This summer, the Environmental Protection Agency will announce its final rule for the Clean Power Plan. Created under section 111(d) of the Clean Air Act, the CPP will limit the amount of carbon dioxide that power plants can emit on a state-by-state basis.

States have the option of using rate-based standards (CO2/MWh) or converting those rates to mass-based equivalents. As states begin thinking about how they will comply with the Clean Power Plan, energy efficiency should receive considerable attention as one of the most effective options.

Under the CPP, energy efficiency is counted as electricity that would otherwise have been used. Under the rate-based standard, states add an estimate of their avoided electricity use due to efficiency programs to their total reported electricity demand. This creates an incentive for states to invest in efficiency, because it means their observed emissions rate per unit of electricity decreases, all else being equal.

If a state opts for a mass limit, the projected amount of electricity saved from efficiency programs is added back into the total electricity load. Because a state multiplies its anticipated load by its rate-based standard when converting to a mass-based equivalent, the more energy efficiency savings a state projects, the higher its mass limit will be.

In either case, states will be required to estimate their savings from energy efficiency. This means states will need to estimate how much electricity they would have used if efficiency programs had not been in place, which policymakers refer to as a “baseline.” But determining how much electricity would have been used is surprisingly complicated.

Take, for example, a homeowner who installs a 15-watt CFL light bulb in her living room. In Vermont, this switch is considered to save 71 kilowatt-hours per year. However, the same exact switch in California is estimated to save only 18 kilowatt-hours per year. This means that even if California and Vermont install the same number of CFLs, Vermont’s estimated electricity savings are four times higher than California’s.

How can these savings estimates be so different?

When utilities and states calculate baselines, the amount of electricity saved depends on a wide range of assumptions. These include factors such as free-ridership (the extent to which customers would have installed efficient lighting or appliances even without efficiency programs), the energy efficiency of the product that is being (in Vermont, the wattage bulb a CFL is replacing is assumed to be 48.7, but in California it’s 37.95), how many hours a day people use their lights or appliances (Vermont assumes an average for lighting of 3.4 hours, but California assumes 2.18), and others.

Many states also have “custom projects,” which have unique assumptions for each installed measure. Making matters even more complicated, not every state considers the same set of factors in the process of deriving its estimates. For example, Pennsylvania doesn’t account for free-ridership, though California and Vermont do. Put it all together, and the estimated impact of the exact same energy-efficiency measure can vary drastically from state to state.

The EPA acknowledges this issue in the draft CPP rule, noting, “many states with energy efficiency programs use different input values and assumptions” and that “this can result in significant differences in claimed energy savings values for similar efficiency measures between states and utilities, even when the same measure type is installed under otherwise identical circumstances.”

The EPA did not make clear how it plans to deal with these discrepancies. What seems likely is that the EPA will publish limited guidance for states to follow when reporting energy efficiency savings in state implementation plans. However, based on EPA’s language, this guidance will likely allow states considerable flexibility in calculating efficiency savings.

Should EPA decide to calculate efficiency savings itself, the agency will have to devise a fair methodology for doing so, which will likely result in reducing the level of claimed savings in some states. However, if states are tasked with estimating efficiency themselves, then even with published guidance, the EPA will still have to review and approve the states’ methodologies. This will require the EPA to account for the many different ways that states estimate savings from efficiency.

At the end of the day, the EPA should be prepared to answer the question: should a state that undertakes few efficiency measures be allowed to claim the same level of savings as a state that implements a dramatically larger number of efficiency measures? The answer to this question will have huge implications for how easily state targets can be met -- as well as how costly it is to meet them.

***

Robbie Orvis is a policy analyst at Energy Innovation: Policy & Technology, LLC.