Asymmetric Information

Information failure

The competitive market model assumes perfect information, that buyers and sellers know the quality, price and consequences of what they are trading. When that assumption fails, the market can allocate resources inefficiently even with no externality, no market power and no public-good problem.

▶

Asymmetric information exists when one party to a transaction has more or better information than the other. The imbalance is the problem, not ignorance in general: if both parties are equally uninformed, there is uncertainty but no asymmetry.

The result is that mutually beneficial trades fail to happen, or that the wrong trades happen, either way, a welfare loss.

Where the welfare loss sits. Because buyers cannot judge quality, they value the good using perceived rather than actual benefit. The demand curve the market responds to is therefore the wrong one, and the quantity traded differs from the quantity that would maximise welfare. Where quality is understated, the market under-consumes; where it is overstated, as with a product whose harms are hidden, it over-consumes. Either way the outcome has the same shape as an externality diagram: two curves, two quantities, and a triangle between them.

Adverse selection

Asymmetric information before the transaction, about the characteristics of the good or the person.

The classic case is the market for used cars. Sellers know whether their car is reliable or defective; buyers cannot tell them apart.

1. Buyers cannot distinguish good cars from bad
2. they will only pay a price reflecting average quality
3. that price is below what a good car is worth, so owners of good cars withdraw from the market
4. average quality falls
5. buyers revise their offer down further
6. the process repeats.

In the extreme the market unravels: only the worst goods remain for sale. This is why the phenomenon is often called "the market for lemons".

The same logic drives insurance markets. People who know they are high-risk are most eager to buy insurance. If the insurer cannot distinguish risks; it must charge an average premium, which is a bad deal for low-risk customers, who drop out, raising the average risk of the remaining pool and forcing premiums higher still.

Note what makes this a market failure rather than merely a bad deal for somebody. Good cars exist that buyers would happily pay for at their true value, and low-risk people exist who would happily insure at a fair premium. Those trades would make both parties better off, and they do not happen. The welfare loss is the value of transactions that never occur.

Moral hazard

Asymmetric information after the transaction, about behaviour.

One party changes their behaviour once they are insulated from the consequences, and the other party cannot observe or control it.

1. Once insured, the cost of a loss falls on the insurer rather than the insured
2. the incentive to take care is reduced
3. the insured takes more risk than they otherwise would
4. losses rise, and premiums must rise for everyone.

Examples: a fully insured driver parking carelessly; a bank taking excessive risk believing it will be rescued if it fails; an employee working less hard when effort cannot be monitored.

That last case has its own name worth knowing: the principal–agent problem, where one party (the principal, an employer, a shareholder, a voter) delegates to another (the agent, an employee, a manager, a politician) whose actions they cannot fully observe and whose interests may differ. Performance-related pay, share options and auditing all exist to align the agent's incentives with the principal's, and all are imperfect.

Telling them apart

The distinction is heavily tested and easy to secure:

Responses

By governments

• Regulation and mandatory disclosure: food labelling, financial product disclosure, safety standards, licensing of professionals.
• Legislation against misrepresentation, giving buyers legal recourse.
• Direct provision or compulsory insurance. Making health insurance universal removes adverse selection entirely, because the pool includes everyone rather than only those who expect to claim.

By private markets

Markets often develop their own remedies, and noting this is a strong evaluative point, information failure does not always require government action:

• Signalling: the informed party credibly reveals quality: warranties, guarantees, brand investment, professional qualifications. A warranty works because it is costly for a seller of a defective good to offer.
• Screening: the uninformed party induces the other to reveal information: insurers offering a menu of policies with different excesses, so low-risk customers self-select into the high-excess option.
• Third-party information: reviews, ratings agencies, independent inspections, comparison sites.
• Excesses and co-payments: leaving the insured bearing part of any loss preserves the incentive to take care, directly addressing moral hazard.

Why a signal has to be costly. The reason a warranty works is not that it reassures buyers but that a seller of a poor product could not afford to offer it, the claims would ruin them. A signal any seller could send carries no information at all, which is why "we care about quality" on a website persuades nobody and a five-year guarantee does. The same logic explains why qualifications signal ability even when the content is never used at work.

Real-world examples

• The 2008 financial crisis is the standard case for both failures at once. Mortgage originators knew more about borrower quality than the investors who bought the repackaged debt, adverse selection, while banks that expected rescue had reduced incentive to control risk, which is moral hazard. Ratings agencies were the third-party remedy that failed, partly because they were paid by the issuers whose products they rated.
• Credence goods: car repairs, dentistry, legal advice, are where the asymmetry never resolves: the buyer often cannot judge the quality even after consuming it. Licensing and professional regulation exist mainly for these.
• Second-hand markets today show the private remedies working: independent inspection reports, seller ratings and return windows have reduced the lemons problem substantially without any of them being compulsory.

Worked example

An insurer offers health insurance at a single premium based on average population risk.

1. People who know they are in poor health value the policy above the average premium and buy
2. healthy people value it below the premium and decline
3. the insured pool is riskier than the population average
4. claims exceed those priced in
5. the insurer raises premiums
6. the healthiest remaining customers now drop out
7. the pool deteriorates further.

This is adverse selection, and it can shrink the market until only the highest-risk individuals are covered at premiums few can afford.

Then moral hazard compounds it. Once insured, some customers use more medical services than they would if paying directly, since the marginal cost to them is near zero. Claims rise further.

Responses. Against adverse selection: compulsory universal insurance, which forces the pool to include low-risk people and removes the self-selection entirely; or screening through medical questionnaires and risk-based pricing. Against moral hazard: co-payments and excesses, restoring some cost to the insured at the point of use.

Evaluation. Risk-based pricing solves the economic problem but raises an equity problem: those most in need of insurance face the highest premiums, or become uninsurable. That trade-off between efficiency and equity is the judgement a 15-mark question wants, and it is why many countries treat health insurance as a matter for government rather than markets.

Note the two remedies are not interchangeable. Compulsory pooling fixes who is in the market and does nothing about behaviour once they are in it; co-payments fix behaviour and do nothing about who joins. A system needs both, and an answer that offers one as a complete solution has only addressed half the problem, which is exactly what the adverse selection/moral hazard distinction is testing.

Common exam mistakes

• Confusing adverse selection with moral hazard. Remember: before/characteristics versus after/behaviour.
• Describing plain ignorance rather than an asymmetry between the two parties.
• Assuming only government can respond, and ignoring signalling and screening.
• Forgetting that compulsory insurance addresses adverse selection but not moral hazard.
• Treating any reassurance as a signal, a signal must be costly to fake.
• Stating that the market fails without explaining the mechanism by which good products or low-risk customers are driven out.

Exam technique

Name which type of information failure the scenario shows and justify it with the timing test, did the hidden information exist before the deal (characteristics) or arise after it (behaviour)?

Then run the unravelling chain explicitly, because the marks are in the mechanism, not the label.

If a diagram is wanted. Draw the perceived and actual benefit curves as two demand curves, mark both quantities, and shade the triangle between them, the structure is the same as an externality diagram, with information rather than a third party causing the divergence.

For evaluation, weigh government responses against market solutions. Note that regulation has enforcement and compliance costs, and consider the efficiency–equity trade-off that risk-based pricing creates.

Quick revision

• Asymmetric information: one party knows more than the other.

Adverse selection: before the deal, hidden characteristics

good products or low-risk customers driven out.

Moral hazard: after the deal, hidden actions

riskier behaviour once insulated from consequences.

• The principal–agent problem is moral hazard inside an organisation.
• The welfare loss is the value of the beneficial trades that never happen.
• Government responses: regulation, disclosure, compulsory or direct provision.
• Market responses: signalling (warranties, qualifications), screening (policy menus), third-party ratings.
• A signal only works if it would be too costly for a low-quality seller to imitate.
• Excesses and co-payments target moral hazard; compulsory pooling targets adverse selection. You need both.
• Risk-based pricing is efficient but raises equity concerns.

Not the topic you were looking for? Describe what you are stuck on in your own words and we will take you to the notes that answer it.