Principal-Agent Problem vs Market for Lemons
Principal-Agent Problem and Market for Lemons are two Game Theory & Information concepts in AP Economics that students often mix up. The principal-agent problem arises when one party (the agent) acts on behalf of another (the principal) but has different incentives and better information. The market for lemons is George Akerlof's model showing that when only sellers know quality, buyers offer average prices and good goods leave the market. Here is how they compare side by side.
Examples include shareholders (principals) and CEOs (agents), or voters and politicians. Because the agent may pursue its own interests, principals use contracts, monitoring, and incentive pay to align goals.
A buyer who cannot tell a good used car from a bad one will only pay something near the average value of what is on offer. That price is below what the owner of a genuinely good car would accept, so those cars are withdrawn, the share of bad ones rises, and the price buyers are willing to pay falls again. The process can feed on itself until only the worst quality trades, or until the market disappears, even though buyers would happily pay more for a good car than its owner would accept. This is adverse selection, a problem of hidden characteristics that exists before any deal is struck, which is what separates it from moral hazard, where the hidden behavior comes after.
Principal-Agent Problem vs the Market for Lemons: Hidden Action Against Hidden Type
| Principal-Agent Problem | Market for Lemons | |
|---|---|---|
| What is hidden | What the informed party does | What the informed party is |
| When it does its damage | After the deal is signed | Before the deal is struck |
| Standard name for it | Moral hazard | Adverse selection |
| Who holds the information | The agent, about their own effort or care | The seller, about the good's quality |
| The visible symptom | Less care than the other side is paying for | The best goods withdrawn at the pooled price |
| What fixes it | Incentive pay, deductibles, monitoring, ownership stakes | Guarantees, certification, inspection, reputation |
| Giveaway phrase in a question | Cannot observe how much care is taken | Cannot tell quality apart before buying |
One remedy manufactures the other problem
The sharpest illustration is a used-car guarantee, which cures the hidden type and then creates a hidden action. Suppose an oil change costs the owner 6 and cuts the chance of an engine failure from 25 percent to 5 percent, with a failure costing 90 to repair. An uncovered owner compares an expected repair bill of 0.25 times 90, which is 22.5, against 0.05 times 90, which is 4.5. Skipping the oil change raises the expected bill by 18 and saves only 6, so the owner maintains the car. Now hand that owner full repair coverage. Their own repair cost is zero either way, the 6 becomes a pure loss, and maintenance stops. The guarantee that made the sale possible has just made the car worse. The usual fix leaves some of the risk with the person taking the action. Attach a 40 deductible and the owner's expected out-of-pocket cost is 0.05 times 40, which is 2, with maintenance, and 0.25 times 40, which is 10, without it. Skipping now costs 8 in expectation to save 6, so maintenance resumes. That deductible is doing something no lemons remedy can do: rather than revealing a fact about the car, it changes what the owner chooses.
Why no contract can turn a defective car into a sound one
Notice the asymmetry in what money can buy. An incentive scheme works because effort is a choice, so raising the reward attached to the outcome the principal wants shifts the agent's decision. Quality in the lemons model is not a choice, so no payment schedule improves it. Offer a seller a bonus for the car being sound and nothing about the car has changed; all you have changed is who wants to claim it is sound. That is why the two problems call for different families of solution. A hidden type needs verification, meaning inspection, a guarantee whose cost differs by type, certification, or a reputation the seller would lose by lying. A hidden action needs exposure to consequences, meaning deductibles, commissions, equity stakes, or monitoring that makes the action observable. A quick check on any proposed fix is to ask whether it works by producing information or by producing incentives. Answers that aim an incentive scheme at a hidden characteristic, or an inspection at hidden effort, are the ones that fall apart under a follow-up question.
Frequently asked questions
Is the market for lemons adverse selection or moral hazard?
The market for lemons is the standard model of adverse selection, since the hidden feature is the quality of the car and that quality exists before anyone trades. Moral hazard would require a hidden choice made after the deal, such as a new owner neglecting maintenance because a guarantee covers repairs. Both can occur in the same used-car market on the same afternoon, which is why exam questions often bury one inside a scenario built around the other.
Can a principal-agent problem exist without asymmetric information?
A principal-agent problem needs the principal to be unable to verify what the agent does, so perfect observation would end it. If a manager could watch and prove every choice a worker makes, the contract could simply require the right choice and penalize anything else, and no incentive scheme would be necessary. Conflicting interests on their own are not enough, because a fully observable conflict is settled by writing the required action into the agreement.
What is the difference between hidden action and hidden type?
Hidden action means a choice one party makes after the contract begins, such as effort, care or risk-taking the other side cannot observe. Hidden type means a characteristic that was already true beforehand, such as a car's condition or an applicant's health. Timing separates them, and so do the remedies: a hidden type calls for verification, while a hidden action calls for putting some consequence back on the person making the decision.
Get AP Econ exam tips in your inbox
Occasional emails with study tips, new interactive graphs, and exam-season reminders. Free, no spam.
No spam. Unsubscribe anytime. Read our privacy policy.
Keep track of what you have studied
A free EconLearn account adds progress tracking, your quiz history, and achievements. Studying here is free either way, and there is nothing to pay for as a student.
Create a free accountAlready have one? Sign in
Last updated