Markets work when both sides know what is being traded. When one side knows more (about the quality of a car, the riskiness of a borrower, or how hard an employee is working), markets can shrink, misprice or collapse. This guide sorts every version of the problem by one question: is the information hidden before the deal, or is an action hidden after it?
The seller knows the car is a lemon; the insurance buyer knows they are high-risk; the borrower knows the project is shaky. The other side can't tell types apart, so the wrong types are drawn in. This is adverse selection.
Once insured, you lock your bicycle less carefully; once hired, an employee may shirk; once lent money, a borrower may gamble with it. Behaviour changes because someone else bears the cost. This is moral hazard.
Adverse selection is about types hidden before a contract; moral hazard is about actions hidden after it. Almost every exam question in this topic is answered by placing the example on the right side of that line.
The uninformed side attracts the worst types. Akerlof (1970).
The informed side proves its type with a costly action. Spence (1973).
The uninformed side designs choices that make types reveal themselves. Rothschild–Stiglitz (1976).
Insulated from the consequences, people take less care or more risk. Arrow (1963).
One party (the principal) hires another (the agent) whose actions it can't fully observe, and their interests differ. Jensen and Meckling (1976).
George Akerlof's 1970 paper (Nobel Prize 2001) asked why a car loses so much value the moment it leaves the showroom. His answer: used-car buyers can't tell good cars (“peaches”) from bad ones (“lemons”), so they pay only for average quality. Owners of good cars won't sell at that price, which lowers the average, which lowers the price again. Step through it:
Buyers can't tell good cars from bad, so they offer 1.2× the average quality on sale, 550: that is ₹660 thousand. Owners of the 4 best cars (700, 800, 900, 1000) would be selling at a loss, so pull out, and the average quality falls.
Bad quality drives good quality out of the market, a version of Gresham's law. Even though every single trade would benefit both sides, hidden information can shrink the market to its worst products, or kill it entirely. The price falls and the average quality traded falls.
At a price based on the average person, the healthy find it poor value and drop out, the pool gets sicker, and premiums rise. It is why people over 65 can struggle to buy cover at almost any price, and why many countries make insurance compulsory.
Raising the interest rate drives away safe borrowers first and attracts risky ones who expect not to repay. So banks may refuse loans rather than raise rates: credit rationing (Stiglitz and Weiss, 1981). It helps explain why informal moneylenders, who know their borrowers, lend where banks won't.
If an employer offers a low wage, the most productive applicants, who have better options, don't apply. Paying above the market wage can attract better workers.
A high-quality type wants to prove it. Words are cheap, anyone can say “my car is good”, so the proof must be an action that is costlier for low types than for high types. Then only high types find it worth taking. Michael Spence's (1973, Nobel 2001) example is education: it signals productivity even if it teaches nothing, because it is easier for able people to complete.
Other signals: a warranty (costly to give on a bad product), a well-known brand, expensive advertising, a firm keeping a high dividend.
Here the side that lacks information moves first, offering a menu of contracts designed so that each type prefers a different one and so reveals itself (self-selection).
A cheap policy with a high deductible and a dear one with full cover. Low-risk drivers pick the first, high-risk drivers the second (Rothschild and Stiglitz, 1976).
Asking for collateral screens borrowers: those confident of repaying are the ones willing to pledge it.
Probation periods and performance-linked pay attract applicants who expect to do well and put off those who don't.
Signalling: the informed party acts first to reveal its type. Screening: the uninformed party acts first to make types reveal themselves.
Fully insured people take less care: fewer locks, less caution, more visits to the doctor. The insurer can't watch, so it pays for behaviour it didn't price.
A bank that expects a bailout (“too big to fail”) or has insured deposits can take bigger risks, keeping the upside while taxpayers bear the downside. A borrower may switch a loan to a riskier project after getting it.
A worker on a fixed salary whose effort can't be seen may shirk. That's why pay is often tied to output, or set above the market rate.
If a firm pays above the market wage, being fired for shirking becomes costly, so workers put in effort even when unobserved (Shapiro and Stiglitz, 1984). It also reduces turnover and attracts better applicants. A side effect: wages don't fall to clear the market, so some unemployment persists.
A principal needs an agent to act for them but can't see everything the agent does, and the agent has goals of their own. The task is to design a contract that makes the agent want what the principal wants. Two conditions:
The contract must be good enough that the agent accepts it at all, compared with their next-best option.
Given the contract, the agent must be better off choosing the action the principal wants (working hard, choosing the safe project).
| Principal | Agent | The conflict | Typical fix |
|---|---|---|---|
| Shareholders | Managers | Managers may prefer empire-building, perks or a quiet life to profit | Stock options, bonuses, takeover threat, board oversight |
| Landowner | Tenant farmer | Under sharecropping the tenant keeps only part of any extra output, so puts in less effort (Marshallian inefficiency) | Fixed rent, which gives the tenant the full return on effort |
| Employer | Employee | Effort is hard to observe | Piece rates, efficiency wages, promotion tournaments |
| Voters | Politicians | Politicians' interests may diverge from voters' | Elections, transparency, term limits |
| Problem | What is hidden | When | Classic example | Private fixes | Policy fixes |
|---|---|---|---|---|---|
| Adverse selection | Type (quality, risk) | Before | Used cars, health insurance, loans | Warranties, reputation, signalling, screening | Compulsory insurance, disclosure laws, certification |
| Moral hazard | Action (care, effort, risk) | After | Insured behaviour, bank risk-taking | Deductibles, co-pay, collateral, monitoring, incentive pay | Bank regulation, capital requirements |
| Principal–agent | Action and interests | After | Shareholders and managers | Performance pay, stock options, efficiency wages | Corporate governance rules |
One attempt each.
Hidden type before the deal: adverse selection. Hidden action after the deal: moral hazard.
Buyers pay for average quality, good sellers leave, average falls, price falls. The market can unravel.
Informed side acts. Works only if the signal costs low types more. Education, warranties, brands.
Uninformed side offers a menu so types self-select: deductible menus, collateral, probation.
Less care once protected. Fix with deductibles, co-pay, monitoring, collateral, incentive pay, efficiency wages.
Contracts must satisfy participation and incentive compatibility. Stock options, piece rates, fixed rent.
From UGC NET, IIT JAM, CUET-PG, JNU and the University of Hyderabad: adverse selection and the lemons market, signalling, moral hazard in lending, principal–agent and efficiency wages.
These are questions from the TestStop bank. A free account answers 10 a day, each with its worked solution; All-Access opens every set with no daily limit.