Why Economists Should Start Behaving Like Scientists

The empirical revolution needs a general theory

Share with your friends

More share buttons
Share on Pinterest

By David Sloan Wilson

I welcome the attention that Noah Smith has drawn to two “big think” pieces, one by Nick Hanauer and Eric Liu and the other by myself, which are both cut from the same broad cloth of evolutionary and complexity theory. Smith comes across as an open-minded skeptic. He likes some aspects and is unimpressed by others. Most of all, he insists on empiricism. Here is how he ends his critique.

“But I think that more important than any of these theoretical changes – or the evolutionary theory suggested by Wilson – is the empirical revolution in econ. Ten million cool theories are of little use beyond the “gee whiz” factor if you can’t pick between them. Until recently, econ was fairly bad about agreeing on rigorous ways to test theories against reality, so paradigms came and went like fashions and fads. Now that’s changing. To me, that seems like a much bigger deal than any new theory fad, because it offers us a chance to find enduringly reliable theories that won’t simply disappear when people get bored or political ideologies change.

So the shift to empiricism away from philosophy supersedes all other real and potential shifts in economic theory. Would-be econ revolutionaries absolutely need to get on board with the new empiricism, or else risk being left behind.”

I can’t help but remark on the irony of this stance. By Smith’s own account, the field of economics is experiencing an empirical revolution. Unlike the past, it has become necessary to test theories against reality. That places the field of economics many decades behind the field of evolution and numerous fields in the human social sciences that have been rigorously evidence-based all along. Earth to the economics profession: Welcome to Science 101!

But there is more to Science 101 than the need to test theories. Let’s imagine that there were ten million cool theories out there. How long would it take to test them? Hundreds of millions of years. Does it really need explaining that the choice of the theory to test is important? Does Smith really believe that any old idea that comes into the head of an economist is equally worthy of attention?

Get Evonomics in your inbox

The main reason that the so-called orthodox school of economics achieved its dominance is because it seemed to offer a grand unifying theoretical framework. Too bad that its assumptions were absurd and little effort was made to test its empirical predictions. Its failure does not change the fact that some unifying theoretical framework is required to prevent the “ten million cool theories” problem. What Nick Hanauer, Eric Liu, and I are saying (Smith misses that our arguments are cut from the same cloth) is that a combination of evolutionary theory and complexity theory offers the best prospect for a unifying theoretical framework. We are not alone in this assessment. A volume titled “Complexity and Evolution: A New Synthesis for Economics” based on a conference organized by Germany’s Ernst Strungmann Forum last year, will be published by MIT Press later this year.

It’s easy to get bogged down in the semantics of whether something counts as a new synthesis or not. This was a major discussion point at the conference, with some of the participants arguing for a “continuous with the past” view and others arguing for a “break with the past” view. Interestingly, a similar conversation is taking place over the phrase “Extended Evolutionary Synthesis” in my field of evolutionary biology, which I will be reporting upon in the Evolution Institute’s online magazine This View of Life. These conversations are not just word splitting. They bear critically upon the hypothesis formation half of the scientific process, which is just as important as the hypothesis testing half.

I end with a concrete example of an empirical research program in economics informed by evolutionary theory. Elinor Ostrom was awarded the Nobel Prize in economics in 2009 for her work on how groups successfully manage common-pool resources. Her achievement was primarily empirical. Based on a worldwide database that she compiled, she derived eight core design principles that were required for common-pool resource groups to successfully avoid the tragedy of the commons. The theory that informed her research was a blend of New Institutional Economics, game theory, and evolutionary theory. The influence of the latter grew throughout her career, including our own collaboration that led to our article titled Generalizing the Core Design Principles for the Efficacy of Groups, part of the 2013 special issue of JEBO to which Smith alludes in his critique. This article places Ostrom’s “core design principles” approach on the foundation of multilevel selection theory and predicts that most groups whose members need to work together to achieve common goals require the core design principles. The theory also predicts that the need for the core design principles should be scale-independent, although the mechanisms required to implement them at a large scale (e.g., a nation) can be very different than at a small scale (e.g., a neighborhood). These predictions are eminently testable on groups as diverse as schools, businesses, and even the cultural evolution of large-scale societies over the broad sweep of human history. Testing the predictions can go hand-in-hand with helping to implement the core design principles in real-world settings, thereby improving the quality of life in a practical sense.

To conclude: I thank Smith for his open-minded skepticism and wholeheartedly agree with the need for empiricism in economics, along with all other branches of science. I hope that he can agree with the need to go beyond “ten million cool theories” for the hypothesis formation half of the scientific process.

2016 March 20

Donating = Changing Economics. And Changing the World.

Evonomics is free, it’s a labor of love, and it's an expense. We spend hundreds of hours and lots of dollars each month creating, curating, and promoting content that drives the next evolution of economics. If you're like us — if you think there’s a key leverage point here for making the world a better place — please consider donating. We’ll use your donation to deliver even more game-changing content, and to spread the word about that content to influential thinkers far and wide.

 $3 / month
 $7 / month
 $10 / month
 $25 / month

You can also become a one-time patron with a single donation in any amount.

If you liked this article, you'll also like these other Evonomics articles...


We welcome you to take part in the next evolution of economics. Sign up now to be kept in the loop!

  • David, this is an interesting article. Thanks.

    You might be interested in a response to your article above:

    Also this is a response (from the same author) to Noah’s post:

  • Here’s another related response to Noah: not directly to you, but it overlaps I think (from a different author):

  • Pingback: Economics — the ‘ten million cool theories’ problem | LARS P. SYLL()

  • Robert Waldbillig

    In my earlier days I met a man who was a big shot in USAID…I asked him why was it that the US was so successful and a country like Mexico was not when I had worked around Mexicans and it was obvious they were hard working and diligent, ,etc…his answer was education, capital reinvestment and a whole litany of reasons, but capital reinvestment really struck me as a major contributing factor in the existence of and continuation of any given enterprise …

  • John Foster

    I am rather surprised at the seeming ignorance here of the existing literature in evolutionary economics attempting to build upon complex adaptive system foundations. This has been going on for at least two decades. I am not going to run through all the people who have contributed (Ulrich Witt, who was involved in the Forum, knows all these people and may well have discussed them) but I would particularly like to point to the work of Peter Allen, a former student and colleague of Ilya Prigogine, who was a standout pioneer in providing complex system foundations for evolutionary economics. I think everyone should read his work and that of others doing related research before going off and reinventing the wheel.

    • David Sloan Wilson

      Thanks for pointing out Peter Allen, but no need to scold me for ignorance! I’ve been reading the economics literature for years now and the degree of non-overlap among its various branches is stunning.

      • John Foster

        I agree about the non-overlap and it’s been getting worse as the decades have passed. There is even very little overlap between the various ‘heterodox’ schools in economics! I seriously don’t believe, after a whole career as an academic economist, that we are not dealing here with a science. So, I applaud your earnest attempts to encourage the construction of a scientific basis for economics!

        • John Foster

          Oops, there is a “don’t” in there that should be omitted!

          • So what is needed for the construction of a *scientific basis of economics*?

            Complexity is only a partial answer, too as it relates to the aggregate level of interactions. However, it points the way through the work of Herbert Simon and Hayek on towards psychology and back to the first Austrian economists who actively tried to incorporate a psychological perspective into economics.

            Thus, my answer is: we need a proper theoretical basis for economics in psychology. As practice shows, It seems to be quite hard, however, based on the institutional foundations and blind spots of 2500 years of ‘western science’.

            It is something Wittgenstein spend a great deal of effort on in the latter part of his life, trying to get a handle on meaning, interpersonal communication, and gestalt perceptions after he had realized the shortcomings of the logical, rational approach of the Tractatus logico-philosophicus.
            Popper started his career with a focus on the psychology of discovery to retreat to a rather sterile critical-rational perspective on the methodology of science as ‘logic of discovery’. Kant didn’t dare publish on the psychological contrapoints to his metaphysical ‘a priori’ and ‘thing-in-itself’, only lecturing and revising rudimentary drafts on psychological topics. Kant’s illegitimate heir apparent, Hegel created instead an idealistic Weltgeist to be complemented by Marx’ materialistic antithesis of social mechanics – both leading nowhere ‘healthy’.

            However, there is a quite productive stream relevant for the buildup of psychology in the sceptics and medical researchers of antiquity, partly affecting Aristotle, leading through Persian and Arabian philosophers to the humanist Renaissance, Scottish enlightenment including Adam Smith, educational reformers, such as Comenius and early psychologists, such as Beneke, who influenced Ernst Mach and in turn Bühler, Külpe and Selz of the Würzburg school.

            That’s where we need to pick up the thread again, it seems.

      • David Sloan Wilson has been reading the economics literature for years now, yet without fail gets the very basics wrong (see e.g., Many advocates of “Intelligent Design” similarly claim expertise in evolutionary biology as a prelude to making a hash of that discipline.

        Here, Wilson, also as usual, openly insults economists (“Earth to the economics profession: Welcome to Science 101!”, not to mention the obnoxious title) without any understanding of the intellectual history or content of the economic literature. The increase in the proportion of econometric research over the last several decades is of course not because economists suddenly learned that empirical work is important, it’s because of the coincident massive increase in computing power and widespread availability of large-scale datasets.

        Modern economics is, and has long been, heavily empirical, does not mindlessly promote market fundamentalism, and is methodologically pluralistic, including often drawing on evolutionary reasoning. Readers of this site could contrast any of Wilson’s claims with the recent piece by Herb Gintis, ( on this very site! As Gintis notes,

        “It is a serious error to reject standard economic theory on the grounds that it supports a free-market ideology. It does nothing of the kind. Correctly deployed, it carefully explains where, how, and when to intervene in the regulation of market exchange. Evolutionary and behavioral game theory are wonderful additions to the economist’s repertoire, but they complement rather than undermine traditional public sector economic theory.”

        Gintis’s observations cannot be reconciled with Wilson’s writing, nor with the substantial majority of the other strawman-bashing on this cite. There is and has long been a vibrant scientific nexus between economics and biology; that productive effort is ill-served by David Sloan Wilson’s offensive and defiantly ignorant public commentary on economics.

        Chris Auld.

    • Adrian S

      Literature is not the same with hard core and reliable data.

      In the end, it seems like you’ve missed the entire point of the article, the author blames the lack of empirical evidence that is used in the actual literature, not the lack of literature. This holds true even for newer fields that attempt to fix this. While their attempts are not recent, their rate of adoption is low and the actual data is still not reliable.

      The paradigm does seem to shift and younger generation of entrepreneurs, mid-level management seem to be very interested in measurable data and making informed decisions. But the reality is that most businesses and public policies do not use reliable data and are mostly driven by hear-say and gut-feeling decision makers.

  • Jan de Jonge

    Citing Noah Smith’s assertion that economic science experiences nowadays an empirical revolution, David Wilson concludes that, unlike in the past, it has become necessary for economics to test theories again against reality. Thus it is Wilson’s view that economic science was for a long period (many decades) not evidence-based.
    I contest that claim.

    Smith refers to an article in BloombergView by Justin Fox, who cites an article of Daniel Hamermesh in the Journal of Economic Literature (2013), “Six decades of Top Economics Publishing: Who and How”. Hamermesh has looked at the articles published between 1963 and 2011 in three major journals (American Economic Review, Journal of Political Economy and Quarterly Journal of Economics) and among other things registered the share of purely theoretical and empirical articles. He notices a small rise in the share of theoretical articles between 1963 and 1983 (from 52% to 62%) and than a stark decline to 40% in 2003 and 27% in 2011. What is important is the share of empirical articles (I do not distinguish between articles in which authors have used existing data or whether they have constructed their own data). This share declined since 1963 till 1983 from 48% to 38% and than jumped to 60% in 1993 and further to 72% in 2011.

    The decline in empirical based articles can be attributed the New Classical School that was dominant from 1975. But the share has never been less than 40%. In 1993 Thomas Mayer published his book “Truth versus Precision in Economics” which contains precisely the discussion within the economic profession between the pro and cons of a formalist (mathematical) and an empirical approach; particularly between adherents of New Classical theory and their opponents.

    On the basis of Hamermesh article it is impossible to claim that economics science in general had abandoned empirical research since 1963. It is understandable that economic science is criticized but critics should not use the wrong arguments.

    I conclude that David Wilson’s claim is not evidence-based.

    • David Sloan Wilson

      Thanks for your comment and the information that you have provided. I take your point that there is plenty of empirical research within the economics profession writ large. Nevertheless, it remains true that neoclassical economics is extremely top heavy with respect to theory vs. empirical research. That’s why the fields of experimental and behavioral economics are accorded such significance–and they lack a theoretical framework of their own, which is why they result in a laundry list of anomalies and paradoxes.

      • Jan de Jonge

        Thanks for your reaction. I really appreciate this. I think that there are a lot of misunderstandings. The theory you oppose is not neo-classical but new classical theory. The important difference is that only new classical theory assumes rational expectations and efficient markets and developed the dynamic stochastic equilibrium theory (DSGE). This is a formal (mathematical) theory with only one representative agent that is criticized inside the economic profession (see the book of Thomas Mayer) and by outsiders. This is the theory that gained prominence in the 1970s and 1980s and displayed the mathematical virtuosity that the AER was keen to promote. But prominent economists as Blinder, Krugman, Akerlof or Stiglitz and many others did not take part in this school.

        In reaction to the critique of Kahneman and Tversky economists started with laboratory experiments. A forerunner was Vernon Smith who received the Nobel Prize in 2002 for his researched with market-designed experiments. His experiments demonstrated, he thought, that the axioms of expected utility theory were not violated in impersonal exchange (in contrast to personal exchange). The violations of these axioms were the core of the critique of Kahneman and Tversky that neo-classical theory was descriptively wrong (though normatively correct).

        Many followers of Kahneman, who called themselves behavioral economists, were strengthened in their criticism of economic theory when new classical theory promoted an even stronger version of the rational economic agent. For decades they have not produced much more than the list of anomalies and paradoxes you wrote about (though I make an exception for Gerd Gigerenzer, who does not call himself a behavioral economist). The main reason is that most of them are not economists but (social) psychologist. I think that the claim that they are behavioral economists is misplaced. They write about a lot of issues from an economic perspective (rational choice theory) but not about macro-economics at all (Ariely is an example).

        A different group is the group around Camerer and Thaler, but Thaler says that if economists abandon the homo economicus the field of behavioral economics will disappear (Misbehaving, 358).

        The latest book of Robert Gordon (The Rise and Fall of American Growth) is a major example of the empirical work done by many economists especially those working for the NBER. Throughout his career Gordon has done empirical research and he was (if my memory is correct) long associated with the NBER.

        A last example of empirical research is the influential article by Reinhart and Rogoff who related the debt ratios of 20 countries over a longer period with economic growth. This article published in 2010 provided empirical foundation in support of austerity policies at reducing the high public debt levels that emerged in the aftermath of financial crisis in 2007-09. But research by Herndon, Ash and Pollin (CJE, 2014, vol. 38) on the basis of more advanced data detected major imperfections in the article of R&R. They conclude that R&R created a false image that high public debt ratios inevitably entail sharp declines in GDP growth. This opened a discussion about fiscal policy. Thus also in the political domain empirical economic research has been important.

  • Sanjit Dhami

    Economics has always been empirical, some fields such as labor economics more so than others. The “causality revolution” in economics and the growth of experimental economics has increased the scope of what we may learn from the empirical data and it allows us to make better choices between theories. It is true that neoclassical theories have not had great success when pitted against field and experimental data. All that is fine.
    However, I am extremely surprised by the lack of awareness shown by the author (and a couple of the comments that I saw) on the lack of a theoretical framework in behavioral economics. Here is what the author writes in response to one of the comments:
    “That’s why the fields of experimental and behavioral economics are accorded such significance–and they lack a theoretical framework of their own, which is why they result in a laundry list of anomalies and paradoxes.”
    This is just false.
    There is now a very well defined and strong theoretical framework in behavioral economics that provides an alternative framework of analysis to make sense of the empirical evidence that the standard neoclassical model is unable to explain. This includes the use of prospect theory to explain data from risk, uncertainty, and ambiguity; the use of a small number of models of other-regarding preferences to explain social preferences; the use of hyperbolic discounting and attribute based models to explain time choices; the use of quantal response equilibrium and level-k models (there are several others) to explain strategic choices; and so on.
    May I invite the author to pursue this in a recent 1800 page book published by Oxford University Press that clearly links together the evidence and the theory (Sanjit Dhami, 2016, Foundations of Behavioral Economic Analysis, OUP). It took 12 years to write and draws liberally on economic theory, psychology, sociobiology, sociology, anthropology, neuroscience, and other related disciplines.
    Here is an amazon link to the book:
    Please, lets cut through the high degree of ignorance on display, if possible.

    • Don Gropp

      Seems to me that one environmental or human catastrophe would nullify any scientific rationale.

  • Macrocompassion

    Economists can’t think like scientists because they have not been trained in science. Were they to begin again with a different science or possibly manage to replace their memory files with a better scientific method for understanding macroeconomics, then there is a chance that they will manage to become knowledgeable in what I claim is actually a true scientific subject and not the past, dismal and pseudo-science that this important subject has become.

    My recent book “Consequential Macroeconomics” can supply this need to convert the past pseudo- into a more precise true-science. Write to me at [email protected] for a free e-copy and get the shock of your life for the better understanding along with a good dose of pure logic and formality of organized thought, for a revelation about how our in our society the social science really works.

  • Don Gropp

    If one of John Kenneth Galbraith’s students presented to him a law of economics, he would give that student an F. He would tell them there are no laws because human activity is too capricious. Has something changed?