Suppose most used cars on sale are fine, and a few are bad. You cannot tell which is which by looking. The seller can, because they have driven it for years. That one gap in knowledge is enough to make a whole market shrink, even though every single person in it would be happy to trade.
In 1970 the economist George Akerlof published a short paper called “The Market for ‘Lemons’” in the Quarterly Journal of Economics. In American slang a “lemon” is a car that turns out to be bad. His argument is simple. A buyer who cannot see quality will only pay for the average car. But an owner of an above-average car knows it is worth more than that, so they keep it or sell it some other way. The cars left on the market are now worse than before, so the average drops, so buyers pay less, so the next-best owners leave too.
The paper had a rough road. It was rejected by three journals before the Quarterly Journal of Economics accepted it, and in 2001 Akerlof shared the Nobel Memorial Prize in Economic Sciences with Michael Spence and Joseph Stiglitz for work on markets where one side knows more than the other.
Below are 100 used cars, lined up from the worst on the left to the best on the right. The height of each bar is what that car is worth to its owner, from nearly nothing up to €10,000. Buyers like cars more than owners do, by a factor you choose. But buyers cannot see quality, so they offer one price: their premium times the average value of the cars still for sale. The dashed line is that offer. Any owner whose car is worth more than the line takes it off the market. Press Next round and watch.
With a premium of 1.5, buyers are willing to pay half as much again as the owner values each car, and yet the market empties out. The first offer is 1.5 times the average of all 100 cars, which is €7,500, so every owner whose car is worth more than that leaves. The remaining cars are worse, the next offer is lower, and the slide continues round after round until almost nothing is left. Nobody was tricked and nobody was greedy. Each owner just made a sensible choice, and the sum of those choices wrecked the market.
Now drag the slider up. When buyers value cars at exactly twice what owners do, the offer always lands right at the top of the cars on sale, so no one is pushed out. Above that, everyone sells. The breaking point depends on the numbers, but the lesson does not: the bigger the gain from trading is compared with the spread in quality, the safer the market.
Real markets are less extreme. Some people have to sell quickly, and cars differ in ways you can see. Akerlof's point is about the pressure, which pushes quality down and thins out trade whenever buyers cannot tell what they are getting.
Press Let buyers inspect. Now buyers can see each car's true value and pay for it individually, and every car sells at once. This is what most fixes try to do: give the buyer a way to know. Mechanic inspections, service histories, vehicle-history reports and warranties all work this way. A warranty is especially clever. A seller who offers one is willing to pay for repairs, which only a seller with a good car can afford to do, so the offer itself is a signal. Brands and reviews do the same job in other markets, because a seller with a name to protect has a reason to be honest.
The same problem appears wherever one side knows more. Akerlof's paper also discussed health insurance for older people, where those who most expect to need care are the most eager to buy, and lending in poor countries. In each case the hidden quality can push good participants out and leave the rest behind.
Markets do not just need buyers and sellers. They need trust, or a way to check. When quality is hidden, the average sets the price, the best leave first, and the average falls again. The cure is not to wish for honesty but to make quality visible.