Megan McArdle and Mark Kleiman have been engaged in a little debate on their blogs over the merits of the new CAFE standards for medium and heavy duty vehicles that were announced last week. McArdle criticizes the standards by pointing out (correctly) that commercial trucking operations are pretty sophisticated and cost conscious, which means that unlike car consumers, they’re quite likely to already buy efficient vehicles when high fuel costs merits that. Kleiman replies in defense of the standards by asserting (also correctly) that there are externalities involved: even if no individual trucker benefits from increasing fuel efficiency, society can gain as a whole, since reducing aggregate oil consumption should cut the price of oil. McArdle responds in his comments with three basic claims. First, the rebound effect for heavy trucks should be large, i.e. truckers will drive more if they get more efficient trucks, which will deeply erode any claimed oil savings. Second, since truckers buy diesel rather than oil, the impact of higher fuel efficiency on oil prices will be limited. Third, there are other externalities arising from CAFE standards, some of which may be negative.
These three claims are reasonable but ultimately either wrong or inconsequential; understanding them can give us some insight into how CAFE standards work. Bottom line: The new CAFE standards should deliver net benefits to society.
The rebound claim is the easiest to dismiss. McArdle makes it this way: “Unlike in cars, it is the load, not the truck itself, that comprises much of the weight that the engine is moving, so making the truck lighter, or reducing its power, actually means more trucks pulling more loads.”
The implicit assumption here is that manufacturers will make their trucks lighter in order to meet the new standards, a reasonable reaction given that it’s the pattern we’ve clearly seen in the case of CAFE for cars. But the regulators anticipated this. The fuel economy standards for trucks are designed differently: they are expressed in terms of gallons per ton-mile, not the familiar gallons per mile that’s used for cars. A manufacturer that increases fuel efficiency by lightening the loads its trucks carry gets no credit toward compliance.*
On to the diesel claim, which is also straightforward to do away with. Refineries can alter their operations to change their product slate (within certain boundaries). Less diesel demand means that refineries can tilt their production toward other products, satisfying demand for those and for diesel simultaneously, all while consuming less oil.**
The concern about offsetting externalities is a more worrisome one, though since McArdle wrongly assumes that trucks will be downsized, she overstates the potential problem. But we don’t need to wave our hands to resolve this one: the EPA and NHSTA have provided gobs of analysis on the regulations’ expected impact.
Their conclusion? The regulations would deliver a net social benefit of $49 billion in present dollars. (This assumes a three percent discount rate for most costs and benefits and a five percent rate for climate impacts; the present value drops to $24 billion if a seven percent rate is used for most costs and benefits, and rises to $86 billion if a three percent rate is used for carbon too.) This figure is dominated by “energy security” savings ($20 billion), which basically means reduced oil prices and less exposure to price volatility, and by reduced local air pollution from particulates and ozone ($25 billion). Negative impacts, including accidents, noise, and congestion, are estimated at $8 billion each year.
One can argue with each of these numbers. (My gut says that the energy security figure, as the regulators define it, is probably on the high side, but I’m considerably more confident in the local air pollution numbers.) It would take a lot, though, to overturn the weight of the evidence, which points strongly toward real if not spectacular benefits from the new rules.
All that said, I don’t want to suggest that the new regulations are perfect. In particular, in one important way, they appear to implicitly discount the economic value of reducing U.S. oil consumption. More on that in a future post.
* There’s also an interesting quirk worth noting here. McArdle and Kleiman both seem to agree that sophisticated truck operators are already capturing all economically attractive opportunities to improve efficiency. Assuming that requires one to conclude that the new CAFE standards will force them to take steps that are uneconomic, which is to say, they will force operators to take steps that increase the cost of delivering a ton-mile (the basic product that truckers provide). But that means ton-miles should go down, not up, i.e. there should be no rebound effect. A rebound effect exists only if one believes, as the U.S. government does, that there are efficiency opportunities that aren’t being captured. (The USG estimates a fifteen percent rebound.) One might try to claim that the marginal cost of delivering another ton-mile is decreased by the CAFE standards, and hence that a rebound effect should exist even if operators are already at optimum efficiency, but that wrongly assumes a static fleet size.
** Another way to see why this argument is wrong is to take is to its logical conclusion. Let’s say we believe that cutting diesel consumption has limited value because we’re not directly cutting consumption of oil. We should then also believe the same thing for gasoline, or for any other refinery product. (Indeed in reality we already do a lot to restrain gasoline consumption.) That would imply that a series of regulations cutting consumption of each refinery product would each have minimal impact. But the combined effect, of course, would be precisely the same as an across-the-board cut in consumption of oil. You can’t have it both ways.