Akerlof's 'market for lemons' demonstrates how information asymmetry about quality unravels markets. Since buyers cannot distinguish high from low quality, they offer average prices. High-quality sellers exit, lowering average quality. This iterative process continues until only low-quality (lemons) remain. The potential welfare gains from trade may fail to be realized entirely when quality is unobservable and heterogeneous.
From your work on asymmetric information and adverse selection, you know that when one side of a transaction has private information, market outcomes can diverge sharply from the efficient benchmark. Akerlof's 1970 "market for lemons" paper provides the canonical demonstration of how this divergence unfolds — not as a static inefficiency but as a dynamic unraveling that can destroy an entire market.
Consider the used car market. Sellers know whether their car is high quality ("peach") or low quality ("lemon"), but buyers cannot tell the difference by inspection. Suppose peaches are worth $10,000 to sellers and $12,000 to buyers, while lemons are worth $5,000 to sellers and $6,000 to buyers. If quality were observable, both types would trade — there are gains from trade for each. But buyers cannot distinguish types. If the market has half peaches and half lemons, a risk-neutral buyer offers the average value: $9,000. At this price, lemon sellers happily participate (they receive $9,000 for a car worth $5,000 to them), but peach sellers refuse (they would receive $9,000 for a car worth $10,000 to them). The peaches withdraw. Now the pool is all lemons, buyers revise their offer down to $6,000, and the market unravels to contain only low-quality goods. The mutually beneficial trades in peaches — worth $2,000 of surplus each — never happen.
The unraveling is an instance of adverse selection: the composition of the market is adversely affected by the terms of trade. Higher prices attract worse risks (or in this case, lower prices drive out better quality). The mechanism is iterative and self-reinforcing. Each round of seller exit worsens the average quality of remaining sellers, which lowers the price buyers are willing to pay, which drives out the next tier of quality. In the worst case, the market disappears entirely. The key insight is that the problem is not simply that buyers are uninformed — it is that sellers' participation decisions are correlated with their private information in a way that undermines the market.
Real markets have developed numerous institutional responses to the lemons problem, and recognizing them deepens your understanding of why markets are structured the way they are. Warranties are a form of signaling: offering a warranty is cheap for a peach seller (the car is unlikely to break) but expensive for a lemon seller (it will). Certified pre-owned programs, inspection regimes, reputation systems, and return policies all function as mechanisms to credibly convey quality information or shift risk. Signaling (as in Spence's job market model, which you studied in adverse selection) and screening (as when insurers offer menus of contracts to sort risk types) are the theoretical counterparts. Akerlof's model explains not just market failure but why so much of market infrastructure — brands, certifications, intermediaries, regulatory standards — exists precisely to prevent the unraveling that would occur if buyers and sellers interacted with no quality information at all.