Irrational Economist_ Making Decisions in a Dangerous World - Erwann Michel-Kerjan [71]
A more plausible explanation from a behavioral perspective is that they do not have enough information to see the value of insurance. A study of the nonpoor uninsured in California provides some confirming, if circumstantial, evidence: Nonpoor people without health insurance are also likely to lack other kinds of insurance (life, homeowners, collision), to be maxed out on their credit cards, and to believe that individual insurance is much more expensive than it really is. These are people who are just not into risk protection of any sort. But there is an economic explanation as well. Group insurance tends to charge explicit premiums that are independent of risk (though with much lower than expected benefits) and thus seems to discriminate against lower risks, sometimes creating a situation where lower-risk employed people are less likely to take coverage than the higher-risk employed. (It probably helps the risk pool that a person must at least be able to work to get group insurance, such that the possibility of severe adverse selection is reduced.) A low-risk person will face the same explicit premium as a high-risk person working in the same firm, but the options available in firms employing mostly lower-risk people and the existence of a lower wage offset for coverage cause premiums to be somewhat reflective of risk.
Here, the best outcome is probably a compromise. Insurance prices facing the uninsured are not especially attractive, and the uninsured tend to underestimate the value of insurance. Behavioral factors and rational factors reinforce each other.
The role of public policy in this context has recently been much debated. Public subsidies for moderate-income uninsured people, which exceed the subsidies they get for group insurance, are a part of every health reform plan. And most plans purport to be able to offer coverage at lower administrative cost through exchanges or pools of some kind. Intererestingly, behavioral issues—motivation, salience, and framing—have not been a major part of the debate.
One approach that sidesteps these issues in practice but makes them more important in theory is the use of mandates for coverage (for children or adults). The strongest argument for mandates has to do with the spillover costs that the uninsured impose on others—costs that recent research shows are much larger than those related to hospital-level charity care, but that extend to lower quality for insured people in the community, because larger fractions of the uninsured make high quality less profitable (whether they use such care or not) (Pagán and Pauly, 2006).
INSURANCE AND A TRANQUIL LIFE
A theme common to the cases discussed in the preceding three sections is that people have other objectives than getting their insurance purchases just right. Howard Kunreuther and I formalized this idea in a study that postulated a decision-making cost to investigating insurance purchasing—collecting information on the best premiums, the best coverage, and the true loss probabilities (Kunreuther and Pauly, 2000). We found that where a risk really matters—fire insurance for the home, collision insurance, an extended warranty for a new car, insurance for dependents against a parent’s premature death—insurance does commonly exist. That is, where expected utility might really be impacted by a loss, people usually care enough to start looking for insurance and eventually buying some coverage, though perhaps not the literally ideal amount. How long they persist in searching for coverage depends on how much a good price matters to them; for example, we found that low-risk young people paid premiums that were less than optimally reflective of their expected expenses for health insurance—a phenomenon consistent with the cost-of-search hypothesis but also with others.
One important consideration that has both rational and behavioral implications is how people