Irrational Economist_ Making Decisions in a Dangerous World - Erwann Michel-Kerjan [69]
My own personal opinion is that practical public policy will tend to ignore this kind of welfare economics-based argument. If people have preferences differing from those of the benchmark model, the most “democratic” reaction is to say “So be it” and to let them waste their money in the way they choose.
However, it is not always possible to take such a hands-off attitude; it’s one thing for preferences to be merely eccentric, quite another for them to be impoverishing. But on top of this, in a society where government caters to what voters want rather than to what welfare economists want, politicians interested in maximizing their tenures or political fortunes may well find it desirable to go along with these mistaken preferences, whether in pursuit of first dollar coverage, dread disease insurance, or even poor investment vehicles disguised as whole life insurance.
Usually it is the extended EU model, rather than the behavioral (peace of mind) explanation, that yields refutable hypotheses. Indeed, the circumstantial evidence supporting the behavioral explanation often amounts to a rejection of the EU hypothesis, and then one turns to the behavioral explanation as the only argument left standing. For example, in principle one could see whether deductible aversion is in fact stronger in cases involving uninsurable (but not unmeasurable) additional loss. Do people who choose low deductibles experience larger uninsured changes in their wealth (in multiple ways) than people who choose high deductibles? The answer to this is not yet known, but could be. If it turns out to be “no,” the behavioral explanation would gain support.
CONSUMER WARRANTIES
Another example of a phenomenon that seems better explained by a behavioral theory rather than by EU theory is the purchase of warranties on consumer durables. Here again, the cost often exceeds the expected value of the benefit by a substantial amount, and yet many people still buy. To be sure, the warranty’s cost is usually small relative to the price of the product (or even relative to the price of a visit from the repairman), but the rational consumer looks at the price relative to the expected benefit, taking into account the probability of the loss as well as the amount of the loss.The proportion of people engaging in this behavior is a minority (in contrast to the majority who choose low deductibles), but a substantial minority, about a third of buyers. One other important factoid: The demand for warranties, even if its existence is irrational, is (rationally) responsive to their prices. (I suppose the same might have been said about the demand for pet rocks.)
Buy why does this behavior occur? The obvious explanation is that buyers of warranties overestimate the probability of needing repairs (especially when experiencing the hard sell after an already costly purchase). Another possible explanation is that, having made the effort to shop for and choose an item, at least some people may attach special value or affection to it—an argument that has been made in a more general framing (Kunreuther and Hsee, 2000), but usually with respect to “old favorite” and difficult-to-replace items, not brand-new merchandise still in the box. Warranties also protect buyers against higher prices for replacements, and may be priced favorably by a manufacturer whose marginal cost of replacement is much below the selling price—a real possibility for many relatively low-priced consumer electronics. Then, too, warranties give buyers peace of mind and freedom from regret.
An interesting question is whether the willingness of some buyers to buy warranties that would not be a good deal even if they were not overpriced results in a lower price for the product itself; could the product be a loss leader (as are razors), with the money to be made on warranties and/or repairs? Perhaps, but this