A Last Word on ‘Cash for Clunkers’

One thing the government's CARS program -- a.k.a. "cash for clunkers" -- has clearly stimulated is commentary. For a policy involving a shade under $3 billion in federal spending, it has enjoyed no shortage of media coverage.

(Photo: Newsday)

In part this is because the program looks like a big success, and certainly congressional leaders and the White House have not been bashful about touting it as such. The original $1 billion allocation for the program was exhausted within days, and as sales data for August begins to emerge it is clear that car sales experienced a banner month.

Was CARS a good policy, all things considered? Let's look at a few of the latest numbers on the program.

There were approximately 1.17 million vehicle sales in August, which works out to a seasonally adjusted annual rate of about 14 million vehicles. June's sales rate was under 10 million and near the recession low, while last August's rate was also about 14 million. Meanwhile, the August norm in good times was about 16 million.

What does that say about the value of the program? Well, let's say that August sales would have matched June's sales in the absence of CARS. They almost certainly would have been higher given economic improvements between June and August, but for argument's sake, let's say they were the same. We can then estimate how many additional sales CARS produced and the actual subsidy per new sale.

Here's economics blogger Calculated Risk:

In other words, CARS just didn't generate that many new sales. Much of the subsidy went to buyers who would have purchased anyway.

As it turns out, much of the subsidy also went to people who weren't interested in purchasing GM or Chrysler vehicles. While year-over-year sales figures rose in August for Ford, Honda, and Toyota, sales declined by 15 percent and 20 percent respectively for Chrysler and GM. To the extent that CARS was designed to help struggling American automakers, it doesn't seem to have had the desired effect.

Particularly worrisome is today's report that sales fell precipitously in the last week of August -- after the CARS program ended. Rather than generate momentum for the automobile industry, CARS may have primarily moved sales around. To a certain extent, it might also have been counterproductive. How so?

Given fixed supply, a purchase subsidy will often just lead to an increase in price:

And while Transportation Secretary Ray LaHood claimed that CARS saved jobs and the the inventory draw-down would lead to increased production, automakers are likely to be cautious in building new vehicles if demand appears unsustainable. The White House's claim that CARS will boost third quarter output by 0.3 percent to 0.4 percent will not prove accurate if September sales fall back below trend.

And then there's the environmental effect of the plan. Much attention has been paid to the fact that purchased vehicles were some 9 miles per gallon more efficient than traded-in vehicles. As I've noted before, much of that gain would likely have taken place without the program, based solely on the fact that oil prices rose steadily over the past decade.

One pair of economists estimated that the carbon savings from the CARS program worked out to roughly $596 per vehicle -- well below the voucher values of $3,500 or $4,500 per new vehicle. Another economist estimated that the implied cost of carbon under the program was somewhere between $237 and $500, much higher than what is assumed to be an efficient carbon price.

No matter how you cut it, CARS was an expensive means to reduce emissions.

That doesn't mean that it was a total waste. There was almost certainly some positive economic and environmental impact from the policy.

But that $3 billion could have been used elsewhere. Other potential programs -- restoring heavily used transit services trimmed by budget cuts or funding weatherization programs, for instance -- would almost certainly have been greener and more stimulative.

In the end, "cash for clunkers" should be understood as a missed opportunity, politically attractive but far from ideal as policy.