in writing.

Use standard contract language. Assume the ESCo will seek contract language that allows it to wring maximum profit from customers and the utility resource plan. Follow the advice of the legal department to guard against fraud and ensure that data quality, inspection, and verification methodology will stand up to a regulatory commission's prudence review, which is not likely to occur until after the ESCo has finished the job and moved on.

Include interest earned by the ESCo in cost calculations. If the utility program requires a co-payment from the customer and if the ESCo offers to finance that co-payment, investigate whether the ESCo is acting as a finance company in disguise. If the total financed co- payment is large enough, the ESCo may be making more money on interest than on the profit per saved kWh. If this is the case, especially if customer interest equals or exceeds the nominal co- payment, the utility should rerun its cost models to factor in what the ESCo is earning in interest.

These 13 suggestions, based on lessons learned from real cases, will help control excesses. It would certainly be unfair to imply that most ESCo's are interested in making unreasonable profits at the expense of the utility and its customers by selling unreliable or illusory savings. Nevertheless, until field results help us better define a truly equitable relationship, any utility embarking upon residential performance contracting must exercise vigilance, control, and guidance over the process.

                           Measured           Measured        
                           savings         savings without 
       Home                 (kWh)        negative savings(kWh)
______________________________________________________________
        1                    100                 100     
        2                    110                 110     
        3                   -200    
        4                    150                 150     
        5                    200                 200     
        6                   -150    
        7                   -110    
        8                    100                 100     
        9                    100                 100     
       10                    200                 200     
______________________________________________________________
  Total savings              500                 960     
  from project

  Average savings             50                 137     
  per home 
__________________________________________________________
This simulation illustrates how the omission of negative 
savings can affect the calculation of savings for an ESCo 
and therefore how it will be paid by the utility. Negative 
savings occur when energy usage increases rather than 
decreases following an installation of energy conservation 
measures. Notice that dropping out negative savings nearly 
doubles the billable savings, 500 kWh versus 960 kWh.

                                 Average Normalized 
                              Annual Consumption (NAC)
                       ____________________________________
                          All households    Only households
                          including those    with positive
                       with postive savings     savings
___________________________________________________________
Pre-"installation"(kWh)       23,300            24,400
Post-"installation"(kWh)      24,500            22,200  
Savings (kWh)                 -1,200             2,200
___________________________________________________________