Poverty Insurance Audit Juries

Government redistribution is usually called “social insurance”, and in fact it does seem to function mostly as insurance against future risks. Which raises the question: why can’t private insurance fill this role?

A number of standard objections can be dealt with relatively easily. Moral hazard is no worse with private insurance, and adverse selection can be cut via parents privately committing their newborns to future insurance arrangements. Correlated risks can be addressed via re-insurance and financialization. Poverty and signaling externalities can be addressed by subsidizing private poverty insurance.

In government welfare policy, a key tradeoff is between the simplicity and uniformity of something like a universal basic income or negative income tax, and the complexity and targeting of managed programs, wherein possibly-ideologically-captured government-agency specialists consider many details of each recipient’s situation to judge just how much of what sorts of help they currently need, and where there are many rules that limit recipient behaviors. (I’m reminded of this tradeoff reading Orwell’s Road to Wigan Pier, and watching Friedman’s Free to Choose.)

Not only is private insurance more likely to accommodate varying individual preferences re this tradeoff, but occurs to me that this tradeoff can be made easier to manage via a random jury auditing approach. Whenever someone requests help, create a prediction market on if that request would be approved by a jury, if it were to consider that request in detail say a year later. (Some class of market traders get access to case details.) Then if that market estimate is high enough, give the requested help. (Or have juries declare an appropriate level of help, have markets estimate that number, and then give that estimated number as help.)

If you fill these juries with random insurance customers similar to the applicant, who are are about to or recently did made their insurance quantity vs. premium tradeoff, you’d avoid ideological capture by those who handle specific cases. And by only rarely auditing choices via juries, you’d greatly lower the cost of allowing detailed case-specific judgements about the need for help.

Of course private insurance needn’t stop generous altruistic help, but it should be legit for such help to consider if someone chose not to help themselves when they could have done so. Ick feelings from seeing poor folks nearby can be addressed by exiling them to far away.