Is Society a Great Big Insurance Company?

A great variety of reasons conspire to create a “safety first” culture. Many originate in one of the perverse features of modern society: that the risks, costs and benefits of a given course of action accrue to different persons. Since September 11, 2001, hundreds of millions of airline passengers have suffered delay and inconvenience and the airlines and airports were made to incur billions of dollars of extra expense to maintain pre-boarding “security”. The security checks may cause an infinitesimal reduction in the risk of a terrorist boarding the plane with explosives round his waist, and possibly an infinitesimal increase of the risk that he will place the bomb in the suitcase he consigns to the luggage compartment. Patting down millions of respectable matrons before letting them board their plane costs a great deal, causes irritation and no perceptible improvement in security. They cause neither irritation nor cost to the regulators who impose these checks at the airports. However, if they relaxed the controls, all the benefits would go to the airline industry and its hundreds of millions of nameless passengers, and none to the regulators. The passengers would not think of thanking them, nor would they have a face-to-face chance of doing so. Should, however, a suicide bomber blow up a plane after controls have been relaxed, the regulators would be lucky to escape being crucified by the media.

The same kind of mismatch between the interests of the risk-bearers and the regulators who “save” them from the risk prevails wherever and whenever collective choice exercised by some authority forcibly crowds out individual choices.

Often in recent history when a situation would inflict losses on some business interest if left to individual choices, collective choice steps in, indemnifies the losers and spreads the loss over all taxpayers. To stay with the airline industry, such a case occurred to it when in April 2010 a volcano blew clouds of ash into the European airspace and the regulators stopped air traffic for six days on the grounds that not a single passenger’s life must be exposed to avoidable risk. The airlines are estimated to have lost $1.12 billion and there is a strong likelihood that they will be indemnified from some European or national government purse.

A somewhat different but equally powerful mechanism operates in the notorious Common Agricultural Policy where the Community shoulders much of the losses that excessive production would otherwise inflict on dairy farmers, pig breeders or wine growers. What used to pass for an ordinary business risk that business was expected to bear, is gradually transformed into a social responsibility. The progressive transfer of responsibility for losses from individual businesses to the state suppresses much of the penalty for losses and hence removes much of the incentive to get rid of the causes of the losses.

Quantitatively the most important field where the “safety first at any cost” principle holds sway is health insurance, especially where the health insurer is directly or residually the government itself that is vulnerable to electoral threats. The service providers have a financial incentive to furnish more and costlier medical, hospital and laboratory tests and seek to avoid the risk, however minute, of a mishap and a ruinous lawsuit in case they did less than the very maximum service. Patients have no interest to check this tendency for in the majority of cases most or all of the cost is borne by the insurer and ultimately by society as a whole.

The very last word in the matter of “safety as the supreme guiding principle” is said by social contract theory that seeks to justify the individual’s obedient subjection to collective authority. Except in revolutionary situations, individuals obey choices imposed upon them by the dictator, the constitutional monarch, an oligarchy, or a democratically formed majority. The principle that such choices are made according to some collective choice rule (that may, but need not be a full-fledged constitution) is matched by the “principle of subjection” that obliges subjects to obey.

David Hume, usually wiser than any rival political philosopher, held that government is born not endogenously from social evolution, but exogenously from foreign conquest or “quarrels with other societies”.1 Once it is installed, individuals acquiesce in its exercise of power over them. In sharp contrast to the Humean view that sees passive acceptance of a tolerable modus vivendi, social contract theory sees people actually seeking a contractually defined command-obedience relation with authority and positive consent to collective choices made according to some implicit or explicit constitutional rule. They consent in full knowledge that collective choices will not be unanimous and some future ones may damage their own interests. On balance, however, they all expect to be winners. If they did not, they would not unanimously consent to the contract, which according to contractarian theory they are supposed to do.

Next to Rawls, the most thoroughgoing version of contractarianism is that put forward by Buchanan and Tullock in their Calculus of Consent2 and in much of James Buchanan’s subsequent work. The basic idea is that everybody, well-to-do and needy alike, will consent to a distribution of incomes which protects them from the risk of adverse changes in income: everyone will opt for a society that, like a great big insurance company, indemnifies losers from the “premiums” it collects from the gainers, the currently well-to-do. The well-to-do willingly accept this one-way redistribution from rich to poor because their preference for a safety-first society is stronger than their desire to remain well-to-do.

At first sight, this is an implausible conclusion drawn from extravagant or at least very strong assumptions. The chief one is that there is a “veil of uncertainty” that hides the future from the well-to-do (though, as some reflection must show, not from the needy). Untoward events may be lurking in the future that would reduce the income of the well-to-do to that of the needy. Suppose that the mean lifetime income is running at 80, with that of the well-to-do averaging 100 and that of the needy 70. There must be an untoward event that would cut the remaining lifetime income of the well-to-do to some low level, the event occurring with a certain probability at some future date which is also subject to a probability distribution. The probability of the adverse event, the reduction in income and its probable date (imminent, medium-term or distant) jointly give a result that would persuade the rational well-to-do to vote for a transfer of 20 of his present income to the needy. His remaining lifetime income would now be running at 80, and so would be that of the formerly needy; there would be equality at the mean level. Should the untoward event in fact occur, the formerly well-to-do would stop transferring part of his income to the formerly needy, but he would presumably still draw an income equal to the mean 80, benefiting as he would from the transfers of other well-to-do contractarians. As long as the mean income did not fall below 80, everybody would be assured to get it one way or another. Society, functioning as an insurance company, would have vanquished risk.

This scheme has the charm of a bedside tale for wide-eyed children. It spoils this effect by the claim that it is describing the presumable behaviour of the rational “risk-averse” utility maximiser. Two remarks undermine this claim. One is that the rational maximiser is not necessarily “risk-averse”. He does buy insurance, but he also takes risks. His behaviour suggests that he is risk-avoider for some ranges of his income and risk-taker for others ranges. The second and perhaps more important commonsense objection to the contractarian fairy tale is that the well-to-do have a better option than the one of willingly accepting redistributive taxation in order to benefit from it if some uncertain future event cuts down their income. Instead of deliberately reducing their income in favour of the needy, they can devote a corresponding part of it to accumulate capital. If the adverse event does occur, they can fall back on their savings. If it does not occur, or occurs only in the distant future, they can leave it to their heirs. Under credible probability distributions, they are better off this way than by deliberately contracting for a redistributive society.

Needless to say, we do have a redistributive society, essentially because he who wants to govern needs to buy popular consent to it. This is something that, with David Hume, we must probably acquiesce in. But I think we should protest against being told that what is happening happens because this is our preferred way of going for “safety first”.