Cars and trucks sold in the U.S. will have to be a little more efficient, according to new Corporate Average Fuel Economy (CAFE) standards released today by the EPA. CAFE standards are our standard tool for promoting vehicle efficiency, and have been the subject of past research here at RFF.
The new standards are largely a product of a compromise between states, the federal government, and auto manufacturers last year.* They are also the end product of the Supreme Court’s Massachusetts v. EPA decision requiring the EPA to address impacts of greenhouse gases under the Clean Air Act. What reading I’ve been able to do so far reveals no big changes over what was proposed last year. The requirements appear relatively modest: essentially the existing requirement of 35 mpg fleet average fuel economy by 2020 is moved up to 2016 and increased by 0.5 mpg. As I’ve written recently, these standards will trigger other Clean Air Act regulation—but let’s focus on the direct effects of the standards for now.
That apparently small change can have a big impact when you consider how many cars and trucks there are in the U.S. and how long these vehicles will remain on the road. The EPA claims that the standards will reduce greenhouse gas (GHG) emissions by 960 million metric tons and cut U.S. auto emissions by 21 percent (over business-as-usual) by 2030. The EPA also estimates that increased up-front vehicle costs of about $1000 will be offset over the course of each vehicles’ life by reduced fuel costs, resulting in a savings of about $3000.
That’s good news for the environment, and good news for consumers, right? The auto industry is (at least for now) OK with the new standards, and the environmental community is generally happy as well. I think the positive spin is broadly correct—we’re certainly better off with stricter CAFE standards than we would be without them—but I’m skeptical about the size of the benefits estimated by the EPA. Performance standards, and in particular efficiency standards, are flawed policy tools—emissions benefits may be lower, and costs higher, than with the best alternative: a carbon price.
The largest problem with efficiency standards is that they encourage increased use of whatever is being made more efficient. If your car is more efficient, it’s cheaper to drive it, and you’ll probably do so more often (and for longer distances). You might even move farther away from work or make other choices that increase your fuel consumption (but not cost, remember that you’re green efficient now). This is great for you—you get increased utility from driving more—but your vehicle emissions won’t go down as much. Even if you “save” more money over the life of the car, the added cost per unit of emissions reduction goes up. Other social costs, like traffic congestion and increased risk of accidents, go up as well. This is called the “rebound effect,” and is the subject of significant research among economists.
The EPA is aware of this effect, and as you might expect from an 837-page rule (with a 475-page regulatory impact analysis and 215-page technical support document), accounted for it in its analysis. Both the EPA estimates of emissions reductions and of costs to consumers assume that owners of new, more efficient vehicles will drive more. Great, right? Maybe.
Digging deeper suggests that while the EPA is accounting for the rebound effect, it might be underestimating it. The EPA assumes the effect will be 10 percent—that is, 10 percent of the emissions reductions that would otherwise be achieved will be erased by increased driving. This estimate, the EPA admits, is lower than that suggested by most studies of the effect for vehicle efficiency standards in the U.S. These studies give estimates from 7 percent to 75 percent, with the average around 22 percent—more than double the 10 percent the EPA chose. The EPA bases its choice on other studies that suggest the effect is declining over time and then projecting that trend into the future. It’s possible the EPA is right here, but it seems like a very aggressive estimate of the effect. That at least makes it more likely that the EPA will have overestimated the benefits of the new standards than underestimated them.**
For me, the worst part isn’t that the EPA may have made overplayed an obscure (but important) assumption—it’s that none of this is necessary. Pricing carbon, whether through an economy-wide cap-and-trade system, a transportation-sector only cap, or (hide the children) a higher gasoline tax—could achieve emissions gains without these perverse effects—if your reason for buying a more efficient car is a higher fuel price, there’s no incentive to drive more. Such a system would also be simpler, easier to administer, and more transparent. It also need not hurt consumers: the revenues from permit auctions or taxes can be returned to households through lower taxes, rebates, or investments in other public goods.
The reason CAFE standards, with all their problems, are and have long been our only policy option for controlling vehicle emissions has nothing to do with efficiency or good government – its politics. It’s not politically possible (or politicians think it’s not politically possible) to support a policy that increases prices at the pump. Instead, we hide costs in suboptimal and inefficient programs like CAFE. That’s a shame, whatever benefits today’s regulations will have.
* Is it still a compromise if one party (the feds) owns a big chunk of another (the US auto industry) and has Supremacy Clause powers over another (the states)?
** If you’re interested in the EPA’s decisionmaking process here, it’s candidly explained in Section 4.2.4 of the Technical Support Document. I’d be interested to hear what experts in the area think about the EPA’s approach.
