ernment Performance and Results Act (GPRA) submission (NREL, 2004). It can be used to generate input data for subsequent use in calculating benefits in a reduced forms model.

The relatively simple calculating procedures outlined above require future projections of basic energy information, in particular of supply and demand for various forms of energy and of marginal costs and/or prices. Such basic information generally can be developed only with a coherent model that projects possible equilibrium conditions for future years. Internal consistency among price and quantity projections for a given scenario, and between scenarios, normally requires use of a coherent model of the energy system. The committee believes that NEMS does provide such projections of prices, marginal costs, and supply/demand quantities.

Developing these net benefit calculations will therefore rely on the use of NEMS only indirectly, through its internally consistent projections of prices, marginal cost, and supply/demand quantities. The procedures recommended by the committee do not anticipate use of the NEMS directly to calculate net benefits. Such procedures can easily be implemented using simple spreadsheets, with particular parameters (such as projections of prices, marginal costs, and supply/demand quantities) taken from a NEMS equilibrium.

Several other calculating procedures, while superficially useful for estimating benefits, are not appropriate for the measurement of economic benefits and should not be used as net benefit estimates when estimating energy efficiency advances: (1) net change in the cost of particular energy commodities for individuals or groups of energy users, (2) net change in the per unit cost of energy commodities for individuals or groups of energy users, or (3) the aggregate change in the total use of energy.

One could measure for individuals or groups of energy users the net change in the cost of particular energy commodities resulting in energy efficiency advances. Note that this measure does not correspond with the impacts diagrammed in Figures 3-4 through 3-6. A simple example illustrates the fallacy. Assume that the elasticity of demand for energy is −1.0, so that a 10 percent reduction in the marginal cost of energy services would lead to a 10 percent increase in the use of such energy services. In that case, in response to a 10 percent reduction in the average or marginal cost of energy services, the energy use would increase by 10 percent, leaving unchanged the product of average cost and usage quantity. But because the product of average cost and usage quantity is simply the total cost of purchasing the energy services, the total cost of purchasing energy services would not change in response to a decrease in the marginal cost or average cost of energy services.

Similarly, the total cost of purchasing energy services would not change in response to an increase in the marginal cost of energy services. However as Figures 3-4 to 3-6 have illustrated, any and all such reductions in the average cost of energy services stemming from energy efficiency advances would have positive net benefits. Thus the first measure—the net change in the cost of particular energy commodities resulting in energy efficiency advances—is not an appropriate measure of net benefits.22

One can illustrate this with the numerical examples given for the lighting program. In the initial situation, total energy services would be 5,475 trillion lumen-hours at a cost of $0.0000008 per lumen-hour, or $4.60 billion. After the increase in efficiency of the lights, consumers would use 6,022 trillion lumen-hours of energy services at a cost of $0.0000004 per lumen-hour, or $2.41 billion. This would give an estimate of benefits equal to $2.19 billion, in contrast to the actual benefits of $2.30 billion.

In a situation where the consumer price exceeds the marginal cost of energy, the error would be greater. Here we assume that the consumer pays a price of $0.12 per kilowatt-hour. In the initial situation, total energy service would be 5,475 trillion lumen-hours of energy services at a cost of $0.0000024 per lumen-hour, or $13.14 billion. After the increase in efficiency of the lights, consumers would use 6,022 trillion lumen-hours of energy services at a consumer cost of $0.0000012 per lumen-hour, or $7.22 billion. If one were to calculate the benefits based on the cost paid by the consumer, the estimate would be equal to $5.92 billion, in contrast to the actual benefits of $2.96 billion. More appropriately, if one were to calculate the benefits based on the total cost of energy to society, a cost of energy of $0.04 per kilowatt-hour would give an estimate of benefits equal to $2.19 billion (exactly as calculated above), in contrast to the actual benefits of $2.96 billion.

The second measure—net change in the per unit cost of energy commodities for individuals or groups of energy users—is also an inappropriate measure of net benefits. These changes in the per unit cost of energy commodities typically are simply the result of price changes. But such price changes lead to benefits for the some members of the economy and equal but opposite net benefits for other members. Summed over all members of the economy, such price changes lead to no net benefits, so the per unit cost of energy commodities is not an appropriate measure of net benefits.

The third measure, the aggregate change in the total use of energy, is also inappropriate. As shown above, if the marginal cost of energy commodities is less than the marginal value of those commodities, increases in the total use of those energy commodities would lead to increases in net benefits. On the other hand, if the marginal cost of energy commodities is above their marginal value, increases in the total use of those energy commodities would lead to reductions in net benefits. Therefore, measurements of the total use of energy commodities will not provide a good measure of net economic benefits.

22  

NEMS, however, measures just this and so does not produce a benefits measure that is consistent with the committee’s definition.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement