about the energy implications of the behavior if it did occur; and so on. Finally, because most of the evaluation studies were commissioned by organizations responsible for the programs being evaluated, who have a stake in the outcomes of the research, the assumptions, methodologies, and conclusions may have been subject to political pressure, with unknown effects. For example, the Department of Energy commissioned an evaluation of a 1979 program in which it distributed water flow limiters and conservation information t o all households in New England (Booz, Allen, and Hamilton, 1980). It found an annual savings of $26 in fuel for each dollar spent by the program. But a review by the General Accounting Office (GAO) (1981) criticized the evaluation for making assumptions that may have vastly inflated the estimates of energy and money saved. The Department of Energy subsequently reestimated energy savings to account for one of the GAO’s major criticisms; this reestimation reduced the estimated savings by one-third. The remaining GAO criticisms did not receive response, but a defensible case can be made that even the revised estimate is several times too high. Although on the surface the program seems to have been cost effective, this conclusion cannot be drawn from the evaluation with much certainty.
Because of the self-interest of most of these conducting evaluation studies, it would be naive to take their conclusions at face value. But what of the general issue: how valuable is energy information? A careful analysis of the methodologies and data convinces us that some energy information programs have been effective in terms of producing economically justified energy savings by consumers. These programs probably include the Energy Extension Service, which was evaluated by the Department of Energy (1980), the automobile fuel economy information program of the Department of Energy (see McNutt and Rucker, 1981), and a few of the programs aimed at promoting major home retrofits (see Hirst, Berry, and Soderstrom, 1981).
Other lines of evidence also suggest that under optimum conditions, energy information programs might be important. It is obvious, for example, that automobile manufacturers know that fuel economy sometimes sells cars: their advertising has often said much more about miles-per gallon than is mandated by regulations. In fact, it has been estimated that the automotive industry spent upwards of $100 million on advertising fuel economy in 1979 (Pirkey, McNutt, Hemphill, and Dulla, 1982). While fuel economy has been less of a concern of automobile advertisers since then, in 1982 it was still one of the major foci (Pirkey, 1982), reflecting the concerns expressed by purchasers of new cars. At the same time, because manufacturers are not trusted as a source of fuel economy information (Pirkey, 1982), information from a standardized fuel efficiency test has an important function—if it is effectively presented.
Overall, most information programs seem to do less than their creators