And all hell broke loose! That is what happened recently after a working paper published by researchers at UC Berkeley and the University of Chicago concluded that it is not cost effective for homeowners to invest in low-income weatherization energy-efficiency programs. For a sample of 30,000 weatherization assistance program (WAP)-eligible households in the state of Michigan, the study found a return on investment equal to -0.8% and that for the society at large to be -9.5%; both of which are needless to say fairly unattractive even when compared to almost zero rate of interest one gets for keeping money idle in U.S. bank accounts.
These results are contrary to the usual rhetoric of energy-efficiency supporters, who claim it to be the lowest-hanging fruit and cheapest action to save the environment from climate change. In its defense the NRDC staff pointed out other credible research such as the one conducted by Lawrence Berkeley National Laboratory in April 2015, which found an average cost of 3.3 cents/kilowatt-hour for residential energy efficiency, 5.5 cents/kilowatt-hour for commercial and industrial energy-efficiency measures, as compared to roughly 13 cents/kilowatt-hour of average cost of supply. They further point out that the unattractive results are due to the fact that the researchers have accounted for complete measure cost and not incremental costs alone (note that the incremental cost is the difference between the cost of an energy-efficient measure and its baseline equivalent). A blogger at the Environmental Defense Fund has offered that it is now the right time to measure and value energy efficiency by introducing new business models, instead of debating old program achievements.
Federal and state governments and policy makers have come a long way in setting up myriad of instrumentalities and administrative structures that provide funding for electric and gas energy-efficiency projects. Given this, one may think that the debate this paper has generated shall wither away in the coming weeks.
However, the key question here is not whether funding for the WAP be curtailed, but how well are policy makers prepared to learn from ex-post data and project experiences such that there could be timely course correction.
One of the primary reasons for difference in expectations versus reality is over estimation of project results based on engineering models. A case in point is performance of combined heat and power (CHP) plants in the United States, which receive support in the form of grants and tax incentives.
At Rutgers Center for Energy, Economics and Environmental Policy, we have analyzed actual performance of CHP plants (installed capacity more than 1 megawatt) over a period of 12 years. Within a sample of studied plants (located in the state of New Jersey), we observed that not all plants run at high capacity factors (CF, which is the percentage of actual plant generation to ideal plant generation, assuming the plant would have run for all 24 hours and 365 days).
In fact, about half of the existing New Jersey CHP plants run at capacity factor of less than 50 percent, which means that they were running for less than 4,380 hours a year. Further, the probability of plants to run at very high capacity factor (greater than 90 percent) is extremely low (0.03); while the probability to run at lower capacity factor (less than 25 percent) is comparatively high (0.27).
This is contrary to Economics 101, which would advocate maximizing the number of runtime to achieve output from the plant (electricity and thermal energy) at lowest cost. Further, this also means that the expected environmental benefits from these CHP plants (higher-efficiency plant and displacement of marginal coal generation with natural-gas plant) is not achieved in reality.
Results for other states and policy implications, along with a proposal for an alternative incentive mechanism have been explained in Utilities Policy paper.
Incorporating lessons from program experience is a key step in program design. Regulators and policy makers should therefore stress on collecting actual ex-post data and its analysis to circle reasons for failure or success towards achieving policy goals. Only then one can reasonably differentiate between hype and reality.