David R. Henderson  

Cash for Clunky Ideas

PRINT
A Fact for Tyler Cowen... Status, Greed, and Power...

Cash for Clunkers Destroys Wealth

Under the current so-called "Cash for Clunkers" program, people who have owned a car for at least a year that gets below 18 miles per gallon can turn in the car as trade for a new car and get a payment of up to $4,500 from the dealer. The dealer then destroys a certain amount of the car to take it out of circulation and then, at least in theory, gets reimbursed by the government.

The program destroys wealth. Here's how.

First, consider the baseline, a $4,500 subsidy that you get simply from buying a new car. The standard analysis of a subsidy to output applies here. The buyer and seller split the subsidy, with the split depending on the relative elasticities of supply and demand. But it's not a wash. There are two deadweight losses (DWL). One is from the incentive to buy a car that the buyer values below the cost. I have no idea how big this DWL is. Let's say it's minimal: $300 on the $4,500 subsidy. The other DWL is bigger, and comes about because the $4,500 in revenue to pay for the program is not a lump-sum tax. It's raised from future income taxes, sales taxes, etc. Each of these has DWL, sometimes called an excess burden. This excess burden is the loss to the economy from the distortion in behavior caused by the tax. A typical DWL number is 30% of the revenue raised. So the DWL from the $4,500 in tax revenue is $1,350.

So now we're up to $1,650 of DWL. But wait; there's more. There are two other factors, one of which could be a small gain and the other of which is a loss. The small gain is due to taking off the road a car that was imposing external costs with pollution and, if global warming is a problem, with more carbon dioxide. I have no idea how to estimate this. The other loss is more substantial and comes about because this is a cash-for-clunkers program, not a cash-just-for-buying-a-new-car program. It comes about, in short, because value is explicitly destroyed.

Imagine a car owner who takes advantage of the program. Ideally, for efficiency, the cars that get destroyed would be the lowest-valued ones. But the program gives the same subsidy to a person with a car worth $3,000 as it does to a person with a car worth only $1,000. So whoever gets there first is the one who gets the subsidy. (At least Congress had the "wisdom"--I use that word loosely--to limit the overall expenditure on the program to $1 billion, but then raised it to $3 billion. Otherwise, with no limit, the program would have caused every car worth under $4,500 to be destroyed and would have acted like a price support for wheat, driving up the price of cars to at least $4,500 just as price supports drive up the price of agricultural products.)

So let's imagine that the average car destroyed is worth $2,000. If you assume a lower number, you'll get a lower wealth loss, and if you assume a higher number, you'll get a higher wealth loss. Even if the $2,000 car is destroyed, maybe the spare parts that are "spared" are worth $200. So for that $4,500, $1,800 in wealth is explicitly destroyed. So, the net loss per $4,500 payment is $1,650 plus $1,800, or $3,450, minus the small environmental gain.


Comments and Sharing


CATEGORIES: Cost-benefit Analysis



COMMENTS (33 to date)
Radford Neal writes:

Aren't you forgetting the negative environmental externalities associated with the manufacture of a new car that would otherwise not need to be built (yet)? These might well exceed the environmental benefits of getting the older car off the road.

Troy Camplin writes:

Oh, and don't forget that many dealers are giving buyers the cash for clunkers on the *hope* that the government will reimburse them. What percentage of dealers won't be reimbursed? Those who will never be reimbursed will have lost that money -- thus, more value is lost in the economy. And, the more foolishly a dealer was to trust the government, the more likely they were to give out cash for clunkers without waiting to see if the government would cover them -- meaning, such dealers could even stand a chance of going out of business, harming the economy even more.

Also, the destruction of the cars also has an environmental effect -- though one could argue that it's merely destroyed now rather than later, so the effect is a wash.

pi writes:

[Comment removed for supplying false email address. Email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Radford Neal writes:

Also, the destruction of the cars also has an environmental effect -- though one could argue that it's merely destroyed now rather than later, so the effect is a wash.

But presumably if this program is praised, programs like it will be instituted at irregular intervals in the future, with the effect that both manufacture and destruction of cars proceeds a continuing higher rate than would otherwise be, with continuing negative environmental effects.

Justin writes:

The environmental impact is likely negative when you account for the used car market. Normally when a middle class family gets rid of their "clunker" and buys a new car, the clunker gets resold to a less well off consumer that is driving a car with even greater environmental externalities (worse gas mileage, worse emissions control, oil and fluid leaks, etc.). The car that is ultimately sent to the scrapyard as a result of this chain of transactions is far worse from an environmental standpoint than the original "clunker." By scrapping the "clunker," you end up keeping cars on the road that are far bigger polluters because the owners of these polluters cannot can't afford a new car even with a $4500 voucher.

Mike writes:

Consumers are happy with CFC because they are getting discounts on a new cars.

Industry is happy because they are selling cars.

Environmentalists are happy because the new cars are cleaner.

Government is happy because they made a program that is popular with consumers.

Libertarians are unhappy because a "progressive" program is being successful.

What is your beef with CFC, other than a contrived ideological argument with numbers you pulled from a hat?

Ironman writes:

That small environmental gain is smaller than you might think....

Turk writes:

The most difficult part of having libertarian views, I have found, is forcing myself to be unhappy with success.

hutch writes:

Mike,

define "successful".

Mike writes:
Consumers are happy with CFC because they are getting discounts on a new cars.

*[hutch: only consumer who have the means to purchase new cars. consumers who are in the markt for a used car will pay higher prices because there are fewer used cars on the road.]*

Industry is happy because they are selling cars.

*[hutch: only the car industry is happy. there is more than one industry in our economy that is ailing, namely airlines, small consumer durables, etc. they aren't happy b/c they didn't get anything.]*

Environmentalists are happy because the new cars are cleaner.

*[hutch: almost by definition, progressive legislation will make progrssives happy. next.]*

Government is happy because they made a program that is popular with consumers.

*[hutch: legislation shouldn't be considered a success simply b/c it gives politicians something to brag about. how much crap has been passed with this end in mind? too much.]*

Libertarians are unhappy because a "progressive" program is being successful.

What is your beef with CFC, other than a contrived ideological argument with numbers you pulled from a hat?

Nathan writes:

Yes Mike, the program appears to be incredibly successful all around. And best of all, the money to pay for it didn't have to be forcibly taken from millions of taxpayers to subsidize others' purchases! Like all good government programs, the money was simply generated out of thin air by the good-intentions fairy! What could be better?

Troy Camplin writes:

A value-destroying update. I just heard on the news that GM is increasing production because of the "success" of the cash for clunkers program.

Now, cash for clunkers is primarily designed to get people who are still working and have enough money to buy a car to go out and buy a car now. They would have done so in the future, of course, but they were encouraged to do so now. One of the results of this is that these same car buyers won't be buying cars over the next several years, so all you did was consolidate buyers into one short time span.

Cash for clunkers has indeed resulted in more cars being bought. This sent a signal to GM (and other car makers) to make more cars. More demand means they need to make more supply to meet that demand. Under normal conditions, that would be a rational move.

But wait! Cash for clunkers distorts economic information. Cars that are on the lot are wanted NOW, but future cars are not wanted -- especially since the program has been cancelled. So GM is making more cars to meet current demand, but current demand is over. This means, they will end up with a large stock of unwanted cars, which will only hurt GM financially. With all the extra cars around, and nobody buying them (for both reasons given above), GM will have to either cut back on hours worked, or lay people off. Which will slow down the economy further.

In other words, cash for clunkers, by distorting economic information, is creating a short-term car bubble, which will burst in the near future.

Tell me, David, how's my economics on this?

Joe writes:

Troy:
Maybe perhaps GM is replacing inventory to remain at a certain level. I think they read the newspapers and have a good understandin that demand is being pushed forward.

Dave:
Maybe we are missing other positive externalities:
Newer cars are safer, drive better and handle better than older cars, especially those worth less than $4,500. So we should have fewer accidents with inuries, thus improving the overall health and productivity of the nation.

Another Possibile Externality:
Buying a new car feels so good, and makes you feel so much better about youself ( At least for a short while) that the total gain in CONFIDENCE is important.

Along the same lines, maybe that confidence boost encourages mating; increase population and increase future productivity.

There is so much fun to be had with the CARS program...

Michal writes:

Nathan hits the nail on the head in his response to Mike. Let me add that I'm unhappy, not because of any libertarian sentiment. I'm unhappy because I have a car, don't want to buy a new one and positively don't want to help somebody else buy one. That money is taken, in part, from me! Mike also forgets the unseen opporuntity cost. The money I'm forced to spend on somebody else's ride is now taken out of the pocket of the guy who was going to fix my roof.

Mike writes:

"only consumer who have the means to purchase new cars. consumers who are in the markt for a used car will pay higher prices because there are fewer used cars on the road"

Only cars that get 18mpg or less are eligible. You can still happily go buy an old Civic for

"only the car industry is happy. there is more than one industry in our economy that is ailing, namely airlines, small consumer durables, etc. they aren't happy b/c they didn't get anything"

Nothing wrong with that. Stimulus need not target every industry all at once.

"legislation shouldn't be considered a success simply b/c it gives politicians something to brag about. how much crap has been passed with this end in mind? too much"

Noted. But keep in mind that politicians being able to brag about something means what? (average voter is happy about it)

I say its successful because almost everyone I've talked to about it thinks its a great idea. The only people complaining about it are Libertarians. Seemingly on the grounds that any government spending program whatsoever is bad.

Now I don't believe that government spending is the answer to everything, but I think this particular program is a good idea.


"Yes Mike, the program appears to be incredibly successful all around. And best of all, the money to pay for it didn't have to be forcibly taken from millions of taxpayers to subsidize others' purchases! Like all good government programs, the money was simply generated out of thin air by the good-intentions fairy! What could be better?"

Is the CFC program raising your taxes? If you believe government deficit spending will raise your taxes in the future, that's your problem, not mine. Its an ideological argument that doesn't bear out in real life. Like it or not the govt has been deficit spending for years and tax rates have stayed (within reason) the same. How that possible? Some combination of long-term inflation and overall economic growth.

Now if you have an issue with that last sentence, then we can have a meaningful discussion, because you might have a point there. But if your argument is just "taxes bad, spending raise taxes, govt bad" then I don't know what to say.

A final comment:

I would call myself a progressive, but I come here for a reason. I come here because I agree with some of what is said by the bloggers here. In particular I agree with Arnold Kling's recent posts about this particular recession being a recalculation rather than something cyclic. I don't subscribe to the belief that any spending the govt does is intrinsically bad. I'm sure not all of you think that, but many of the commenters here talk that way.


El PResidente writes:

David,

If we have to "destroy wealth" to restore employment and incomes, that's unfortunate. Please pass me the hammer.


Michal,

That money is taken, in part, from me!

Which part? I'd like to know how proportionately outraged I should be.


hutch,

*[hutch: only consumer who have the means to purchase new cars. consumers who are in the markt for a used car will pay higher prices because there are fewer used cars on the road.]*.

Only the engines are destroyed (seized). The other parts may very well stock salvage yards, keeping many existing cars serviceable with used parts for many years to come. I don't think this will impact the used car market very much immediately, and in a few years many of these newer, more efficient vehicles will be sold into that market for others to buy. It accelerates the transition to greater fuel efficiency for those who will genuinely be in the market for a used car. Their next step up in fuel economy will be a bigger one, on average.


Justin,

By scrapping the "clunker," you end up keeping cars on the road that are far bigger polluters because the owners of these polluters cannot can't afford a new car even with a $4500 voucher.

Are we going to use Civic parts on Suburbans? Even if we keep older cars on the road longer, they will be identical to the ones we have taken off, unless we are going to build Frankenstein cars from the junk parts. I don't think your average Joe is going to do something like that. Are you suggesting that even though the models of cars will be identical their condition will deteriorate and with it their average fuel economy?

Troy Camplin writes:

Joe,

Nobody is "in charge" of GM that way. That was one of the few economic realities Steinbeck got right in "The Grapes of Wrath." GM, like every corporation of its size, has a system that runs pretty much on automatic. Information comes in that x number of units were sold at their dealerships, indicating that there should be a production increase. It would take a considerable amount of direct intervention to override the system set in place. Further, it would take this kind of economic reasoning to understand what is really going on and, sadly, very few people (and likely nobody at GM) undergo such reasoning and, therefore, understand what is really going on. So, no, GM doesn't know that the information it is receiving is bad -- it is an information-processing system, but an unknowing, unreasoning one.

ERV506 writes:

Dave (or any commenters), I have a (somewhat unrelated) mental exercise which asks a question relating to GDP (flow) and wealth (stock).

Suppose there are two identical islands, A and B, equally endowed in all respects. Suppose that in one year, each island builds exactly one widget, equally valued at $100. Thus, annual GDP would be $100 in both A and B. Suppose that the only difference between them is that a hurricane wipes out the widget in island A every year, while they keep accumulating in island B. Assuming away depreciation, after 5 years, total wealth would be $0 in island A (after the hurricane in the fifth year) and $500 in island B, yet both would have $100 worth of GDP every year.

This is nothing more than an application of Bastiat's broken window. My question then is, is this a correct criticism of using GDP as a measure of welfare? While "cash for clunkers" might boost measured GDP, its net effect would be wealth dimishing.

Comments/criticisms?

TomB writes:

This may not be the most novel observation, but this is the broken window fallacy only with the government doing the breaking. Maybe the broken window fallacy should have a little brother: the destroyed car fallacy. Perhaps someone more clever than I could come up with a better name.

Also, the government has hired dealerships to do the breaking. And there is paperwork. And I am told that the government is behind on payments. And the government had to triple the initial budget. And apparently there are unintentional consequences: some charities get a substantial portion of their budgets from car donations, and anticipate getting fewer cars. (Cafe Hayek has stories on many of these things)

Shayne Cook writes:

I'll be interested in seeing the repossession rate following/associated with this government sponsored "surge" in auto purchases.

It will be interesting to see if it at all correlates with the housing foreclosure rate associated with government influences on the housing bubble.

Radford Neal writes:

I say its successful because almost everyone I've talked to about it thinks its a great idea. The only people complaining about it are Libertarians. Seemingly on the grounds that any government spending program whatsoever is bad.

Have you actually read the post and the subsequent comments? Certainly there is a standard, doctrinaire libertarian argument against the program. But that's not the argument the post and most comments have been making. We're arguing that it's a bad program because it does not actually accomplish any reasonable economic or environmental objective.

Of course, if you're right about most people thinking it's a great idea, it's accomplished a political objective, which is all that its originators care about.

Jacob Oost writes:

mike: Government, shmovernment. The reason this program is a bad idea is for a few very simple reasons:

1) All it does it shift resources around with no new wealth being created. It represents, AT BEST, a lateral move in the nation's stock of wealth. In reality, it represents a loss due to the inherent inefficiencies involved in bureaucracy versus a large group of unorganized people spontaneously deciding to spend money in their own way. Economic growth is the natural output of voluntary exchanges between individuals, i.e. trade on the free market. You cannot "create" economic growth by artificially creating coerced exchanges. That is putting the cart before the horse.

2) It causes people to spend money on cars rather than on other things that, in the absense of a subsidy, they would have bought instead, all because the full cost of the car is not passed onto them, it is passed directly to the tax-payer. Read what Nathan said, and then refute it. If you can't do that, then stop talking about this.

3. It is the broken window fallacy in action. It requires the tax-payer to take a loss (a broken window), in order to "stimulate" the economy. This is a fallacy, it's been debunked many times, in theory and in practice. It doesn't matter how many hack economists can cook the numbers enough to make it seem to "work," it's a bunch of hooey.

4. It serves an industry with political connections (why not a cash for betamax players program? Cash for old computers?), giving politically-connected industries less incentive to be efficient, and creating the expectation that the government will bail out politically connected industries. Unfortunately, politically connected industries have had that incentive for years, but this just exacerbates it.

5. It literally destroys cars. Cars that would have had more value AS CARS are now sold for scrap. That is a loss of wealth.

However many special interests are made "happy" by a program at not the measure of the success of a program.

The Cupboard Is Bare writes:

The destruction of the engines and trannies is a major loss to salvage yards. Some engines can go for as much as $1,500.

The shortage of engines/trannies may very well result in the junking of cars simply because they can't get the parts. Auto collision shops frequently include used parts in their estimates. This can sometimes make the difference between a car staying on the road or being "totaled".

zach writes:

"perhaps GM is replacing inventory to remain at a certain level. I think they read the newspapers and have a good understandin that demand is being pushed forward."

Ha, ha! Your assumption that GM (or at least the decision makes of the company) reads the newspapers and makes intelligent decisions with regards to future demand. That's laughably ignorant of recent history on your part. If GM was so good about predicting future demand, how do you explain them having record numbers of cars floundering on lots last year, going bankrupt, and needing various government bailouts (including the CARS program)? Seems like maybe they're not very good at predicting future demand.

Thanks for your brilliant insight though.

Chris writes:

The article at the following url provides a good framework for those interested in calculating the remaining piece of David's analysis, the carbon benefit of the CARS program.

By the writer's rough estimation, the program is effective in reducing carbon output, including the scrapping costs, but as is typical for government programs the benefits could have been had much more cheaply.

http://www.slate.com/id/2224306/

Joe writes:

Zach,

Snide comments aside, working in industrial supply chain management, these input do impact decisions, recent history included. Nobody is great at predicting future demand. But you make your best guess using all information available. See Wal-Mart...

Justin writes:

El PResidente,

I don't believe that it is reasonable to expect that the cars we're taking off the road via Cash for Clunkers are identical to the cars that are being kept on the road as a result of the program. The people that have "clunkers" and also have the means to afford a new car are generally going to be relatively well off families. Their cars are generally going to be comparatively well maintained and will probably get something approaching their rated mileage.

If one of these families turns in a $3000 or $4000 car, that car would normally then generate one or more used car purchases that ends with someone that trades in a $500 pile of junk. The $500 pile of junk is likely to be poorly maintained, it probably lacked a catalytic converter when it was built, it has probably had more than its share of fluid leaks, etc. It's the car whose emissions you can see and smell when driving behind it that is ultimately going to be kept on the road longer because the "clunkers" like a 2000 Camero aren't available in the used car market.

El Presidente writes:

Justin,

That is a good point, and I cannot disagree. I think the magnitude of this effect becomes important to your case. I also think the value added for the automobile owners who cannot afford a new car and must make their old cars last longer should be included in our conisderation of the policy's "wealth destroying" effects.

This prompts some additional questions with respect to those who will be nursing older vehicles. Will their fuel economy decline even more? How much? What are their alternatives? What are the effects of the alternatives? Will these things change in the foreseeable future as a result of this policy?

tjames writes:

Assuming Cars for Clunkers was going to happen, it seems far more benefit could have been derived (or far more loss avoided) if the cars traded in had, instead of being destroyed, simply been traded to folks driving the worst cars now on the road, and then those cars destroyed. Yes, this is another round of subsidy, but if government already "owns" the cars, everything else is sunk cost, and this subsidy has as its only alternative the nearly complete destruction of the wealth in hand. At this point, its about getting the most benefit out of the first trade that has already been made. Get the serious clunkers off the road.

El Presidente writes:

tjames,

Excellent suggestion!

The Sheep Nazi writes:

Surely the best refinement of this idea would be simply to have the unemployed UAW workers remove the engines from the clunkers, and then put them right back in again. This way at least no valuable equipment is destroyed, and El Prez gets to swing his job-creating hammer too.

El Presidente writes:

The Sheep Nazi,

Cute, and not too far off the mark. If you wanna go down that road, I would invoke Keynes' suggestion of digging holes and filling them back up. Sooner or later, I think even the devout libertarians here will have to recognize that our current troubles are the result of lagging demand which is inexorably linked to the distribution of wealth and therefore the flow of income. I don't hear many libertarians saying, "This recession is what I always wished for." I said it would be unfortunate to have to destroy wealth to create jobs, and I meant it, but wealth is not destroyed through downward redistribution. A dollar is a dollar is a dollar. Neither your pocket nor mine can impart greater significance to it. If we would allow more downward redistribution, we could have less destruction. Or, we can insist that people must waste their time and effort so that they can demonstrate that they have "earned" their income. Either way, the money is going to have to go from the top to the bottom, in order to go back up again. Choose your medicine, hold your nose, and swallow.

Troy Camplin writes:

Money and wealth are two different things.

Tom Brady writes:

The boost to the OEMs (car builders) is arguable because they have already made their money, as have the parts suppliers. The assembly line is paid to build the car regardless of demand for the car. Dealers pay corporate to have the vehicle on their lot.

Enter C4C. Government doles out a reimbursemnt check to the dealer for giving a customer $3,500 or $4,500 toward a new vehicle. The dealer gets back their money and they are also out spared from the cost of having that vehicle on the lot. The profit is minuscule. If the vehicle returns for maintenance, then they make some solid profit.

Alternatively, had they given the customer $2,500 for the trade and flipped it for $3,500 or $4,000, there is $1,000 to $1,500 gross profit for maybe the cost of a detail and oil change.

The excitement will wear off when these dealers review their YTD sales and don't see an increase.

Comments for this entry have been closed
Return to top