Yves here. While there have been some good posts expressing consternation at the recently announced winners of the so-called Nobel Prize in Economics, this one is particularly useful in explaining how economists come to treat obviously ridiculous theories as gospel truth.
Readers may be familiar with one of the major mechanisms: the fondness of economists for theoretical work that presents itself as rigorous due to the use of proof-like expositions (but people who have examined large samples of theoretical papers find even they fail to meet their professed standards, since the critical parts of the argument are more often than not in the narrative parts, and the mathed-up sections are trivial). This post gives a good layperson exposition of a second major failing of economics: that data sets and regressions can be tweaked enough that one can eventually find a relationship that looks meaningful. But those relationships are typically well short of robust; they often fail if key variables (time periods, definitions, data sources) change. As data scientist Cathy O’Neil noted apropos the Reinhart/Rogoff debate over debt levels and growth:
I’d have no problem with economists if they behaved like the people in the following completely made-up story based on the infamous Reinhart-Rogoff paper with the infamous excel mistake.
Two guys tried to figure what public policy causes GDP growth by using historical data. They collected their data and did some analysis, and they later released both the spreadsheet and the data by posting them on their Harvard webpages. They also ran the numbers a few times with slightly different countries and slightly different weighting schemes and explained in their write-up that got different answers depending on the initial conditions, so therefore they couldn’t conclude much at all, because the error bars are just so big. Oh well.
By James R. Crotty, Professor Emeritus of Economics and Sheridan Scholar at University of Massachusetts. Cross posted from Triple Crisis
Eugene Fama just received a Nobel Prize for his contributions to the theory of “efficient financial markets,” the dominant theory in financial economics that asserts that markets work ideally if not constrained by government regulation. The fact that economic “science” teaches that unregulated financial markets work effectively helped financial institutions and the rich accomplish their goal of radical financial market deregulation in the 1980s and 1990s. Deregulation, in turn, not only contributed to the rising inequality of the era, it helped cause the global financial market crisis that began in 2007 and the deep recession and austerity fiscal policies that accompanied it.
The theory of efficient financial markets requires the union of two ideas: the “efficient market hypothesis” (or EMH) and optimal (security) pricing theory (OPT). Both the EMH and OPT are built on crudely unrealistic assumptions that would lead anyone not indoctrinated in a mainstream PhD program to conclude that efficient financial market theory is a fairly-tale rather than serious social science.
The EMH is simply an assumption or assertion with no supporting evidence that all information relevant to the correct pricing of securities is known by all market participants. For long-term assets such as stocks and bonds, the relevant information is the cash flows associated with each security in every future time period. Yet it is logically impossible for anyone to know this information because the future is not yet determined in the present; the future is uncertain. Nevertheless, defenders of efficiency adopted the “rational expectations” hypothesis, perhaps the most ludicrous assumption in the history of social science, which asserts that all investors know the correct probability distributions of all future security cash flows and believe that they will not change over time.
The assumed complete and correct data about the future is then plugged into one of the basic mainstream models of optimal security pricing, such as the capital asset pricing model (CAPM), which specifies agents’ preferences concerning the risk and return associated with every possible portfolio of securities. The combination of EMH and a theory of optimal pricing determine security prices that are efficient in that every investor has selected the risk-return profile in a portfolio that maximizes her welfare, and financial resources are made available to those who can make the most productive use of them. Market prices are assumed to be in equilibrium at all times, even though the data show that market prices are much more volatile than would be compatible with the assumption of perpetual equilibrium.
The capital asset pricing model itself embodies a large number of grossly unrealistic assumptions in addition to the assumed knowledge of the future embedded in the EMH. For example, it assumes that every investor holds the same portfolio (those who want more risk borrow money to build a larger version of this portfolio), no one trades securities, and no one ever defaults on debt.
One might think that the whole financial market-efficiency project should have been rejected out of hand because it is founded on a large set of unrealistic assumptions about how financial markets work. Yet not only is it still the dominant theory of financial markets, Nobel Prizes have been awarded to its originators.
Why would an academic profession sanction the use of theories based on such unrealistic assumptions? The answer given by proponents of efficient financial markets theory is that the economics profession relies on the theory of “positivism” associated with Milton Friedman as its guide to the acceptance and rejection of theoretical propositions. Friedman’s positivism states that the realism of assumptions does not matter: it has no relation whatever to the acceptability of a theory or its derived hypotheses. As Friedman put it, “[T]ruly important and significant hypotheses will be found to have assumptions that are wildly inaccurate descriptive representations of reality.” The only acceptable test of a theory “is comparison of its predictions with experience.”
There are at least three serious problems with this method. First, if patently false assumptions are adopted, as in efficient financial market theory, and impeccable logic is used to deduce hypotheses from them, they cannot—as a matter of logic—be accurate reflections of reality. Fairy-tale assumptions can only generate fairy-tale hypotheses.
Second, econometric tests can at best provide suggestive, not conclusive evidence in support of the empirical validity of predictions generated by economic theories. With today’s computing power, it is possible to run literally millions of regressions to test a theoretical proposition. Such regressions may use different data sources, time periods, empirical measures of theoretical variables, functional forms, lag structures, and so forth. For example, investor expectations of future cash flows from all available securities are a central determinant of efficient equilibrium security pricing, yet there are numerous ways to choose empirical measures of expectations. And the theory itself does not tell us what the appropriate choice among this vast menu of possible alternatives measures is. As a result, virtually any hypothesis can be shown to be statistically significant if enough different regressions are run. This is why both sides of every important debate in economics can provide econometric evidence in support of their positions. And it is why economists should not rely exclusively on econometric hypothesis-testing in assessing alternative theories as positivism demands. The realism of assumption sets is crucial to this task, as are historical and institutional analysis, surveys, and experimental studies.
Third, when positivist economists insist that econometric “prediction” is the sole judge of the acceptability of a theory, they put the entire burden of proof on econometric tests. But when the preponderance of such tests turns out to be inconsistent with their favorite theory, they never reject the theory, as their methodology says they must. Rather, they move on to additional econometric tests on alternative specifications in a potentially endless process of data mining. In a widely discussed survey of empirical tests of hypotheses derived from the CAPM in 2004, Eugene Fama and a coauthor arrived at a striking conclusion: “despite its seductive simplicity, the CAPM’s empirical problems probably invalidates its use in applications.” The tenets of positivism require that the CAPM should be rejected. However, financial economists kept mining the data in an endless effort to find econometric results that fit the theory. Meanwhile, CAPM sustained its canonical status and efficient market theory remained unscarred in spite of its lack of empirical support.
Why would an academic profession adopt a methodology such as positivism that supports theories that are based on unrealistic assumptions? After all, there is an obvious alternative—begin with a realistic assumption set and use it to derive realistic hypotheses about the behavior of financial markets. This is the method used by Keynes and Minsky to show that financial markets have no efficiency properties and are properly thought of as gambling casinos. The answer is that the economics profession is committed ideologically to a defense of the proposition that financial markets are efficient, yet it is impossible to derive this proposition from a realistic assumption set. Thus, the profession had no choice but to adopt a positivist methodology that sanctioned the use of even absurdly unrealistic assumptions in theory construction. Since realistic assumptions lead to theories that show the strengths, but also the myriad dangers and failures of unregulated capitalism revealed in the historical record, they had to be replaced by the large number of absurd assumptions required to sustain support for economists’ inherent belief that unregulated or lightly regulated markets create the best of all possible worlds, maximizing both economic efficiency and individual liberty. Positivism is the magic that makes it possible to construct a “scientific” defense of the proposition that free-market capitalism has no serious flaws and dangers.
The objective of the ideological project of the economics profession in the current era is to provide a theoretical foundation for unregulated financial markets and unregulated capitalism. The fact that the project has succeeded in the face of logic and history is admittedly a fantastic conjurers’ trick, but it is ridiculous to award Nobel Prizes to the conjurers. We should not give prizes to people for the creation and propagation of an ideologically-based theory that strengthened the drive for the radical financial deregulation and thus helped create a global depression.