Climate Change and the U.S.: Is the Environmental Protection Agency under-pricing carbon?

Frank Ackerman

What will the US do about climate policy? The ongoing paralysis in Congress makes it clear that failure actually is an option. For those who long for success, the best hope may be EPA regulation of carbon emissions. Yet far from setting an ambitious new course, EPA may be following the worst of traditional economics – effectively arguing, in opaquely technical language, that climate change isn’t such a big deal, so the right policy is to do almost nothing.

The story began when several states and environmental groups filed a lawsuit, Massachusetts vs. EPA, to force EPA to regulate greenhouse gases. In 2007 the Supreme Court ruled for the states, ordering EPA to determine whether greenhouse gases harmed the environment and public health. The Bush EPA ran out the clock, saying almost nothing on the subject, but Obama’s new EPA administrator, Lisa Jackson, swung rapidly into action. Last April, EPA declared greenhouse gases to be pollutants that endanger public health and welfare, qualifying them for regulation under the Clean Air Act. A proposal was published for comment in September; my comments were among thousands that EPA received. A final ruling is expected soon.

Buried in the 336-page EPA proposal is a short section on the “social cost of carbon” – the dollar value of all damages caused by emitting one ton of CO2. There is, of course, no such number. Many damages to human life and nature are priceless, as I’ve argued elsewhere. In a cost-benefit analysis, however, the price tag attached to damages is all that matters. The higher the price, the more it’s worth doing to avoid them. That’s why the strength of US climate policy may depend on EPA’s social cost of carbon.

Every $1 per ton of CO2 is about a penny per gallon of gasoline, so $5 per ton would be a trivial price incentive of 5 cents a gallon. At $50 per ton, or 50 cents a gallon, you’d start to notice. An increase of $500 per ton, or $5 per gallon, would put us in the realm of gas prices in many European countries where people buy smaller cars and use public transportation a lot more than we do.

$500, though, isn’t in the running. In the September proposal, EPA offered a range of values from $5 to $56. It sounds to me like the high end was included to mollify critics, while the low end is what EPA’s economists prefer.

How could US climate policy shrink down to a mere nickel per gallon of gas? There are three steps in the construction (or constriction) of EPA’s social cost of carbon estimates; each of them draws on the narrowest and most dated version of climate economics. (For details, see my comments on the proposal.)

First, EPA only considers peer-reviewed economics articles – specifically, only the latest estimates from three simple models, DICE, FUND, and PAGE. The peer review criterion excludes the Stern Review, the massively researched, widely discussed study commissioned by the British government. Stern’s social cost of carbon estimate was $85 per ton, well above EPA’s recommendations.

The focus on three models erroneously suggests that they represent the state of the art. In fact, each of the three contains hidden biases which serve to minimize climate damages or even imply spurious benefits from warming. Meanwhile, many other models and studies contain valuable additional information about climate damages.

Second, EPA considers discount rates of 3% and 5% for future damages. EPA’s proposal quotes, but ignores, an Office of Management and Budget recommendation to consider discount rates below 3% for intergenerational analyses. EPA cites five academic sources on discount rates, four of which recommend or use rates of 2% or lower. The lower the discount rate, the greater the weight of future damages, and the higher the social cost of carbon will be.

Finally, EPA mentions but fails to incorporate recent research by Harvard economist Martin Weitzman on catastrophic climate risks. As Weitzman demonstrates, the value of reducing catastrophic risk can, technically speaking, be infinite – a paradox that calls for a precautionary approach to climate policy.

In the best of all possible worlds, if Congress came to its senses, this debate might be unnecessary. But I wouldn’t bet the fate of the earth on it. Controlling climate change is an experiment that we only get to try once. To have a good chance at a happy outcome, we’ll need to start with a much bigger number for the social cost of carbon.

You can follow this discussion, and related work on climate economics, on Public Goods, the new blog just launched by my colleague Liz Stanton.