Pattern/Economics/No. 0630
Moral Hazard
Your insurance pays the whole repair bill, so you accept the first quote instead of shopping around.
- Evidence
- Well established
- Read
- 6 min
- Links
- 8 connections
01You've seen this when…
- in life
Your health plan makes each physical therapy session nearly free. You book more sessions than you would if you paid the full fee.
- at work
A contractor bills your company for every hour worked under a contract that allows unlimited spending. When a complication appears, adding another week is easier than finding a cheaper approach.
- out in the world
Depositors expect the government to repay them if a bank fails. They choose the bank offering the highest interest without checking how risky its lending is.
02The idea
The arrangement changes what the person making the decision stands to lose. Insurance covers a bill. A guarantee protects a lender. A contract reimburses expenses. Once some consequences fall elsewhere, behavior can change too.
That response is moral hazard: protection from a consequence changes incentives in ways that can increase risk or costs borne by others. An honest person can follow every rule and still take less care because their personal cost has fallen. That lower cost can also lead them to spend less time comparing prices or use more services.
The key is changed behavior, not merely a transferred loss. If insurance pays for storm damage and the owner behaves exactly as before, that’s risk sharing.
Moral hazard describes an incentive problem. Its name can sound like a judgment about character. Often the protected person is responding sensibly to the deal they’ve been offered.
03Why it happens
- The personal price falls below the full cost. A service costing $200 may cost an insured customer only $20. The customer weighs the benefit against $20, while the remaining $180 falls on the insurer and ultimately its funding pool. More services become worth buying from the customer’s perspective.
- Prevention becomes less rewarding. If someone else pays most of a loss, spending your own time and money to prevent it saves you less. Protection can weaken precautions before an accident, not just increase spending afterward.
- The relevant choices are hard to supervise. An insurer can see a claim more easily than the care taken to prevent it. A client sees billed hours more easily than a contractor’s effort to avoid unnecessary work. This is often an information imbalance.
- Contracts leave gaps. Even observable behavior may be difficult to control through a practical contract. Specifying every sensible precaution or necessary treatment would be expensive, intrusive, and sometimes impossible. This connects moral hazard to the principal-agent problem: one party’s choices affect another party’s interests.
04A worked example
The RAND Health Insurance Experiment, conducted in the United States during the 1970s and early 1980s, assigned families to insurance plans with different levels of cost sharing. Some received free care; others paid part of their medical bills, subject to limits.
What it looks like People with more generous insurance use more medical care. In ordinary records, that could simply mean people expecting high medical bills choose better coverage.
What’s actually going on The random assignment helped separate that selection effect from the effect of coverage itself. People assigned to pay less out of pocket used more care and incurred higher medical spending. Changing the price they faced changed their behavior. That’s moral hazard in the economic sense, even when the additional care was valuable and nobody acted irresponsibly.
What would have helped An insurance designer needs to examine changes in spending together with changes in the kinds of care people receive. RAND found that cost sharing reduced both more and less effective care. Charging patients more across the board can therefore save money while deterring useful treatment. A better design would try to preserve access to valuable care while reducing low-value use, and test whether its incentives actually achieve that.
05How to spot it
06What to do about it
- Keep some consequences with the chooser. Deductibles, copayments, warranties, and performance-linked fees can restore a reason to consider costs. That’s skin in the game. But the retained loss must be affordable; exposing someone to financial ruin isn’t a sensible default.
- Reward the behavior you actually want. Pay for a resolved problem rather than unlimited hours, where quality can be assessed. Seek incentive compatibility: a deal where doing well for yourself also serves the other party’s goal.
- Make important actions easier to check. Require documented maintenance, inspect completed work, or audit a sample of claims. Target choices that meaningfully affect losses rather than adding paperwork everywhere.
- Separate essential protection from avoidable spending. A single charge on every service is blunt. Consider lower charges for high-value care or safeguards against unnecessary work, while recognizing that classification can be difficult.
- Count the costs of the remedy. Monitoring takes money. Incentives can encourage corner-cutting. Reduced coverage shifts risk back to people who may be poorly equipped to bear it. Compare the whole arrangement, not just the payer’s bill.
07When it isn’t moral hazard
Adverse selection concerns who enters an arrangement: people expecting expensive treatment may be more likely to buy generous insurance. Moral hazard concerns how incentives change behavior under that arrangement. Both can operate together, but they require different evidence and remedies.
Risk compensation is narrower: people may take more risks when they feel safer. Moral hazard also includes spending and effort responses when people’s appetite for danger stays the same or falls.
More use can be beneficial. Insurance may let someone obtain treatment they previously couldn’t afford. That price response can still count as moral hazard in economic models, while improving their welfare.
Protection has benefits precisely because people are often risk-averse. Eliminating moral hazard by eliminating insurance would also eliminate valuable protection. The practical goal is a workable balance that preserves protection while allowing some behavioral response.
08Roots
In 1963, Kenneth Arrow set out to explain the differences between medical care and an ordinary market. A sick patient cannot inspect a treatment the way a shopper inspects a pair of shoes. The patient faces an uncertain outcome under the care of a doctor who knows more, while insurance changes what the patient pays. Arrow brought moral hazard into this account of medical care.
The label was already old. Nineteenth-century insurers distinguished hazards arising from the property itself from hazards arising from its owner. A fire underwriter could inspect a warehouse’s construction more readily than the owner’s honesty or care. The term carried judgments about character and encompassed worries about fraud as well as carelessness.
Mark Pauly’s 1968 comment on Arrow sharpened the price argument: when insurance lowers the patient’s cost of care, increased use is an ordinary economic response that can occur even when patients act responsibly. The idea traveled from insurance into banking, employment, and contracts. The old, character-laden name survived as economists increasingly used it to describe incentives.
09How solid is this?
Randomized health-insurance studies establish that cost sharing changes use, and the incentive mechanism is well developed in economics. Its size varies across settings; increased use or risk alone does not establish waste or net social harm.
10Connections
- Often confused with Adverse Selection, Risk Compensation
- Countered bySkin in the Game, Incentive Compatibility
- Can follow from Asymmetric Information
- Part of Principal-Agent Problem
- Includes Cobra Effect
- See alsoRisk Aversion
11Origin and sources
The term originated in nineteenth-century insurance practice. Kenneth Arrow (1963) and Mark Pauly (1968) helped establish its modern economic treatment through the study of medical insurance.
- [1]Arrow, K. J. (1963). Uncertainty and the Welfare Economics of Medical Care. The American Economic Review, 53(5), 941–973.
- [2]Pauly, M. V. (1968). The Economics of Moral Hazard: Comment. The American Economic Review, 58(3), 531–537.
- [3]Einav, L., & Finkelstein, A. (2018). Moral Hazard in Health Insurance: What We Know and How We Know It. Journal of the European Economic Association, 16(4), 957–982.
- [4]Baker, T. (1996). On the Genealogy of Moral Hazard. Texas Law Review, 75(2), 237–292.
Suggest an edit· Updated 2026-10-02