The Verdict on Cash for Clunkers: a Clunker

From a new working paper by Atif Mian and Amir Sufi:

A key rationale for fiscal stimulus is to boost consumption when aggregate demand is perceived to be inefficiently low. We examine the ability of the government to increase consumption by evaluating the impact of the 2009 “Cash for Clunkers” program on short and medium run auto purchases. Our empirical strategy exploits variation across U.S. cities in ex-ante exposure to the program as measured by the number of “clunkers” in the city as of the summer of 2008. We find that the program induced the purchase of an additional 360,000 cars in July and August of 2009. However, almost all of the additional purchases under the program were pulled forward from the very near future; the effect of the program on auto purchases is almost completely reversed by as early as March 2010 – only seven months after the program ended. The effect of the program on auto purchases was significantly more short-lived than previously suggested. We also find no evidence of an effect on employment, house prices, or household default rates in cities with higher exposure to the program.

Levitt was skeptical two years ago. A recently departed Obama official, meanwhile, disputes the finding.


"A recently departed Obama official, meanwhile, disputes the finding."



Isn't saving just 'delayed spending' according to economists? Why was this a good idea again? The only reason I can think of is increasing liquidity for troubled car manufacturers during that time frame...

Joe D

Putting money in people's hands is only stimulative to the extent that they *must* spend that money (see "paradox of thrift"). CfC only stimulated those who needed a new car. Tax cuts for the wealthy are not stimulative. Unemployment benefits are.

Direct government demand (infrastructure maintenance, e.g.) is also stimulative, of course.


so it didn't provide economic stimulus,? however, it did get inefficient cars off the road and replace them with more fuel efficient cars. anyway to measure the economic impact of that?


Ron Paul said this was dumb. Ron Paul is always correct. End communication.


What was the total impact on average car gas mileage and fuel efficiency? Can it be said that those people now have more buying power because they spend less on gasoline, thus raising their standard of living?

Drill-Baby-Drill Drill Team

A spending program should try to achieve multiple goals: improve the economy and improve transportation, but how about improving quality of life and encourage exercise in a sedate, couch-bound population.

I believe transportation requires multiple modalities to get us to our destination. We cannot discount walking. Even SUV riders have to leave their vehicles and walk the last 100 yds and walk throughout building in an 8 hour workday. But walking takes time and is not viable for long distances.

Most of our daily errands are within a few miles of home. A bicycle multiplies the human effort by several fold. We should be able to accomplish trips up to 7 miles on our own pedal power (Less than 30 minutes transit).

For middle distances around city arteries from 10-40 miles, we should plan our homes and cities around mass transit. Buses, street cars, trains, and even high speed rail. Of course this infrastructure takes decades and billions of dollars. But it should not stop us from relocating our homes close to city centers, hubs and major arteries of transit.

Our society currently uses the car exlcusively for distances from 100 yds to 100 miles. We should be able to use the proper tool for the proper distances. Note that combining exercise and time for conversation and our own private thoughts may be integrated into our commutes.

Lost Opportunity Cost: 3 billion dollars helped the purchase of 360,000 cars. But as an alternative, the government could have purchased 30 million bicycles and given them FREE to active adults for use in short errands, exercise and commutes. They could tie the gift, to a promise to either use the bike daily or give it to someone who will. Even if only 3% used them daily, we would have 1 million people not using their cars, not contributing to parking problems and traffic jams, diminished smog and noise, getting regular cardiovascular exercise, combating global warming, connecting them with their local neighborhoods and improving quality of life without the expense of a single barrel of oil.

Call it Government Bicycle Give Away. That may be a billion dollar bonanza that will accomplish many goals: improve quality of life, make us healthier, and last for decades. It's Win-Win. Or you could watch your new 2008 Mini SUV become a rusty, oil leaking, clunker of tommorrow.



Two points:

1. An important objective of Cash for Clunkers was to replace less fuel efficient vehicles with more eco-friendly vehicles. As opposed to the Bush era tax credits for small businesses to buy gas guzzling SUVs.

2. Moving up purchases could have positive effects on the economy. This does not seem to be examined by the paper.


I suppose I could read the source article, but I didn't....

What are they using as the baseline for when the total purchases of cars was reversed? In this economy people are purchasing much less. I'd be willing to bet that one of the things people would cut back on first would be new cars. What's to say the slower car market now has nothing to do with the cash for clunkers program and everything to do with current economic conditions.

I suppose this was probably dealt with in the paper, but as I said I'm too lazy to read it.

Jeff #3

#7 Drill-Baby-Drill Drill Team

Sometimes I agree with you, sometimes I dont, but I always find your comments interesting to read.


This isn't new. Only Obama's brilliant economists seemed to skipped this lecture on economic stimuluses of the past in Econ 101. And, considering how unsuccessful the cash for clunkers was, they repeated for it for housing. Surprise, surprise they got the same result.

Don't dispair. Employment could get better with the "Green Money" that went to China to purchase wind turbines whose manufacter creates so much pollution that only the Chinese can supply the world. And, there is the stimulus money sent to Africa to clean foreskins!


Only problem with this conclusion is we can't know what would have happened if Cash for Clunkers hadn't be conducted.

For all we know, the economy could have slowed further in the near term, provoking even worse economic conditions subsequently. We don't know. We can't know.

Note: Not a fan of most of the administration's policies. Didn't vote for President Obama.

Dan D

How come no one ever talks about cleaning out 2009 inventories to make room for 2010 models?

Isn't that a problem in a recession? Especially a GREAT one.


The paper's conclusion is that the same cars on the road in March 2010 would have been on the road in March 2010 without the program, right? If that is correct, then I think that from March 2010 on the effect on fuel efficiency raised by comments Nos. 4 and 6 is 0. The only effect the program had on fuel efficiency would have been between July 2009 and March 2010.


It was for political gain, someone that helped the administration was either getting paid back, or now owes the Obama administration. That's how politics work. By the way, if someone sends me some money, I will blog in agreement of your opinion.
Ask yourself, "What is the relevant scarcity hindering a better outcome?" When we answer that we'll know where to direct the incentives.


This program was the worst kind of socialism, forcing taxpayers to buy cars for others.


But the point of the program was to advance demand from the future because the car companies were in desperate conditions THEN. So it did what it was designed to do.

That you disagree with the premise of the program is obvious, but the words you quote say the program did exactly what it was supposed to do. I'm sorry you don't want to remember how desperate things were just a short time ago and that the fear was for literally millions of auto related jobs.

It's also somewhat funny - but illuminating - that a huge argument about the limited Gulf drilling moratorium is that it would cost jobs. (Which it apparently has not because the companies would incur cost to move rigs and let people go only to hire them later, which was predictable but not mentioned because that wasn't politically expedient.)

I guess that means government action to support jobs is bad and government action that might cost jobs is bad because I gather that all government action is bad. Is that what you believe?


Max Andronichuk

Paying 100 people to go around smashing street light and then paying 100 more people is "stimulus" in the eyes of the retard economic philosophy of keynesians and social democrats....

Their rationale left them before it even had time to arrive.

Any sane person will no doubt conclude:
why not just use that money and those 200 people on a project/s bringing in NEW, efficient and socially useful activities as outlined by the Market...

But no, on they go...

I would love to see these people try and pitch an idea like this to investors in the private sector where they would be held to account... Good luck raising the money, I won't be buying any shares.

Fire these bastards


It is incorrect to state that NO cars were purchased that wouldn't have been purchased anyway. We had no intention of buying a new car. However, getting $4K for a 15-year-old end-of-life gas guzzler was indeed the incentive that got us to trade an inefficient minivan plus the clunker for a new 40 MPG car. So it got one actively driven plus one rarely driven gas guzzler off the road and we own a new car we wouldn't have bought otherwise.

(And for those who claim it's wrong to use exceptional situations are arguments, tell it to Ronald Reagan [welfare Cadillac] and Karl Rove/Lee Atwater [Willie Horton].)

Julien Couvreur


Yes, empirical economic analysis is difficult. So it is difficult to prove that the situation would not have been worse, absent said policy.

Similarly, though, it is difficult to prove that the situation would not have been better, absent said policy.

So, if you're saying that the study is inconclusive, because of the method it uses, it is inconclusive both ways. The question is what method would you recommend to prove that the policy was justified (policies do have to be justified, not just voted on political whim, right?).