What are moral hazard and adverse selection?
Two distinct problems created by information being unequally held between insurer and insured — and almost every feature of how insurance is designed and priced is a response to one or the other.
Adverse selection — the problem before the policy is sold. People who know they are high risk are more likely to buy insurance, and to buy more of it, than people who know they are low risk.
Why this is destabilising. If an insurer charges an average price, low-risk people find it poor value and decline, leaving a higher-risk pool, which requires a higher price, which drives away the next tier of lower-risk buyers. Taken far enough, the market unravels — this is the classic demonstration, and it explains why voluntary insurance markets for some risks do not exist at all.
How insurers respond: underwriting questions; risk-based pricing; medical evidence; exclusions for pre-existing conditions; waiting periods; and compulsory or default participation, which is why some insurance is mandatory and why group schemes covering everyone in an organisation work where individual policies would not.
Moral hazard — the problem after the policy is sold. Being insured changes behaviour, because the consequences of a loss are borne by someone else.
Two forms: taking more risk — leaving a bicycle unlocked, driving less carefully — and claiming more readily, or inflating claims.
How insurers respond: excesses, so the insured bears part of every loss; no claims discounts, so claiming has a cost; policy conditions requiring reasonable care and specified security; premium increases after claims; limits and exclusions; and investigation of claims.
Why it is not a moral judgement. "Moral hazard" sounds accusatory and is a technical term about incentives, not character. The behaviour change it describes is frequently unconscious and entirely ordinary.
The tension both create. Every response to adverse selection — more questions, finer risk pricing — makes insurance more expensive or unavailable for those most likely to need it, which is the central social tension in insurance. Genetic testing in life insurance is the sharpest current example, and is subject to a moratorium and code precisely because perfect information would destroy the pooling that makes insurance valuable.
Pooling requires some ignorance to function at all.
General information, not advice.