Back in July and also a couple of other times over the past two years, Mike has written here about the Jevons effect — when an increase in energy efficiency reduces the per-unit cost to the consumer of doing the energy-consuming action, moving her down along her energy demand curve and increasing her quantity consumed. This fascinating and nuanced issue, like so many things in energy policy, is a real phenomenon, but a very dynamic one. How big is the Jevons effect; how much of a rebound will there be from a given increase in energy efficiency? Paraphrasing Jevons’ neoclassical successor Alfred Marshall, answering that question starts from a comparative statics-partial equilibrium analysis, but ceteris is never paribus, so the model you use is likely to be wrong, and you’ll have a hard time capturing the actual dynamics accurately because the underlying system you are trying to model is not just complicated, but also complex (i.e., non-linear and non-deterministic).
The size of the rebound depends, among other things, on the price elasticity of demand for the energy (typically thought to be pretty inelastic), but the demand for energy is always a derived demand, so it depends on the demand for the goods and services into which the energy is an input. So, say, if I’m going to drive 10 miles a week no matter how much I spend per week on gasoline, an increase in energy efficiency will not have a rebound effect from my behavior because it’s not going to change. To complicate matters further, though, my demand for those goods and services is a function of both substitutes and complements, and is a function also of the context in which I am making my consumption choices. So at the margin the reduction in the per-mile cost of driving may, at the margin, induce me to drive slightly more because then I can group my errands more efficiently, weighed against the other alternative ways that I might be able to achieve what I want to achieve (in the case of transportation, alternatives like bus, train, bike, walking, scooter, etc.). Those preferences are sure to be quite heterogeneous across a diverse population, as are time preferences and discount rates, which can also change in a highly distributed way as people change their behavior based on their expectations of future environmental harm from current energy consumption. Plus, ceteris is never paribus in the sense that other technological and energy efficiency changes are likely to be occurring simultaneously with the one you are trying to isolate, which will confound the analysis.
But I digress (sorry). Like Mike, I’ve been thinking about the Jevons effect and have read Jevons, although I have not yet read The Myth of Resource Efficiency: The Jevons Paradox yet. A couple of weeks ago I also discussed the new Gayer and Viscusi paper from Mercatus on the consumer irrationality hypothesis and whether energy efficiency regulations actually do benefit consumers. Mike and I aren’t the only ones; Ken Green at AEI also took note of some Jevons effect work (referencing the Breakthrough Institute work that Mike mentioned in his most recent post on this) and the Gayer and Viscusi work on consumer irrationality, thinking of those two issues as pervasive fallacies in energy policy.
And then in this morning’s Wall Street Journal, Robert Michaels takes on the rebound effect in a short opinion piece, The Hidden Flaw of ‘Energy Efficiency’. In this piece, which draws on his recently-released Instituted for Energy Research study, The Rebound Dilemma, Michaels analyzes the direct and the indirect effects of energy efficiency mandates:
Higher efficiency reduces the cost of cooling. A family that once had only a single air-conditioned bedroom may now choose to install a central unit, and one that suffered in the heat may purchase its first one. Direct rebounds like these, however, are only the start of the story.
Technology that improves energy efficiency and reduces its cost means people can consume more goods and services that use energy—home electronics, appliances and the like. And of course, businesses will use additional energy making them.
One reason why I think research and debates are important on complex effects like the Jevons effect is that too often, energy efficiency mandates are naively offered up as a cost-effective way to reduce energy use, and therefore to reduce greenhouse gas emissions. We’re fooling ourselves (or, as Ken notes in his post, suffused with fatal conceit) if we think that such effects and relationships are going to be so straightforward and yield the desired outcomes predictably and reliably.
In the context of Mike’s earlier posts, I think Bob’s conclusion fits right in:
Rebound gives critics of regulation both philosophical and practical rationales for their views: Some object to efficiency standards on libertarian grounds and rebound research now tells us that many standards will fall short of their initial promise. But for the Breakthrough Institute in Berkeley, Calif., which gives primacy to climate change, rebound increases the urgency of introducing large-scale governmental management of both markets and technologies.
The growing body of research on rebounds means that both the left and the right must rethink their stances on energy policy. Some efficiency regulations may be worth their costs, but the existence of rebound means that the nation can no longer accept legislation to improve efficiency without further thought.
To that I would add another recommended reading: Richard Epstein’s Simple Rules for a Complex World.