Kenneth Arrow, Moral Hazard and How Extremely Intelligent People Can Be Led Astray
By Matthew McCaffrey
Nobel Laureate Kenneth Arrow has passed away at the age of 95. There are few people who can lay claim to a more profound influence on 20th-century economics. In fact, Arrow was the mainstream economist par excellence, contributing to many fields over the course of an almost seventy-year career. He will likely be remembered mainly though as a pioneer of the modern mainstream approach to general equilibrium theory and welfare economics (together with fellow Nobelist Gérard Debreu).
However, despite (not because of) his enormous influence, Austrians tend to be critical of Arrow’s most famous achievements, particularly his role in cementing general equilibrium as the key framework used to do economic theory. There are many criticisms of equilibrium theorizing in its various forms, but one major problem is its lack of realism: general equilibrium contains little room for the dynamic, uncertain, and above all, entrepreneurial market process (for a related debate, see Peter Klein’s discussion here).
Arrow’s work is a good example of how extremely intelligent people can be led astray by ideas that are inappropriate to the task at hand. In Arrow’s case, the trouble stems from his focus on deriving rigorous mathematical proofs for abstract equilibrium models rather than on the practical problems of real prices and economies. A good example of this tendency is Arrow’s work on moral hazard, which shows how ignoring entrepreneurship leads to unrealistic theories and mistaken policy conclusions.
Moral Hazard, Insurance, and Market Failure
Once an obscure insurance concept, the financial crisis made moral hazard almost a household term. It’s also a concept Austrians have been interested in for some time (see here and here for some discussion). For Austrians, moral hazard is mainly a branch of the theory of interventionism. It deals with the implications of choice when actors don’t bear the costs of their actions. This occurs, for example, when actors can shift the costs of their decisions to others without these others’ immediate consent. Lower costs in turn make decisions that were previously dangerous or costly more attractive. The result is that people benefit from making choices they normally wouldn’t make, and other people bear the consequences. Markets tend to minimize these kinds of situations, whereas intervention institutionalizes them. Bailouts are a classic example of moral hazard: firms are more likely to engage in risky behavior if they’re convinced they won’t be allowed to fail when they make bad decisions.
Unfortunately, moral hazard is one area where Arrow’s contribution represented a step back for economics. In 1963 he published a paper titled, “Uncertainty and the Welfare Economics of Medical Care,” which is now considered a classic as well as the starting point of the modern literature on moral hazard. In this article, Arrow argues that markets underprovide valuable forms of medical care and insurance coverage. If this is the case—that is, if the quantity or quality of medical care are too low in a free market—then the existence of moral hazard may justify government intervention to increase the quantity or quality of medical care. Importantly, Arrow’s theory marked a key move toward explaining moral hazard as a kind of market failure, especially one produced by asymmetric information.
In the case of insurance, Arrow argued that moral hazard could prevent insurance companies from creating many beneficial forms of coverage. The idea is that doctors act as the agents of insurance companies. They cannot be perfectly monitored, and therefore the cost of their poor decisions is shifted to the insurers. One result is that the wider the coverage options that companies offer, the more likely doctors will be to over-supply treatments, and the more likely patients will be to demand them, either by taking greater risks or simply over-medicating. Insurers bear the burden of this behavior, but due to incomplete information they cannot price their services accurately, which encourages them to reduce their coverage to suboptimal levels. To rectify this situation, companies can be taxed and subsidized in order to stimulate the desired forms of production, or government itself may step in as an insurer.
The fatal flaw of Arrow’s theory is that it compares the real world to an abstract and idealized competitive model. Predictably, the real world fails to live up to “optimal” conditions of this model, which are based on unrealistic assumptions such as that all goods and services relevant to costs and benefits in the medical care industry can be priced (p. 945). For Arrow, the fact that many are not—for example, there are few or no markets for bearing many types of risk—implies real-world markets are flawed, in that they cannot create the preconditions for the emergence of a Pareto-optimal equilibrium.
Arrow’s logic made a serious impression on the economics profession. However, he overlooked some key developments in economics that made his own theory largely irrelevant. In particular, Frank Knight brought the concept of moral hazard into economic theory almost four decades before Arrow’s article appeared, explaining it as an entrepreneurial problem. Knight’s view, which was close to Mises’s, is that the function of entrepreneurship is to bear uncertainty. In practice, this means that entrepreneurs constantly make judgments about how best to deal with incomplete information, especially about the behavior of their customers and the people working in their organizations. Using economic calculation, entrepreneurs try to work out the best uses to make of society’s scarce resources. Rather than using a neatly-defined model as a benchmark, entrepreneurs pit their abilities against the data of the real world.
In other words, there’s no predetermined optimal quantity or quality of medical care (or any other good or service), only the one entrepreneurs produce through their constant efforts to satisfy consumers in the face of scarcity and uncertainty. This optimum can’t be determined in the abstract: it’s created from moment to moment in the market through the interactions of millions of consumers and producers.
By focusing on a competitive equilibrium, Arrow also overlooks a vital institutional problem: governments create moral hazard by socializing the costs of dangerous behaviors. Unlike other market actors, policymakers are in a unique position to systematically redistribute costs and benefits by force. They thereby increase the demand for regulatory capture and rent-seeking. Rather than impartial referees that reduce moral hazard, governments are its most common cause. Arrow definitely recognized some problems of government intervention and production, and he likewise understood that markets and civil society generate protections for consumers. Yet he never made the leap to understanding the power of entrepreneurship to eliminate moral hazard, and the power of intervention to create it.
The Morals of Moral Hazard
However, Arrow did make a positive contribution by helping to clarify the moral aspect of moral hazard. Following the publication of his paper, other economists criticized him for implicitly smuggling value judgments into his theory. In this view, Arrow’s theory of moral hazard implies a negative and unnecessary judgment about the character of consumers who respond to moral hazard, i.e. that people who act in risky or costly ways when they can shift the costs to others are immoral. This upset some critics, who claimed that there is no moral dimension to moral hazard. For instance, Mark Pauly argued that “the response of seeking more medical care with insurance than in its absence is a result not of moral perfidy, but of rational economic behavior” (1968, p. 535). Arrow’s reply is clever and insightful:
We may agree certainly that the seeking of more medical care with insurance is a rational action on the part of the individuals if no further constraints are imposed. It does not follow that no constraints ought to be imposed or indeed that in certain contexts individuals should not impose constraints on themselves. Mr. Pauly’s wording suggests that ‘rational economic behavior’ and ‘moral perfidy’ are mutually exclusive categories. No doubt Judas Iscariot turned a tidy profit from one of his transactions, but the usual judgment of his behavior is not necessarily wrong. (1968, p. 538)
We can’t neatly seal off economic behavior from all other kinds of action, or excuse it simply on the grounds of a narrowly-defined rationality. Although “pure” theory is value-free, real-world action never happens in a moral vacuum. Possible moral elements implications of moral hazard should not be dismissed out of hand.
Arrow’s reputation will continue to live on in economics. Certainly, for anyone interested in understanding how mainstream economics came to be the way it is, his work is essential reading. However, Austrians will likely continue to regard Arrow’s research as a challenge to the Mengerian tradition rather than its guiding light.