About a year ago — when the economy was merely bad, not yet apocalyptic, and automakers were still solvent — economist Alan Blinder proposed a left-field stimulus plan: the government should pay people to take their old, crappy automobiles off the road.
Cash-for-clunkers programs supposedly offer three benefits: environmental, because old cars pollute more; social, because the money primarily flows to the poor; and economic, because the rebate circulates back into the economy.
Now the idea is actually getting some traction. Germany has implemented a rebate that has resulted in a car-buying boom, and in the U.S. two separate versions of a program are working their way through the house and senate.
Does the idea make any sense? Freakonomics guy says no: “policies which might be a good idea if implemented as one time, short term programs, can be much less attractive if made permanent because of the way they distort incentives.” The article lists a number of ways that a cash-for-clunkers program could be gamed or create weird downstream effects, and Freakonomics guy is right that we should be careful about how we design such a program. But the criticisms remain frustratingly vague. Even if the program doesn’t work as well as advertised, could it still work? Alan Blinder offers up a lot of compelling data nuggets of his own (“cars 13 years old and older accounted for 25 percent of the miles driven but 75 percent of all [presumably non-carbon] pollution from cars”) and it’s not at all clear — to me, at least — how the numbers would wash out in the end.
George Monbiot criticizes the program from a different angle, one dear to my heart: he calculates the per-ton price of carbon reductions from such a program to see how well it stacks up against other uses of the funds. The answer, according to Monbiot, is not very well. His math is a bit hard to follow, but I think he comes out with a figure of about $740 per ton of CO2 reduction, which compares unfavorably to, say, geothermal energy at about $5 per ton.
There are two problems with Monbiot’s analysis. The first is that it applies to a British program that differs in some important ways from the versions being considered in the U.S. The second is that Monbiot is so obviously motivated by a hatred of car companies that it becomes difficult to sort out analysis from spleen. So I ran some figures for a U.S.-based program, and came up with a figure closer to $210 per ton. A lot of assumptions go into a calculation like this, and I wouldn’t put too much stock in the exact figure. It’s certainly more favorable than Monbiot’s, but it’s still quite high.
Is the program worth it? Hard to say. Weighed against the high per-ton price are any stimulative effects on the economy, plus other unquantified benefits, like a substantial drop in particulate emissions. Certainly there are more cost-effective ways of reducing greenhouse gas emissions (e.g., building weatherization). Just as certainly, there are worse ways of spending stimulus money (e.g., building massive new highways). Personally, I’d like to see some more data come out of the trial cash-for-clunkers programs underway in California and Texas before rolling out a national program.