Insurance is not available for every misfortune. Cover depends on conditions that some risks simply do not satisfy, and where they fail the product cannot exist at any price.
The loss must be uncertain
Insurance transfers uncertainty. Where an event is effectively certain, there is nothing to transfer and the premium would have to approach the loss itself plus expenses.
Wear, deterioration and gradual damage fall into this category, which is why policies exclude them. They are maintenance costs rather than contingencies.
Pre-existing conditions raise the same problem in a different form, since the event is already in progress when cover is sought.
Losses must be largely independent
Pooling relies on claims arriving separately so that most policyholders fund the few who suffer. Correlated events break that arithmetic by producing claims simultaneously.
Systemic financial losses, pandemics and widespread natural catastrophes can affect an entire portfolio at once, which is beyond what accumulated premium can meet.
Where these are covered at all, it is usually through reinsurance, government backstops or explicit limits on aggregate exposure.
The loss must be measurable
A claim requires a determinable amount. Where the loss is subjective or extremely difficult to quantify, settlement becomes contentious and pricing becomes guesswork.
Some policies address this by paying a fixed sum on a defined trigger rather than assessing actual loss, which removes the measurement problem at the cost of precision.
That structure only works where the trigger itself is objectively observable and difficult to influence.
The insured must not control the outcome
Cover that pays out on an event the policyholder can bring about creates an incentive problem the insurer cannot price around. Deliberate acts are excluded universally.
Less obviously, cover can change behaviour in ways that raise the underlying risk, which is why deductibles and limits exist alongside exclusions.
Selection is the mirror image: those most likely to claim are most likely to buy, which pushes premiums up and drives lower risks out unless classification counters it.
Where the market withdraws instead
Availability can disappear without the risk becoming formally uninsurable. If expected losses rise faster than premiums can, insurers reduce exposure in an area or line.
Property cover in high-hazard locations shows this pattern, with terms tightening and capacity narrowing before withdrawal.
The outcome resembles uninsurability from the buyer's side, but the cause is pricing and capital allocation rather than a failure of the underlying conditions.