Tuesday 22 July 2008

The lemons market (updated)

The seminal paper on the lemons market was by George Akerlof - The market for "Lemons": quality uncertainty and the market mechanism, Quarterly Journal of Economics 84 (1970), pp. 488–500. This is the paper that won him the Nobel Prize. His result was that asymmetric information between the buyer of a used car and seller of that car can cause failure in the proper functioning of the market because the seller of a used car knows the quality of his car but the buyer is unable to discern good from bad cars. The upshot being that only bad cars get sold, good cars are driven out of the market.

In a recent paper (Lemons hypothesis reconsidered: An empirical analysis) in Economic Letters, Arif Sultan sets out to test this idea. He argues that used cars, if inferior to new cars, would require higher maintenance expenditures. His paper tests the hypothesis that there is no difference in the average maintenance expenditures required for cars acquired used and those acquired new.

The results he gets show that there is no evidence that cars acquired used required more maintenance expenditures than those of a similar age acquired new. The conclusion to the paper reads, in part,
The purpose of this paper was to examine the difference in the quality between cars acquired used and those acquired new. I measured the quality of the car by using maintenance expenditures incurred on a car [... ] I found that cars acquired new required the same maintenance expenditures as those acquired used, all else being equal, implying that cars acquired used are of same quality as cars acquired new of a similar age. If cars acquired used were of lower quality, they would have required more maintenance expenditure.
Update: The Visible Hand in Economics comments on The lemon hypothesis vs evidence and points us to other comments at Division of Labour and Marginal Revolution.

8 comments:

Anonymous said...

Which just means the asymmetric information problem in that market has been addressed.

A mildly interesting question is how it has been solved - is it private or public institutions that are mainly responsible?

Matt Nolan said...

"Which just means the asymmetric information problem in that market has been addressed.

A mildly interesting question is how it has been solved - is it private or public institutions that are mainly responsible?"

Not necessarily. It could just mean that the sellers of the best quality cars value the cars at a lower rate than the price that corresponds to the market with uncertainty.

If this is the case we don't end up with the lemon problem - but the asymmetric information between the buyer and seller would remain.

Paul Walker said...

"It could just mean that the sellers of the best quality cars value the cars at a lower rate than the price that corresponds to the market with uncertainty."

But in the standard model, how is this possible? If buyers are willing to pay the expected value of a car, then the good car seller's valuation has to be higher than the buyer's offer. And why would a buyer be willing to pay more than the good car sellers valuation given that the buyer doesn't know the true state of the car?

Matt Nolan said...

"And why would a buyer be willing to pay more than the good car sellers valuation given that the buyer doesn't know the true state of the car?"

Because buyers and sellers don't receive the same value from the same product, ergo they don't have the same utility function - that is why we would still have trade in a world of equally endowed individuals.

Paul Walker said...

"Because buyers and sellers don't receive the same value from the same product, ergo they don't have the same utility function - that is why we would still have trade in a world of equally endowed individuals."

I agree, in fact trade is more likely when they have large disagreements over the value of the good, but I don't see that helping here. If buyers do value the good cars more than the sellers and they offer to buy at their valuation then both good and bad car owners will offer to sell to them and they are back where they began. They can't tell the cars apart, so there is a risk they will buy a bad car, so why would they offer such a high amount?

What could happen in this case is that because buyers value cars more than sellers they could be willing to pay to have, say, the AA check the car to see if it is in fact good. Buyers valuation would have to equal, at least, sellers valuation plus cost of AA check.

Matt Nolan said...

"They can't tell the cars apart, so there is a risk they will buy a bad car, so why would they offer such a high amount?"

It all still depends on their relative valuation.

If a seller has a reservation value for a good car of 2k and a reservation value for a bad car of 4k, while a buyer has a maximum willgness to pay of 5k for a bad car and 7k for a good car then we will have a fully functioning market even if the asymmetric information issues associated with the lemons model still held.

In fact, as long as the expected value to the buyer is greater than the reservation value of the seller of the highest quality car, then the entire market will still function.

Paul Walker said...

"In fact, as long as the expected value to the buyer is greater than the reservation value of the seller of the highest quality car, then the entire market will still function."

True, but this isn't the model Akerlof used in the lemons problem. Also it doesn't explain the no difference in maintenance expenditures results since under the above assumption low quality cars could still have higher maintenance expenditures.

Matt Nolan said...

"True, but this isn't the model Akerlof used in the lemons problem."

It is the same general model - however, given the realised set of parameters in the used car market the market failure result does not occur.

This evidence does not imply that the asymmetric information problem has been addressed - which was the point I was initially addressing by commenting :) .

"Also it doesn't explain the no difference in maintenance expenditures results since under the above assumption low quality cars could still have higher maintenance expenditures."

The Sultan paper stated that it adjusted for the age of the vehicles right - in which case if we had no market failure we would see no difference in maintenance expenditures - which we didn't. However, this simply tells us that we don't have a market failure - it doesn't tell us that the general asymmetric information model does not hold.

I agree that the evidence shows that we don't have a market failure. However, I don't think the evidence is sufficient to tell us that the asymmetric information prevalent in the market has been addressed. This only matters insofar as it implies the market may be unstable in the face of an exogenous shock in preferences.