Brad Delong has a defense of the clunker program based on its environmental benefits. However, I am not completely certain its a bad program on the basis of its economic benefits.
Now, Cash for Clunkers sounds a like an application of the broken window fallacy – that we can improve the economy by destroying things which then have to be re-created. While this is clearly a fallacy during full employment, its not clear to me that its a fallacy during recession.
The difference is that during a recession the service stream from labor and capital are being destroyed anyway. You have workers with no factory and factories with no workers.
The workers produce a service stream, their labor, that is being effectively destroyed. The work an unemployed worker would have done today is just gone and can never be recovered. Much the same is true for the capital he would have used. Though, the calculation with capital is a bit different since presumably it depreciates faster with use.
If instead this labor and capital is employed fixing windows or building cars then it is not lost. So, if I destroy a clunker the economy looses a car but I prevent the loss of capital and labor services.
If it were a one for one trade – every clunker destroyed represented a new car built – then this would be a no brainer. Destroying the clunker would definitely be worth it.
However, many of the people using the program would have bought a car anyway so we have to destroy multiple clunkers to get one new car. There is some tipping point where destroying clunkers would not be worth it but I am not sure how close we are to that point.
The program has additional value, however, because it lowers the potential downside risk for auto dealers and potentially auto manufacturers. This in turn should reduce their incentive to hoard cash, which in turn is simulative itself. That is, the clunker program should have multiplier effect.
Taken together this means that we could probably destroy quite a few clunkers per new car built and still be in the positive economically. How many clunkers we are in fact loosing, I don’t know.