In defence of mainstream economics
After the 2008 global financial crisis, nearly every month or so an essay goes viral that discusses, often at length, how the profession of economics needs an overhaul. Not that there is no truth in the statement but the offered reasons are often seriously deficient in their understanding of what economists do and the strengths and limitations of their tools and methods.
The usual criticisms are the following:
1. Economists ignore the importance of psychology, sociology, history and culture.
2. Economists think of human beings as some abstract mathematical machines.
3. Economists write very simplistic models of the economy forgetting that they don’t exactly depict the real world and that these models rely on very strong, and perhaps untenable, assumptions.
4. However much the economists wish it to be, economics is not a natural science.
Let us take these in order. A criticism like “Economists ignore the importance of culture” sounds profound until we realize that it is banal. For example, for most goods the quantity demanded decreases if the price increases. This statement is valid both in downtown Manhattan and in a tiny Indian village. Obviously, in making this statement I did ignore the cultural aspect but I leave it to you to judge the importance of culture and history here. Arguably my example was too simplistic, and indeed it was. It is only when economists are seeking such general principles that they ignore culture, which is rather rare. If an economist were to calculate the price elasticity of demand for a Mercedes—percentage change in demand for Mercedes with a one percent rise in price—(s)he would most certainly not overlook the culture and context of the setting. One could glance at the recent issues of The American Economic Review, a top journal in economics, to see that an empirical paper set in sub-Saharan Africa will never, based on the evidence there, would indulge in a policy recommendation for Scandinavia, or even India. So, it may break the hearts of many but economists are indeed aware of this issue called “culture” and have tried to study it since a long time. A visit to Google Scholar might be helpful in finding several papers that study culture and economic phenomena.
The same holds for criticisms like “Economists don’t talk enough to psychologists, sociologists etc”. Those offering this advice should also tell the economists, at some point, what to talk. Contrary to the myth of economists being closed-minded it is perhaps the most interdisciplinary topic. John Nash was a mathematician, Roger Myerson (Nobel Prize winner of 2007), Al Roth (Nobel Prize, 2012) and a number of others came from operations research, Daniel Kahneman (Nobel Prize, 2002) is a psychologist to name a few examples. To not leave the sociologists out, Mark Granovetter will probably win the Nobel Prize in the next few years. In any case, that some of his articles are among the most cited in economics bears testimony to the open-mindedness. In the competitive market for ideas if hanging out with sociologists on a daily basis was the most fruitful avenue of research most economists would have pursued it. There may be a reason why they aren’t.
Similarly overdone, and often misplaced, criticism relates to the use of mathematics and modelling in economics and the assumptions of rationality. The criticisms of the rationality assumption implicitly assume that no economist ever thought of relaxing this assumption, which is rather unfortunate given how mainstream the fields of behavioural economics and experimental economics that focus exclusively on bounded rationality are. In the 1970s perhaps yes, it was difficult to depart from rationality but not anymore. Also, the critics of rationality fail to appreciate the slippery slope. For example, one can easily answer the question “Why wages don’t fall during recessions?” by saying that firms are stupid. Imagine Newton saying that an apple came down from the tree because it liked the ground more than the tree. For obvious reasons a serious discipline would choose to avoid such tautological explanations of observed phenomena. It is not that economists think that modelling humans as homo-economicus is a perfect depiction of reality but perhaps it is often a better alternative than most others.
On the use of models we must compare it to economics done without models. That takes us to 19th century where most arguments were verbal and sounded compelling. Formal modelling helps us understand these arguments better as they enable us to understand one piece of an argument in isolation. And for that purpose a model has to leave out some details. A model of gravitation, for example, fails for both really tiny and huge bodies. And yet, it’s a great model and is also practically useful. As Dani Rodrik, an economist from Harvard University, says, “Models are never true; but there is truth in models.” So a good model has to be simplistic so long as it does not leave out the important details. No economist thinks that his/her model is a perfect depiction of reality in any situation but what is sought is a reasonable approximation. Yes, there exists a problem of “chameleon models,” as Paul Pfleiderer of Stanford University calls them, that talk about policy implications without checking the reasonability of the assumptions but its magnitude is not entirely clear.
On the use of mathematics, I would again quote Dani Rodrik: “Math essentially plays two roles in economics, neither of which is cause for glory: clarity and consistency. First, math ensures that the elements of a model—the assumptions, behavioural mechanisms, and main results—are stated clearly and are transparent. Once a model is stated in mathematical form, what it says or does is obvious to all who can read it. This clarity is of great value and is not adequately appreciated.”
To give an example, if in explaining a certain macro phenomenon I assume that a demand curve is upward sloping, a formal mathematical model will not leave any room for me to hide this assumption. Hence, the success or failure of the reasoning I offer will often depend on the sensibility and validity of my assumptions which is where, perhaps, the idea of doing economics as “science” comes from. Indeed, there have been recent critiques such as Paul Romer’s paper “Mathiness” where he argues that models in the theory of economic growth often hide substantive assumptions in mathematical symbols. But the prescription here is not to do away with mathematical models.
Having said all of this, I would not like to suggest that everything is perfect in economics today. There are several shortcomings especially in our predictive power. A lot many experienced practitioners have written extensively about other possible future avenues to improve the practice of economics. I personally quite enjoyed an essay by Robert Shilleron the importance of building in narrative and story-telling in the practice of economics. I strongly feel that a number of economists lack the art of journalistic story-telling, rare in journalism itself these days. Ariel Rubinstein from New York University has written numerous provocative and insightful essays, on the purpose of economic models and economic theory. One of my favourites is his contemplative essay titled Dilemmas of an Economic Theorist that discusses the role, if any, of economic theory in understanding the real world and improving it. I hope the critics from the media making sweeping conclusions like “Economists don’t understand the difference between correlation and causation”take the time out to get informed about what is really happening in this part of the world. It would improve both their understanding and the quality of criticism. Ill-informed criticism can create an illusion of profundity in some echo-chambers but is unlikely to attract the (likely) targeted audience.
Aditya Kuvalekar is an assistant professor of economics at the Universidad Carlos III de Madrid.