Energy efficiency has become a pillar of government policy in efforts to reduce greenhouse gas emissions and improve energy supply. In the United States, it is a fixture of the Obama Administration’s Climate Action Plan. The European Union, meanwhile, has an aspirational goal of increasing bloc-wide energy efficiency by 27%. India, for its part, is hoping to save as much as 8.9 billion dollars (US) by overhauling its markets for energy-intensive appliances.
Why is energy efficiency such an attractive platform for countries? The oft-cited “McKinsey Curve” for the industry cost of greenhouse gas reduction exemplifies the answer. The McKinsey analysis suggests that, against a business-as-usual prediction of 70 Gigatons of carbon dioxide-equivalent (GtCO2-e) emitted globally in 2030, approximately 11 GtCO2-e can be abated at negative cost through energy efficiency upgrades. If true, these upgrades would be a “win-win”: consumers privately profit from energy savings, and society at large benefits from mitigation of climate change and air pollution. As suggested by two economics researchers, the McKinsey analysis (which is not alone in its conclusions) essentially implies that “the ground is littered with $20 bills that energy consumers have failed to pick up.”
But that’s just it – if energy efficiency is a “free lunch”, then it must be a free lunch that we are continually declining. Why would we do such a thing? Economists have come up with a number of explanations. Market failures, for instance, could be a barrier to investment in energy efficiency: individuals may not have adequate information to identify profitable investments; landlords may not want to upgrade energy efficiency when their tenants are the ones who actually save on energy costs; or consumers may not have access to the credit required to pay upfront investments. Just as easily, our own behavioral anomalies could explain low investment. Chief among these is our tendency to undervalue the future: even if that super-efficient air conditioner (or refrigerator; or car) is going to save us money over the long haul, the price tag on the appliance itself sure is hard to swallow.
All of the above are problems that can, in principle, be resolved by policy. For example, the government could either provide information to consumers on the performance and expected savings of energy-using goods, or mandate suppliers to do it themselves. This might help correct the inertia and present bias of consumers. However, the purported “energy efficiency gap”, as it has been called, could also be explained simply by measurement error. There may, for instance, be “hidden” costs that analysts do not count but consumers nonetheless experience (think of the inconvenience of undertaking renovations). Alternatively, our “engineering estimates” of the monetary savings from efficiency upgrades may be too high. If our forecasts of the net benefits of energy efficiency are wrong, then it is likely that our policy emphasis on it is also overdone.
This is exactly what has been found in some instances. Consider the Weatherization Assistance Program (WAP) – the U.S.’ largest residential energy efficiency program. WAP provides low-income households with home improvements like furnace replacement and attic and wall insulation, at zero out-of-pocket cost. Two recent studies of WAP’s take-up and impacts illustrate the challenge of energy efficiency. In one of these, a random subset of eligible households were actively encouraged to apply and were offered application assistance. Only 6% of these “nudged” households actually participated in WAP, and only 1% did so in the control group. Moreover, the ultimate savings on energy costs from those who did participate were much lower than predicted by engineering estimates. The second study, meanwhile, focused exclusively on WAP participants and provided information on how to interpret their energy bills as well as specific energy conservation practices. The authors again find the engineering estimates of energy cost savings to be overestimated. They also find that their behavioral and informational interventions drive reductions in energy usage at much lower cost than the efficiency upgrades subsidized by WAP. In some cases, apparently, it is better to first focus on changing behavior than to try to drive investment in new technology.
WAP is just one program, affecting one specific set of energy-using goods, in one particular country. There are now studies of energy efficiency in various markets, from residential utilities and home appliances, to automobiles and manufacturing processes. What is the upshot? In the words of two economists, “the apparent energy efficiency gap has multiple explanations whose relative contributions differ across both groups of energy users and types of energy uses”. As usual, there is no silver bullet for, nor one single explanation of, the energy efficiency gap. That means “blunt” policymaking is not going to work very well. Instead, policymakers need to understand the particular market and its consumer base, and design policy based on what we know from experience and analysis.