TCS Daily

Sweetwater vs. Saltwater

By Arnold Kling - November 5, 2002 12:00 AM

Economists agree on many things. We are all for free trade. We are all more persuaded by Bjorn Lomborg than by Edward O. Wilson. We all believe that government support for scientific research helps the economy. We all support Lawrence Lessig in his attempt to overturn copyright extension (the Sonny Bono Act).

However, the economics profession divides on a number of issues. This year's Nobel Prize, awarded to Vernon Smith and Daniel Kahneman, may prove to be the most controversial in its 34-year history. In general, I would expect that Saltwater economists (from Berkeley, MIT, and Harvard, near the oceans) would approve, while Sweetwater economists (from Chicago, Minnesota, and Rochester, near the Great Lakes) would not.

This year's Nobel laureates are known for evaluating the validity of two sacred assumptions in economics: the assumption that individuals optimize and the assumption that real-world markets lead to competitive outcomes. Kahneman has found systematic exceptions to the assumption that individuals optimize. Smith uses human experimental subjects to simulate market behavior, and on occasion his simulations have found that different market structures lead to different outcomes.

Can We Live Without Our Assumptions?

The Sweetwater and Saltwater schools differ on the significance of the underlying postulates. Here are two representative points of view.

"Economics is that way of understanding behavior that starts from the assumption that individuals have objectives and tend to choose the correct way to achieve them."
--David Friedman, Hidden Order: The Economics of Everyday Life, p. 3

"The only sane answer to the questions 'do markets work well?' and 'do people act in their own interest?' is 'sometimes.' Smith and Kahneman have done a huge amount of productive work in helping us to understand that 'sometimes.'
-- Brad DeLong

David Friedman, the son of Milton Friedman (Nobel 1976), represents the Sweetwater school. The standard definition of economics is "the study of the allocation of scarce resources for competing ends," which does not rule out irrational behavior. However, for Chicago-trained professors, economics is the study of rational behavior. In their view, work such as Kahneman's, which looks at irrational behavior, is outside of economics by definition. Although Milton Friedman is the most well-known exponent of the Sweetwater school, within the profession Gary Becker (Nobel, 1992) is considered the economist most closely identified with the doctrine that economics is the study of rational behavior.

Milton Friedman's defense of the rationality assumption is pragmatic. He does not ask us to believe that people do behave rationally. Instead, he asks us to evaluate the predictions that he makes when he treats people as if they behave rationally. In Friedman's famous metaphor, he says that we know that billiard players do not use the laws of physics to line up their shots. However, we can best predict how they will line up their shots by treating them as if they were using those laws.

Brad DeLong, who is also the son of an economist (James V. DeLong), represents the Saltwater school. Saltwater economists believe that examples of irrational behavior and imperfect markets are interesting and important.

In principle, a Sweetwater economist might agree with DeLong that people are rational only sometimes and that markets are perfectly competitive only sometimes. However, the Sweetwater economists do not view this as a reason to pursue economics differently.

The Sweetwater school views the economist as a prediction-making machine. As long as the economist's predictions are nontrivial and accurate, the underlying assumptions themselves do not need to be examined. By the same token, Sweetwater economists would argue that in order to show that irrational behavior and imperfect markets matter, you must use your theories to make nontrivial and accurate predictions. It is not sufficient to demonstrate in a psychological experiment that people miscalculate. It is not sufficient to demonstrate in a simulated market that strange outcomes occur. The relevance of such findings depends on one's ability to use them to make valuable predictions about the behavior of prices and trade patterns in the real world.

By these standards, I believe that this year's Nobel laureates fall short. The "experimental economics" that follows Vernon Smith has generated very few predictions about real-world markets that differ from the predictions of standard mathematical models. To the extent that there have been unique predictions, to my knowledge they have not been demonstrated to be valid in any real-world setting.

Similarly, the "behavioral economics" that follows Daniel Kahneman has not developed much empirical traction. There are some hints of interesting results, particularly concerning how much income individuals choose to save and how they manage their portfolios, but giving a Nobel Prize to behavioral economics at this stage is like putting a promising rookie in the Hall of Fame.


In the 1970's, the great divide between Sweetwater and Saltwater economists concerned macroeconomics. Saltwater economists, such as James Tobin (Nobel, 1981) battled Sweetwater economists, including Milton Friedman and Robert E. Lucas, Jr. (Nobel, 1995).

The conflict began over the importance of monetary policy. Saltwater economist Robert Solow (Nobel, 1987) once quipped that "Everything reminds Milton Friedman of the money supply. Everything reminds me of sex, but at least I keep it out of my papers."

Eventually, however, the war spread to the very issue of whether unemployment in recessions is involuntary. In the latter part of the 1970's, Sweetwater economists began to respond to the natural gravitational pull of the assumptions of optimization and well-functioning markets, which tend to rule out the possibility of involuntary unemployment. Saltwater economists were outraged - Franco Modigliani (Nobel, 1985) once protested "Was the Great Depression nothing but an outbreak of laziness?" Sweetwater economists, notably Edward Prescott, believe that unemployment indeed can be explained by changes in the incentive to work, rather than the Keynesian mechanism of effective demand failure.

In my view, if we use the "prediction machine" concept of economics, then Keynesian Saltwater economics works better. For example, using standard Keynesian thinking, I was able to predict in January of 2000, that the collapse of the Internet Bubble (which was still inflating at the time) would lead to an economic slump.

The "prediction machines" of Keynesian macroeconomists have been far from perfect. However, Herbert Stein once wrote that although economists do not know very much, noneconomists know even less. Similarly, I would say that Saltwater economists do not know very much about macroeconomics. However, Sweetwater economists, who deny that there is such a thing as involuntary unemployment, know even less. For more on the Saltwater point of view, see this interview with Solow.

Policy Differences

Sweetwater and Saltwater economists tend to differ on policy issues. Saltwater economists see opportunities for government to fix private-sector outcomes. Sweetwater economists are less likely to view private sector outcomes as broken. Rather than assume that government seeks to improve social welfare, Sweetwater economists are more likely to adopt the cynical view of the Public Choice school, associated with James Buchanan (Nobel, 1986).

Here are a few examples of policy differences:

The Distribution of Income

Sweetwater economists tend to favor ensuring a minimum level of income. However, they would take no further steps to reduce income inequality.

In contrast, Saltwater economists believe that government must pay more attention to the distribution of income. What the money supply is to Milton Friedman and sex is to Robert Solow, the distribution of income is to Paul Krugman and other Saltwater disciples.


When it comes to monopoly, Saltwater economists tend to see government as the solution, while Sweetwater economists tend to see government as the problem.

Historically, governments granted monopolies as a revenue source. Most Sweetwater economists believe that to this day no monopoly can persist without government protection. Indeed, residues of the symbiotic relationship between monopoly and government revenue can still be seen today, for example, in the high tax rates on regulated telephone service.

In the Microsoft case, there are Saltwater economists who take the view that Microsoft is a monopoly and that government should take strenuous action. Sweetwater economists tend to believe that the market is capable of providing checks and balances against Microsoft. I myself cited Thomas Sowell, a disciple of Milton Friedman, in my skeptical survey of the case for government intervention in the market for operating systems.

Growth Policy

Saltwater economists tend to think in terms of opportunities for government to make positive contributions to economic growth. These might include spending on health and education, increasing taxes to reduce government deficits and interest rates, and intervening in situations where markets produce flawed outcomes.

Sweetwater economists tend to think that government's best contribution to growth would be to get out of the way. They favor deregulation, lower tax rates to reduce the deadweight loss from taxes, and letting the private sector sort out economic choices.

Staying in School

Few of the prominent exponents of the various schools are active on the Web. The leading economics weblog belongs to the younger DeLong, representing Saltwater. The Sweetwater contingent includes Megan McArdle and Zimran Ahmed. Economists with deep roots on the Web but unpronounced aquatic tendencies include John S. Irons and Hal R. Varian.



TCS Daily Archives