Government is not very good at price discovery. That’s one lesson, I think, of the cash for clunkers program. Whoever set the rebate at $3,500 and $4,500 (let’s average it to $4,000 for illustrative purposes) evidently calculated that 250,000 car owners would trade in their vehicles for new cars that get at least four miles per gallon more between July and November. That calculation proved to be hilariously wrong. Washington Post media reporter Howard Kurtz asks the question, “Also, isn’t is apparent that the $4,500 payments for older gas-guzzlers was extremely generous? That’s a huge chunk of change to spur people who probably would have bought a new car anyway.”
Well, yes. If Congress had set the rebate at $2,000 rather than $4,000, it might have netted 500,000 trade-ins over the four-to-five month period—or maybe fewer or maybe the $1 billion would have been exhausted earlier. The fact is that price discovery—figuring out how big a rebate is needed to produce the desired number of transactions—is very difficult for even the most competent of individuals, much less for Congress. Look at it this way. If Congress had set the rebate at $100,000, almost all of us would have rushed to trade in our cars, and the $1 billion would have probably have been exhausted in 24 hours. Only people with net worths on the order of Bill Gates could afford to ignore an incentive like that. If Congress had set the rebate at $25, almost no one would bother to trade in a car except those who would have done so in any case.
So Congress needed to set the rebate at somewhere between $25 and $100,000, and not surprisingly it got the number wrong. Markets are good at price discovery; government isn’t. If the House bill adding $2 billion in stimulus funds to the cash for clunkers program passes the Senate, the government will spend $3 billion for 750,000 trade-ins. If Congress had set the rebate at $1,333 instead of $4,000, it might well have had to spend only $1 billion for the same number of trade-ins. In which case Congress will have spent an extra $2 billion for no good reason. Note that I am assuming that it’s a good idea to spend government money to induce such trade-ins or subsidize those which would have occurred without any rebate.
My sense is that voters had got this kind of thing figured out. Pollster Scott Rasmussen reports that voters opposed cash for clunkers by a 54%-35% margin and that now they oppose spending additional money by an almost identical 54%-33% margin. Why should we let Congress which couldn’t design an intelligent cash for clunkers program redesign the health care system which comprises one-sixth of our economy? It’s a very good question which members of Congress are already hearing from their constituents.
