by (Originally published 2 February 2009)
In “The War of the Economists“, Will Wilkinson comments on the question of ‘economics’ and ‘science’, saying
There is a reason extremely smart economists are out there playing reputation games instead of trying to settle the matter by doing better science. The reason is that, on the questions that are provoking intramural trashtalk, there is no science.
Wilkinson is at least partly right, but not terribly helpful.
It seems to me that discussions like this — about the status of ‘economics’ as a “science” — often generate much more heat than light, at least in large part because people fail to recognize or acknowledge that ‘economics’ is not actually a discpline (as a professor of mine once said regarding ‘political science’). Rather, ‘economics’ (generally stated) is a subject matter, studied by (at least) three different disciplines, which can be broken down into 1) economic science; 2) economic theory; and 3) economic practice.
The first discipline is indeed a science, and there are indeed economists who practice it. This involves using science’s hypothetico-deductive methodology, examining evidence, forming hypotheses, and testing them against the evidence (to the extent possible for social sciences). The second is really philosophy: the question of “what is a good economic system?”, like “what is a good life?”, is not a purely factual one, but one based (at least in part) on what we think to be important. Of course, any good philosophy will be informed by facts and evidence (provided, perhaps, by economic science); if not, it becomes merely ideology. But economic theory is not the same thing as economic science. The third is a practical discipline, as practiced by the FRB, the Secretary of the Treasury, and those involved in finance. This is neither science nor philosophy, but is involved in the question of “what do we do now to avoid or emerge from a recession, or make a profit. Of course, economic practice should be informed by both theory and science, but it is also different from both of them, in that it demands practical action in the present, when neither theory nor science can give definitive answers.
Thus, the question of whether ‘economics’ (in general) is “a science” is unanswerable, because ‘economics’ (in general) both is and is not. The sub-field of economics comprising the discipline of ‘economic science’ is “a science” (though it may sometimes be practiced badly). ‘Economic theory’ and ‘economic practice’ are not “science” — but they aren’t intended to be.
But, more important to the point made by Wilkinson, “the questions that are provoking intramural trashtalk” are not questions of econonomic science, but questions of economic practice. It is a mistake to say that “there is no science” relating to such questions, because there is indeed economic science relating to them. The problem is that the science is inconclusive. But be that as it may, leadership (in business, war, or politics) involves making important decisions with incomplete information. Even lacking conclusive scientific answers to questions about how much economic stimulus is required (or even whether it is required!), and what the “best” stimulus would be, government must act in some way.
But the issue is still more complicated. For just as economic practice must rely on economic science in order to be most effective in achieving its goals, it must also rely on economic theory in forming its goals. Even if economic science were able to provide some kinds of conclusive answers, these would not be sufficient for economic practice, for they would be scientific answers. That is, they would be answers to the question: “if you do X, then Y will result”. But this does not tell you whether you want to achieve Y. Even if, for example, one could “prove” (in some sense) that letting destruction, unemployment, etc. run their course is “better” for enabling long-term economic growth, this would hold only if one accepted that “growth” is the fundamental goal — or at least more important than any other. And on this point there is no general agreement.
If there are fundamental disagreements about the purpose of economic activity (in economic theory), then it should not be surprising that there are deep disagreements about the actions that should be taken by societies or states in economic matters (economic practice). And on many of these matters, it does not appear that economic science will be helpful in resolving the dispute (at least for the foreseaable future). This does not make economic science useless. At least some of the time, economic science can tell us that, in order to achieve some goal X, the evidence suggests that Y will work better than Z.
But the answers will very often be suggestive rather than conclusive. As in many of the social sciences, conclusive experiments in economic science are hard to come by. Small-scale contolled experiments can be performed, but it is difficult to ascertain that they measure what is relevant on the larger scale. And large-scale “natural” experiments are rare and tend not to have anything like proper controls that would allow conclusive results.
But unlike many other social sciences, the difficulty for economists is that many economists wear more than one hat. The same economist can at one time engage in economic science, at another in economic theory, and yet another in economic practice. I would suggest that the most fruitful course of action for economists would be to be clear about which hat they wear at any given time. This means clear to others, but even more importantly to themselves. As Richard Feynman said in Cargo Cult Science,
The first principle is that you must not fool yourself — and you are the easiest person to fool. So you have to be very careful about that. After you’ve not fooled yourself, it’s easy not to fool other scientists. You just have to be honest in a conventional way after that.
This means, among other things, to be honest with oneself when advising on economic practice, recognizing to what extent it is informed by one’s economic science, and to what extent by one’s economic theory.
