"Economic Analysis and Social Obligation"

Professor Kenneth J. Arrow, 1972 Nobel Laureate in Economic Sciences

A major question of government policy is the extent to which it should intervene in the workings of markets. There is a general theoretical proposition, which also conforms to empirical experience, that, under competitive conditions, market outcomes are efficient in a certain precise sense. It follows that the government should intervene when markets fail to operate properly, that is, when individuals benefit or impose costs on each other in ways that are not mediated through the market (e.g., pollution). An additional reason for government intervention, not as widely accepted, is to make the income distribution more equal than the market outcome. But in that case the redistribution should interfere with consumers' choice as little as possible, which means that the redistribution should take the form of money transfers (taxes and payments).

Economists typically argue, then, that externalities be recognized by taxes which are a kind of substitute for prices and redistribution by money payments.

In fact, though, there are all sorts of felt social obligations that are not encompassed by this simple proposition. First of all, there are the obligations of parents to children (the obligations of children to parents are fading), which have long been recognized in law. Second, there are the collective obligations of the present generation to the indefinite future, displayed historically by wars in defense of a country and by constitutions and other legislation and faced currently in concerns about exhaustible resources and climate change. Third, we find that, by now, most government expenditures even in the United States are for education, retirement, and health, goods which have very few externalities as currently understood.