In conversation with

Exordium

The Strangeness of the Discipline

A Good Scientific Conversation

The Strangeness of the Science

Not that economists care about all this abuse and miscomprehension. "There is something wrong here" does not resonate in academic hallways and at international conferences. For the most part, economists seem content with and even conceited about their discipline since they consider it superior to any other social science. They brush off criticism easily. When students and other outsiders complain that the theory of rational choice and the abstract models are unrealistic, economists typically will refer to the scientific character of their discipline. "Economics is a science and like any other science it proceeds by means of unrealistic assumptions and abstract models.". And: "scientists do not speak in comprehensible terms about the real world either." "This is a science, to be sure; if you seek our advice you will have to take our scientific analysis for granted unless you want to immerse yourself in our scientific inquiry." That should shut the critic up, shouldn't it?

But if economics is a science, it must be a rather strange science. Real scientists subject theories to empirical tests, reject those that fail and continue with the ones that are successful. Economists try to do the same, running regressions of theories that involve enormous sets of data. But while they do produce empirical results, it is impossible to pinpoint the series of empirical tests that prove a theory wrong and cause its rejection. Empirical evidence may favor one theory over another, but if an economist believes in her theory, she can — perhaps even innocently — work the data for evidence that suits her model better. No empirical evidence is decisive. There are no parallels in economics to the Morley experiments that proved Einstein's theory of relativity. If empirical research fails to falsify its theories, how scientific is economics?

The empirical failure of economics as a science goes further. Despite a vast amount of research, economists cannot provide a set of empirical results that is interchangeable, like physicists can. Economists cannot talk about a standard for the price elasticity of the demand for electricity. For all its importance and amount of time spent on the empirical testing of it, economists cannot state the impact on investment of a one percent change in the interest rate. As a textbook author, I find that I can demonstrate nothing more important about empirical research in economics than economists do a great deal of it. Where, then, is all this effort in empirical research going?

The hallmark of "real" science is a reliable, reproducible, and predictable result, yet economists fail famously to predict what will happen in the real economy. They miss the onset of recessions and are often surprised by strong growth just after they predicted a slowdown. Economists are bad in predicting the future course of the dollar, or the interest rate. Don't ask them about the stock market because they wouldn't know. As a matter of fact, serious predicting has become a low priority. Some research institutes will still generate predictions for public consumption — usually on the basis of large econometric models — but the work is not taken seriously in academic circles. After so often proving to be obsolete and superfluous, the scientific heyday of large models, on which my youth hero Jan Tinbergen had vested all his hope, is over. Again strange, predicting the outside world is not a skill that graduate school teaches, yet it is what the outside world expects.