Cash for Clunkers: Not Much

The popular program was even less successful than we initially thought.

The initial view of the Cash for Clunkers program was very positive. Lots of cars were being sold and there was the additional secondary benefit that people were getting more fuel efficient cars–i.e. it was also good for the environment. Now a new analysis by Resources for the Future suggests that perhaps that initial view of the program is faulty.

In the United States, the Cash-for-Clunkers program provided eligible consumers a $3,500 or $4,500 rebate when trading in an old vehicle and purchasing or leasing a new vehicle. Many other countries, such as France, the United Kingdom, and Germany, have similar programs, which generally share the same two goals: to provide stimulus to the economy by increasing auto sales, and to improve the environment. The U.S. program received enormous media attention and many considered the program to be a great success; during the program‟s nearly one-month run, it generated 678,359 eligible transactions and had a cost of $2.85 billion.1 But as a matter of economic theory, it is typically quite difficult to achieve multiple goals with a single policy.

Indeed, one of the problems with the environmental aspect of the policy is Jevon’s Paradox.

Jevons paradox (sometimes Jevons effect) is the proposition that technological progress that increases the efficiency with which a resource is used, tends to increase (rather than decrease) the rate of consumption of that resource.

This is a relatively straight forward application of micro-economic theory. With an increase in fuel efficiency the price of the fuel will fall and demand for the fuel will increase. How much of an increase there is in demand depends on the price elasticity of demand. While price elasticity is often inelastic in the short run, over the longer run it is more elastic. Still, the Jevon’s/Rebound effect is often less than 100%. That is, even in the long run the price elasticity of demand is still inelastic. Then there is the Khazzoom-Brookes postulate that argues the rebound effect is but one way in which energy usage can increase. There are two additional factor that will increase demand for energy. First is that there is something like an income effect. If you save $100/month in fuel expenses you’ll likely spend most, if not all, of that $100 on something else which will increase economic growth which in turn will increase demand for energy/fuel. Also, improving the efficiency of one limited resource will have a beneficial effect on all complimentary resources. Thus, energy efficiency gains paradoxically tend to increase energy consumption vs. reducing it.

As for the increased car sales one of the main criticisms was that the program may end up time-shifting car purchases. That is a person might already be planning on purchasing a new vehicle, but is going to be doing it in November or December. Cash for Clunkers merely moves that purchase forward via a subsidy. Thus, the initial surge in sales in July and August would be at least partially offset by a decline in sales in the months that follow.

To test these effects the analysis uses Canada as a control group since Canada did not have a Cash for Clunkers style program. In addition the authors argue that qualitatively Canada car market is similar to the U.S. market. They use data from 2007 to 2009 on new car sales.

The results of the analysis indicate that of the 0.66 million cars sold during the Cash for Clunkers in July and August 2009 program 0.3 million would have been sold anyways and that only 0.36 million were car sales that other wise never would have happened. Outside of July and August the program had negligible impact and when looking at the total number of new cars sold from June to December of 2009 the effect was close to zero–i.e. there was almost no stimulus effect when factoring in time shifting over a longer horizon. And as for the benefit to U.S. automakers?

In addition, our simulation results show that Toyota, Honda and Nissan benefited from the program disproportionally more than other firms: with a combined market share of around 38 percent before the program, they accounted for more than 50 percent of the increased sales.

So in the end, the impact of Cash for Clunkers appears to be minimal and this program is considered to be one of the more successful programs. Hope and change you can believe in, or something like that.

FILED UNDER: Economics and Business
Steve Verdon
About Steve Verdon
Steve has a B.A. in Economics from the University of California, Los Angeles and attended graduate school at The George Washington University, leaving school shortly before staring work on his dissertation when his first child was born. He works in the energy industry and prior to that worked at the Bureau of Labor Statistics in the Division of Price Index and Number Research. He joined the staff at OTB in November 2004.

Comments

  1. Not to mention raising the cost of owning a car for those at the bottom end of the car market by removing a lot of inexpensive inventory. Let them take bullet trains!

  2. jakeman says:

    A few weeks ago, I asked my friendly neighborhood mechanic how CFC had impacted their business. He said it hammered them hard, and they’re only now beginning to recover. The unintended consequences went far beyond the (apparently negligible) impact on new car sales.

  3. Hey Norm says:

    Um….wasn’t the prime purpose to help the auto industry? And isn’t the auto industry on sounder footing than anyone predicted? That’s what I thought.

  4. @Hey Norm: No and no. Next question.

  5. Rob in CT says:

    Now this is a program that never made sense to me. Thankfully it was pretty small.

  6. David M says:

    Even time shifting purchases was a positive in the summer of 2009. Also, I don’t see a lot of cooperation from the GOP on anything else that might reduce emissions, so I’ll give it a pass there.

  7. jan says:

    @charles austin:

    Not to mention raising the cost of owning a car for those at the bottom end of the car market by removing a lot of inexpensive inventory. Let them take bullet trains!

    Trying to buy a reasonably priced, good, used vehicle became next to impossible after this “program,” as the provisions, allowing you turn a car in, included having the car crushed, shredded, or dismembered for parts, except for the engine and drivetrain which had to be ‘frozen’ to insure they would never operate again. In other words, the used car inventory artificially dried up.

    According to the research done by Edmunds.com. this program was not a good investment for the government, either.

    The average rebate was $4,000. But the overwhelming majority of sales would have taken place anyway at some time in the last half of 2009, according to Edmunds.com. That means the government ended up spending about $24,000 each for those 125,000 additional vehicle sales.

  8. David M says:

    I will point out the idea that cash for clunkers caused a significant increase in used car prices is just silly, especially given the complaints that the program didn’t do much to improve overall fuel economy (because it was too small).

    Only 690,000 vehicles were taken out of circulation by this program in 2009, and the used car pricing trend has much more to do with the much larger slowdown in new car sales and the economy tanking in 2008.

  9. Only 690,000 vehicles. It’s the margins that count though.

  10. Hey Norm says:

    “…Trying to buy a reasonably priced, good, used vehicle became next to impossible after this “program…”

    That’s just pure bullshit. BULLSHIT. [can I get something more than just bold???]
    Have you no shame woman? Have you heard of Craigslist? eBay? Or my personal favorite by far…bringatrailer.com?
    Friggin’ unbelievable.

  11. jan says:

    @David M:

    I will point out the idea that cash for clunkers caused a significant increase in used car prices is just silly, especially given the complaints that the program didn’t do much to improve overall fuel economy (because it was too small).

    After this program used car prices went up as the inventory went down. It’s that whole idea of supply and demand that most of you don’t agree with. But, in reality when an item’s numbers are reduced (this time artificially tampered with by the government), their per unit monetary value usually increases.

    Below is some documentation:

    Used car prices up 30% in three years

    I’m not implying the 30 percent increase in the average price of a used car is a direct result of the Obama administration’s Cash-for-Clunkers program, but it certainly is a factor, and nobody is mentioning it.

    Dealers are paying an average of $11,660 for a used car or truck, up almost 30 percent since December 2008.

    Boston Globe editorial…

    IN THE market for a used car? Good luck finding a bargain: The price of “pre-owned’’ vehicles has climbed considerably over the past year. According to Edmunds.com, a website for car buyers, a three-year-old automobile today will set you back, on average, close to $20,000 — a spike of more than 10 percent since last summer.

    Clunkers program could drive used car prices up

    That means up to 750,000 vehicles will never find their way into the hands of another owner. Many are at the end of their useful lives, but others, with years of life left in them, normally would be resold.

    “Those are the cars that lower-income families need,” says Geoff Smartt, owner of Smartt Cars in Caldwell, Idaho.

    Too few older cars at reasonable prices could put some dealers out of business, says Tim Swift, general manager of the Corry Auto Dealers Exchange, an auction operation for dealers in Corry, Pa.

    “It’s was tough finding them before, and now, it’s even worse,” says Mike Williams, owner of Auto Wise in Shelbyville, Ky. “The $3,000-to-$5,000 car is just gone.”

    Finally…..

    Cash for clunkers muted effect on economy

    The Obama administration’s program to encourage new car sales may have encouraged people who would have bought new cars anyway to do so sooner rather than bringing new buyers into the market, according to a UC Berkeley economist.

    When the incentive was offered, automakers sold an additional 360,000 vehicles in cities with lots of used cars, only to see sales plummet once the program expired, said Atif Milan, who teaches economics and finance at Cal.

  12. jan says:

    @Hey Norm:

    That’s just pure bullshit. BULLSHIT. [can I get something more than just bold???

    You just fly by the seat of your screaming pants, don’t you?

  13. Trumwill says:

    Cash for Clunkers was a ridiculous program, not the least of which because it destroyed functioning capital, but it’s hard to place too much blame on it for the used car pricing problems*. The makes and models where the price went up the most (presumably because of shortages) were not eligible for C4C. Used car sales fell by a whopping $6mil. C4C was a fraction of that. The bigger issue was that in the economy at the time, people were holding on to their cars rather than buying new ones and trying to buy used rather than new. People who don’t buy new cars don’t sell their existing one, and people who buy used aren’t adding to the used car market.

    * – Which, yes, do exist. Or did. I was car-shopping throughout 2008-09 and it was because of the price uptick on used cars that we ended up buying new (something we always swore we would never do).

  14. David M says:

    @jan: Thanks Jan, you refuted your own claim with those links. Nothing you linked to comes close to supporting the idea that cash for clunkers has had any lasting or significant impact on used car prices.

  15. Trumwill says:

    Note, the $6mil should not have a dollars sign in front of it. It was a reference to new cars moved, not dollars. Also, here’s a link from Edmunds after C4C mentioning that the biggest price hike was in late-model used cars and not new ones. Cash For Clunkers conspicuously took *old*, not late-model used, cars off the road.

  16. Ben Wolf says:

    Jevons paradox (sometimes Jevons effect) is the proposition that technological progress that increases the efficiency with which a resource is used, tends to increase (rather than decrease) the rate of consumption of that resource.

    Show quantitatively that we burned more fuel than we might otherwise thanks to this program, rather than implying it. Furthermore, it’s bizzare to complain the economic effects of a $2.8 billion program were minimal when consumer spending fell by over $1 trillion per year. Minimal stimulus, minimal impact.

  17. Ben Wolf says:

    The reaction of the right toward stimulus borders on the obscene. It’s the equivalent of a doctor prescribing three strong doses of medicine for an ailment, then having Steve Verdon tell the doctor, “I administered half a dose and it didn’t work, so obviously your treatment is wrong.”

  18. Steve Verdon says:

    @Hey Norm:

    Norm,

    What part of negligible impact on sales from June-December for 2009 don’t you understand. It did not boost car sales. So maybe the auto industry is doing better than anticipated, but not because of this policy…at least that is the conclusion of the study.

    @David M:

    Even time shifting purchases was a positive in the summer of 2009.

    How? Moving a car purchase a couple of months probably does little to help the economy. Moving a car purchase forward a couple of years….

    @David M:

    Lasting impact? No of course not since the program itself had no lasting impact. I’m not sure about short term transitory effects though.

    @Ben Wolf:

    Ben,

    Sorry, this is a blog. Going out and doing that kind of analysis is not going to happen. But it is a well known effect. Look at the OPEC oil crises. Those spurred on fuel efficiency but fuel consumption increased. So the effect can and has happened.

    You can read Jeffrey Rubin’s commentary here regarding energy efficiency and why it doesn’t reduce energy demand, but actually increases it.

    Furthermore, it’s bizzare to complain the economic effects of a $2.8 billion program were minimal when consumer spending fell by over $1 trillion per year. Minimal stimulus, minimal impact.

    Oooohh, goal post moving noted.

    It’s the equivalent of a doctor prescribing three strong doses of medicine for an ailment, then having Steve Verdon tell the doctor, “I administered half a dose and it didn’t work, so obviously your treatment is wrong.”

    Sorry false analogy. Try again Ben.

  19. Ben Wolf says:

    Sorry, this is a blog. Going out and doing that kind of analysis is not going to happen.

    In other words we used more gas because you say we do, i.e. you know because you know. Gotcha.

    You can read Jeffrey Rubin’s commentary here regarding energy efficiency and why it doesn’t reduce energy demand, but actually increases it.

    In other words efficiency never works because Jeffrey Rubin says so, i.e. he knows because he knows. Gotcha.

    Oooohh, goal post moving noted.

    What goal post? To move it I would have had to make an argument, then been challenged, then changed my argument to neutralize the challenge. I’m not seeing that in this thread.

    Sorry false analogy. Try again Ben.

    In other words “Nuh-uh, so shut up.” I think I’ll stand with what I wrote.

    Oh and Ben, if I wanted to shut you up, I can do it very easily. So stop being such a ding bat.–Steve Verdon

  20. David M says:

    @Steve Verdon: I didn’t have many disagreements with what you wrote, more with the asinine idea a couple people brought up that somehow cash for clunkers caused the increase in used car prices.

    The program did about what I expected it to do, little as that may be. Knowing what we know now, I’d still support it although it’d be nice if it could have done more to increase fuel economy, even if that meant making it a larger program.

  21. john personna says:

    Jevons’ Paradox is not a law. This is not physics.

    And as it happens the disproof is trivial. VMT has fallen, and has not increased, since C2C.

  22. Steve Verdon says:

    @Ben Wolf:

    You are clearly quoting me out of context. No, I’m not going to do a brand new econometric analysis for a f*cking blog post. Especially one that is going to require obtaining fairly hard to get data.

    Second, I pointed out that even though OPEC and the oil crisis spurred on considerable energy efficiency we are now consuming more oil than ever. Granted it is in part due to population growth, but it is more than that. There are economists out there who have analyzed this. In particular Khazzoom and Brookes as well as David Saunders.

    So it isn’t just one other guy. It is a number of researchers who have looked at the issue and come away saying, yeah as weird as it sounds it appears energy efficiency improvements increases energy consumption. When you keep in mind that improvements in energy efficiency basically imply a decrease in the relative price for energy it makes alot more sense and doesn’t seem so weird. Factor in that it can enhance macro-economic growth as well and make compliments look more attractive economy wide the result seems even less paradoxical.

    Bottom line if we are going to rely on energy efficiency improvements to solve environmental problems they need to be coupled with taxes that strip away the economic benefit of the energy efficiency improvements, and the chances of that kind of policy happening are very slim.

    What goal post? To move it I would have had to make an argument, then been challenged, then changed my argument to neutralize the challenge. I’m not seeing that in this thread.

    My argument was not:

    Cash for clunkers was small, therefore it is a failure because it didn’t address the $1 trillion fall in consumer spending. You set up a completely different metric and used that…hence goal post moving.

    In other words “Nuh-uh, so shut up.” I think I’ll stand with what I wrote.

    No, it is me saying, that isn’t my argument, so try again.

  23. Hey Norm says:

    @ Jan…
    Only when I see pure unmitigated BULLSHIT.
    I’ve bought two vehicles since C4C that match the qualities you claim cannot be found.
    You are simply full of shit. Period.

  24. Hey Norm says:

    And I looked at a beautiful M3 convertible on a dealers lot last week that was a great deal.
    You spout infactual BULLSHIT.

  25. Steve Verdon says:

    @Hey Norm:

    Okay Norm we get it you got money and are special. Now cool the tone of your comments down.

  26. Steve Verdon says:

    @john personna:

    Nobody wrote or implied Jevon’s paradox is a law. Nobody implied or claimed this was physics.

    That “disproof” is not a “disproof” since the claim isn’t that people simply drive more, but that they use that resource more.

  27. john personna says:

    @Steve Verdon:

    So what are you saying, they consume less gasoline and consume more of something else?

    Because a decline in VMT means that they definitely didn’t expand use of the *same* resource. They bought new cars, and drove *less*.

  28. john personna says:

    Evidence Mounts That Car Usage Is Declining

    That is totally contra-Jeavons, and certainly looks good for those of us who have been calling BS on the Paradox as a *determinant* for years.

  29. john personna says:

    From the WSJ, on April 13, 2009:

    And the demand picture is indeed weak, especially in the U.S., where gasoline demand apparently peaked in 2007 and won’t return. That’s due to a cocktail of higher fuel-efficiency, shifting transportation habits, and the growing role of biofuels

    I think Jevons is one of those seductive concepts that actually makes the believer stupider. It is a pat answer that presumes an outcome in what is actually a varied and multi-factored world.

  30. Rob in CT says:

    It’s rather hard to know, isn’t it? I mean… you could have an increase in driving due to greater efficiency, but if oil prices are rising at the same time, the effect would be swamped.

    Or there isn’t really an effect. That’s gotta be hard to untangle.

  31. john personna says:

    @Rob in CT:

    My contention was that while efficiency allows expanded use, all else being equal, those rising prices could indeed drown out the effect.

    But as far as knowing, we know that gasoline demand and driving are actually down, so the global price pressure was sufficient to reduce use in the US.

    There are a lot of interesting choices apparent in that. Many people chose to buy faster cars with lower MPG, rather than hybrids they could drive more at the same cost.

  32. Rob in CT says:

    Also, wouldn’t we expect to see less driving with 9% unemployment?

  33. john personna says:

    @Rob in CT:

    Yes, and there is a well-known correlation between GDP and VMT.

  34. EddieInCA says:

    @Steve Verdon:

    ….for a f*cking blog post

    Steve Verdon says:
    Wednesday, November 9, 2011 at 23:40

    @Hey Norm:

    Okay Norm we get it you got money and are special. Now cool the tone of your comments down.

    Pot… meet Kettle.

  35. robert wilson says: