05 August 2009

A Clunker of a Policy

The "cash for clunkers" program to encourage people to scrap their old cars in favor of a new one looks to be getting an infusion of new cash, taken straight out of DOE's stimulus package budget for energy innovation. If this isn't bad enough, the entire basis of the plan fails basic economics, as explained by Willem Butier:

Fiscal incentives to induce automobile owners to trade in their jalopies and buy new cars have been introduced in many car producing countries, including Germany, France, Italy and Spain. A number of US states and Canadian provinces also have introduced such schemes. The rationale is partly a general Keynesian demand stimulus, partly a sector-specific subsidy to workers, managers, share holders and creditors in the automobile industry and other industries that depend on them. If the programme is temporary and the cash incentive substantial, such programmes are bound to work.

This artificial shortening of the economic life of a car seems nuts. It’s worse than getting paid to dig holes and fill them again. It’s like being paid to burn down your house to encourage the residential construction industry. In Iceland, where economic calamity has befallen a population that was until the autumn of 2008 among the richest in the world, people torch their SUVs for the insurance money. Iceland doesn’t produce any cars, let alone SUVs, so this does not do their GDP any good, but think of the global externalities! Perhaps the G20 could propose the world-wide legalisation and subsidisation of the willful destruction of consumer durables, residential property and infrastructure (schools, hospitals, prisons etc.) as a global stabilisation policy measure. . .

If the replacement incentive is sufficiently short-lived, there need not be any long run effect on the level of car production, and even little if any effect on the (undiscounted) cumumulative volume of car production. A given cumulative volume of current and future car production would simply have its time-profile shifted from the future to the present. But if the scrapping subsidy were longer-lasting, cumulative car production would increase, resulting in greater environmental damage.

Cash for clunkers has both real costs and opportunity costs. The environmental costs vs. benefits are uncertain, however, a simple calculation using official figures indicates that the cost per ton of projected CO2 emissions avoided is about as much as $160 to $1,400, which makes it a bit difficult to accept Senator Diane Feinstein's claim that "This program has done much better than we ever thought it would for the environment." Maybe they thought that the costs would have been higher than $1,400 per tonne of CO2 avoided!

But even the projected figures are questionable because they assume that the new cars will be driven the same number of miles as the clunkers. Ken Belson at the NYT Wheels blog notes of the trade-in vehicles under the program:
The vehicles turned in were driven about 6,000 miles a year, he said. If the new vehicles are driven about 12,000 miles a year, the rough annual average, then consumers will actually use more fuel, not less.
If the shiny new cars get a more drive time than the old clunker, then it will result in more demand for gasoline. More broadly, lessons from Germany's experience with a similar program are worth keeping in mind:
Retailers, for example, have complained bitterly that the program sucks spending from other categories. German retail sales fell 1.5 percent in March -- the third monthly decline in a row -- a decline that retail industry groups blame partly on incentives to buy cars rather than other goods.

The rebate also is expensive. Nominally it will cost $6.6 billion if Germans take full advantage of the program. The real cost is harder to figure. Increased sales will boost sales tax revenues, and the state will avoid the cost of unemployment benefits for workers who might have lost their jobs. On the negative side of the balance sheet, the program will kill jobs in other parts of the economy, for example auto repair shops or used-car dealers.
Bottom line? Cash for clunkers fails basic economics and environmental tests, and in the case of the US it takes money from a program focused on investments in energy innovation. The politics of cash for clunkers should not lend anyone to optimism that far more complicated and far-reached legislation on cap-and-trade will be able to avoid such political trade-offs.


  1. It's widely commented that the clunker bill was converted from a environmental policy to a stimulus policy. The exact environmental impact, positive or negative, will depend on some statistics that we may never know, such as how many more miles the turned-in cars would have been driven without this policy. I heard on the radio that the time it will take the newer more efficient cars to make up for their mfr carbon emissions ranges from 18 months in the rare case where the buyer bought a Prius and drives a lot, to a more typical 5 years, 8 years in the worst cases. But they didn't mention the cost per ton of carbon that this post refers to.

    Roger closes by saying that this issue doesn't give optimism for climate policy-making. Is there an optimist left anywhere on this issue? Even supporters of W-M don't usually call themselves optimists. Most of the supporters consider it inadequate and only support it in hopes that it will be improved or that it will help bring China on board.

    I'm certainly no optimist. We have a dysfunctional system of governance that is highly resistant to reform.

    (and thanks to Tom for the hints re pasting)

  2. I'm not sure I agree with the idea that this fails basic economic tests. Some argue that the very reason our economy is so strong in the U.S. is because companies strive so hard to manufacture "perceived obsolescence": i.e. to make consumers believe that, even though their material goods work perfectly fine, they need to replace them with a newer version of the product. In some cases it's for good reason (e.g. a faster computer) but in other cases it's unnecessary (the simple example is clothing). Obviously in the latter cases this is incredibly inefficient--we really are basically "burning down houses to build new ones". But it also appears to drive massive consumer spending that spearheads our economic growth.

  3. 1) Examine my comments on this lunatic program.

    2) This program may be third in line behind Kyoto and the $45 Trillion the IEA says we need to spend as the best examples of just how insane The Green Religious Cult can be.

    3) Dr. Pielke, this may be a good opportunity -- respectfully -- to offer a friendly reminder not to forget your commitment to explain what led you to conclude that it is advisable to make the current CO2 famine even more severe.

    As you are a reasonable and sensible person, I look forward to your explanation.

    Again, I would find directly cited peer reviewed science far more convincing than the heavily biased interpretations of peer reviewed science offer up by the purely political IPCC.

    Least convincing of all would be reliance upon the laughable IPCC computer model forecasts.


  4. Where do you get $1400/ton? Jacoby's figure in your linked source is $160/ton. I guess ($3500 to 4500)/(3ton/car/yr) yields about $1400/ton if you have a high enough discount rate.

  5. Tom-

    I did not use a discount rate, I just divided projected CO2 savings by program cost.

  6. OK, but you can't divide the up-front cost by the one-year savings, unless your discount rate is 100%/year. You'd have to amortize over some reasonable vehicle life. You'd also have to correct for how much of the purchase decision depends on the clunker rebate, and various indirect effects (e.g., effect of early clunker retirement on used car prices in general). Jacoby's $160/ton seems a bit optimistic (roughly a 10-yr vehicle life with no correction for other stuff) but $1400 is way pessimistic.

    Either way, it's a sorry excuse for GHG policy.

  7. -6-Tom

    Yes, fair enough. I think "somewhere between $160 and $1400" is a reasonable estimate, and I certainly agree with your conclusion. Thanks for the additional info.

  8. Outside the environmental question, there is the basic economic fairness question. A program like this just transfers money from one group (the taxpayers) to another group (the auto industry). There are other sectors that would like the help, why wouldn't they get the money? Most small hardware stores are in trouble, so why not offer discounts to them? At the end, less productive endevours are promoted, making everybody poorer. It's easy to create jobs (split every job in two). Creating productive jobs is the real challenge. The only way to achieve this is by letting people choose the goods or services they value most (using tax cuts fo example).