The Prius Paradox Paradox: Rebound Effects Are Relative

Walter Borden

WILL money saved from using clean technology simply be spent on using    more energy? Jevons paradox (or the Jevons effect) is named for economist William Stanley Jevons.  In the 1860’s, he observed that technologically driven increases in the efficiency of coal-use increased coal consumption in a wide range of industries. Counter-intuitively to some, he argued that technological improvements could not be relied upon to reduce fuel consumption. Buyers simply use the savings to buy more energy. Such rebound effects as a batch of recent research reveals, are at work in energy markets yet are often overdetermined and misunderstood. Their occurrence suggests the need for carbon taxes in order to price environmental risk in energy costs. The basic logic of such taxes was sketched out in the 1920’s by another economist, Arthur C. Pigou, as the Pigovian Tax. He argued that landowners who allow their rabbits to overbreed and spill over to neighboring land, therefore damaging  crops, have a financial responsibility for the damage. Such activity, often uncorrected by markets, is seen as a market failure. So its remedy is a tax or law to protect the rights of neighboring landowners.

Interest in both is keen among policymakers, thinktankers, bankers, and the general public as the tension between energy demand  and supply increases. Pollution, global warming, declining oil reserves, and increasing demand for energy in the neoliberalized global marketplace underlie both the interest and the tension.

To the extent that they are at work, Jevons rebound effects in a system vary based on the scale of the market considered. For example Richard York of the University of Oregon finds:

A fundamental, generally implicit, assumption of the Intergovernmental Panel on Climate Change reports and many energy analysts is that each unit of energy supplied by non-fossil-fuel sources takes the place of a unit of energy supplied by fossil-fuel sources 1, 2, 3, 4. However, owing to the complexity of economic systems and human behaviour, it is often the case that changes aimed at reducing one type of resource consumption, either through improvements in efficiency of use or by developing substitutes, do not lead to the intended outcome when net effects are considered.

Dr. York’s work appears to reveal an instantiation of the effect.  Across most nations of the world, developed and developing, he reports an average pattern, “…over the past fifty years is one where each unit of total national energy use from non-fossil-fuel sources displaced less than one-quarter of a unit of fossil-fuel energy use. When looking at electricity specifically, the displacement of each unit of electricity generated by non-fossil-fuel sources is less than one-tenth of a unit of fossil-fuel-generated electricity.”

These conclusions put a useful empirical foundation under recommendations found in’s clean energy innovation study: meaningful suppression of fossil fuel consumption requires adaptation of mainstream energy policy. Also looking at the international scale, published a chart this week titled The mind-boggling rise in Asian coal consumption shown as Exhibit 1.

Chinese Coal Consumption vs. Developed World
Exhibit 1: Chinese Coal Consumption vs. Developed World. Source:

Coal going unconsumed in the U.S. is being burned with little scrubbing in China and India, further arguing for the need to decarbonize via international agreements. Liberalized trade (neoliberalism) needs alignment with a flow of trade that balances externalities – pollution – created by exchanges of resources and capital. This also complements York’s finding: shifts to renewables will be inconsequential if the total decarbonization rate isn’t decelerated, that is, if amounts are merely shifted from one market to another.

When Rebound Effects Are Perceived But Not Found

Then there is the contention of the paradox at work in driver behavior popularized as the ‘Prius Effect” in sources such as Conundrum and the Wall Street Journal. Their argument is that Prius owners drive more and thus erase their net carbon and energy savings for the system. However, the work of Ken Gillingham of Yale University and analysis from CO2 Scorecard show Prius owners rack up comparatively the same vehicle mileage as non-Prius owners.

This Prius Fallacy has a dual premise: Prius drivers drive more because they are paying less for gas, and/or they use their savings on carbon-intensive goods and activities.

Gillingham’s micro-dataset on personal automobiles contains information – further analyzed by Thinkprogess – which refutes premise one as the scale of the consumer. The plot in Exhibit-2 shows no significant difference in Vehicle Miles Traveled (VMT) by Prius owners vs. the rest of  California’s drivers. (For those interested in statistical details on the data and diagnostic regression Thinkprogess’ analysis is worth a good study). Prof. Matthew Kahn of  UCLA writing in the Christian Science Monitor reinforces these conclusions.

So in these cases when consumers switch from conventional cars to a fuel-efficient hybrids a meaningful reduction in gasoline consumption – up to 430 gallons per year for an owner who switches from an SUV— is also observed.

As for premise two, money saved on fuel is spent on carbon intensive purchases, there is a scarcity of data.  It’s a safe bet though that a great many Prius drivers may simply piggy bank their savings, therefore no immediate rebound in carbon emissions, or they invest in a cleantech venture (a case of negative rebound). And, indeed, some may use it to fly to Cancun for the weekend.

There is a lack of published study on the purchasing habits and behavioral economics of hybrid drivers and conventional car drivers. Further data and analysis will be greeted with interest. Most likely, whatever the breakout, the average Prius driver isn’t driving a catastrophic rebound.

Some argue that a dollar spent on energy has a two-or three-fold effect. Thinkprogress goes on to point out:

It is hard to see how that is possible—if up to 8% of our GDP accounts for energy use, it already includes the energy component of the rest of the 92% of the GDP. Both direct and indirect energy use within the economy are included in the 8% share. Adding a two- or three-fold multiplier on top of that would lead to phantom accounting. There is little in the way of a solid theory or verifiable empirical estimate that proves the existence of multiplier effect in this particular context. And even if we give the benefit of the doubt to the proponents of rebound and assume the existence of multiplier effect, the share of the $1500 savings will on average account for rebound worth $201 (13%) and $301 (20%) for two- and three-fold multipliers respectively….So, at her worst, an average Prius driver is re-injecting $101-$301 of her $1500 savings into energy use from fossil fuel. And that’s near the worst-case scenario. If Prius drivers don’t drive more than conventional drivers, and if Prius drivers must (lacking evidence to the contrary) be considered among average American consumers, where’s the real fallacy?

So it seems, there are many more questions around the stipulated Prius Paradox than the popular press examines.

The world’s energy economy is a complex, adaptive system

While the Jevon’s paradoxical effect is absent in the case of Prius drivers in one of the world’s largest, emissions regulated, and ecologically conscious economies, it is found at different energy economic scales. The real but inconsistent functional evidence of the effect opens up several questions in need of further exploration.

  • Elasticity in energy market economics: The paradox can be seen as applicable when demand for the product is relatively elastic. But, isn’t the demand in this case relatively inelastic? People are not going to drive twice as much because their car uses half as much gasoline per mile, and they are not going to heat their house to 90C because they have added insulation.

    Exhibit 2: Comparative Density Plot -- Miles Travelled Prius v. All Cars, California. Source:
  • Behavioral considerations: Proponents of the paradox like Owen and analysts at the Breakthrough Institute mistake the effects of economic growth for those of efficiency. Doesn’t the rebound effect apply mostly to the changes in energy use directly attributable to efficiency only? Larger houses, bigger cars – these or other choices which are driven by technology trends, greater personal wealth, attitudes, and preferences should be distinguished from the rebound effect.
  • Macroeconomic considerations: Net savings contribute to a macroeconomic re-spending effect which may be related to the change in the marginal cost of delivering energy services when efficiency technologies are installed. Do grid costs drop with clean energy?
  • Externalities and market failures: Less carbon pollution leads to lower environmental costs of the real damage to the commons in the form of polluted air, deforestation, acidified oceans, and displaced populations.
  • In information technology: The electrical efficiency of computing (the number of computations that can be completed per kilowatt-hour of electricity used (see Dr. Chis Toomey’s very interesting and informative blogging on the subject) has also doubled every year and a half since the dawn of the computer age. Its not clear that these efficiences are conserved or idled. Likely they are utilized, but how to measure and understand the implications of this relation is a matter of increasing importance and research.

A common report from home retrofittings is that the owners can lower thermostats due to less draft and less cold walls. Such anecdotal evidence is buttressed by the well documnted ~38% decrease in energy use for the the Empire State Building after super efficient windows and retrofits. People simply reported being more comfortable.

In the specific case of the Prius:

  • Cost of ownership: Do owners spend most of the savings in energy costs paying for the higher initial cost of buying a Prius rather than a conventional car?  Is the upfront cost of improving energy efficiency basically equivalent to savings in energy expenses?

Motivated cognition and confirmaiton bias in attribution of Jevon’s paradox must be addressed as well. Do some observers use the paradox to unintentionally discourage energy conservation? It does appear that some actors use the presumed paradox to argue for the reversal of policies that develop markets for renewable energy because their economic interests derive from fossil fuel marketing. This range of data contradicting the precept is scarcely reported in mass media and among institutional observers, and so the narrative of the Jevon’s paradox at times trumps evidence, suggesting a bias.

If the paradox holds, then notions of degrowth and low-growth economies may need to be reconsidered and perhaps abandoned, i.e. slowing growth is impossible as any efficiency gains are zero sum among markets. This argument is popular with U.S. Chamber of Commerce.

Yet here again, some conventional wisdom seems to fail. One can think of very plausible scenarios where the economics of energy efficiency are better than what the rebound effect proponents describe. Assuming a positive savings rate (as most models of capital accumulation do) efficiency is always partially saved to some extent. This lowers the cost of capital as its benefits accrue. Given that since both efficiency and renewables are capital intensive, efficiency could conceivably further lower the costs of both.

Rebound effects vary, break down when localized, and behavior and trends influence them.

The media and thinktanks often decontextualize and oversimplify instances of Jevon’s paradox. Owen’s Rebound are not found in the case of Prius drivers in one of world’s largest car markets. Given economic preferences, values, and worldviews, Prius owners may well actually drive less than others with similar wealth, location, and age. Additional assertions of rebound effects with efficient refrigerators cannot be confirmed either.

Yet, recent work does indeed find rebound effects emerging at the global scale.

Is this then a paradox within a paradox? That’s unclear. But, what is clear is that energy consumption patterns are not always as deterministic and elastic as some analysts suggest. With this in mind technological innovation needs integration with policy innovation – not separation from it. This can only be achieved with a greater understanding of energy consumption behavior. When the demonstrated bias in favor of the energy industry in the media is taken in account, reports on market behaviors around clean energy merit added scrutiny.

Jevons paradox and associated effects do occur though they are localized and overdetermined. Deliberate changes in economic behavior and cultural values, for example, often counter such effects at the indivdual scale. Aggregate costs of capital for society are lowered in many cases by efficiency. However, owing to the complexities of human systems and behaviour, it is often the case that changes aimed at reducing one type of resource consumption, increase it in other channels. Nevertheless, improvements in clean energy and efficiency do not always have fixed outcomes of net increased usage.