What Milton Friedman Got Wrong: Biologists Destroy Homo-Economicus

Homo-economicus vs biological adaptation

Share with your friends

More share buttons
Share on Pinterest

By David Sloan Wilson

One of the most influential articles published in the field of economics is Milton Friedman’s (1953) “The Methodology of Positive Economics”, in which he argues that people behave as if the assumptions of neoclassical economic theory are correct, even when they are not. One of the most influential articles in the field of evolution is Stephen Jay Gould and Richard Lewontin’s (1979) “The Spandrels of San Marcos and the Panglossian Paradigm”, which argues against excessive reliance on the concept of adaptation.

Different disciplines, different decades. No wonder these two classic articles have not been related to each other. Yet, there is much to be gained by doing so, for one reveals weaknesses in the other that are highly relevant to current economic and evolutionary thought.

The reason they can be related to each other is because Friedman relied upon an evolutionary argument for his “as if” justification of neoclassical economics. I cannot improve upon his own framing of the problem:

The abstract methodological issues we have been discussing have a direct bearing on the perennial criticism of “orthodox” economic theory as “unrealistic” as well as on the attempts that have been made to reformulate theory to meet this charge. Economics is a “dismal” science because it assumes man to be selfish and money-grubbing, “a lightning calculator of pleasures and pains, who oscillates like a homogeneous globule of desire of happiness under the impulse of stimuli that shift him about the area, but leave him intact”; it rests on outmoded psychology and must be reconstructed in line with each new development in psychology; it assumes men, or at least businessmen, to be “in a continuous state of ‘alert,’ ready to change prices and/or pricing rules whenever their sensitive intuitions … detect a change in demand and supply conditions;” it assumes markets to be perfect, competition to be pure, and commodities, labor, and capital to be homogeneous.

Friedman admits that the orthodox theory’s assumptions about human preferences and abilities, which are often labeled Homo economicus as if they are a description of a biological species, are manifestly unrealistic. Yet, he claims that they are still predictive of human economic behavior by way of three analogies. First, trees distribute their leaves as if they are maximizing their exposure to sunlight, yet no one pretends that they are performing optimization equations. Likewise, an expert pool player acts as if he is performing complex calculations when making his shots, when in fact his behavior has been molded by countless hours of play. Finally, a firm acts as if it is maximizing its profits, when in fact its continuing survival is the result of a selection process in which the non-optimizing firms were eliminated.

Get Evonomics in your inbox

The first is an example of genetic evolution, the second is an example of individual learning, and the third is an example of cultural evolution. In all cases, a process of selection results in entities that behave adaptively, as if they are solving complex optimization equations, when mechanistically they are doing nothing of the sort.

Evolutionary biologists will recognize Friedman’s point as a distinction between ultimate and proximate causation. Ultimate causation explains why a trait exists, compared to many other traits that could exist, based on the outcome of a selection process. Proximate causation explains how the trait exists in a physical sense. Sunflowers turn towards the sun because selection has favored phototropism (the ultimate explanation); but within each individual sunflower is a physiological mechanism that causes the plant to do so. The proximate explanation need bear no resemblance to the ultimate explanation, other than to reliably cause the adaptive behavior to come into existence.

So far, Friedman is standing on firm evolutionary ground with his “as if” argument. Evolutionists frequently reason about the properties of species “as if” they are maximizing their fitness, without worrying about the proximate mechanisms. As a simple example, we can confidently predict that many desert animals are sandy colored to avoid detection by their predators and prey. The prediction holds true for different kinds of desert animals, such as insects, snails, reptiles, birds, and mammals, even though different proximate mechanisms in these animals cause the sandy coloration to develop. The ability to predict the properties of organisms in functional terms, without reference to proximate causation, is one of the most powerful features of evolutionary theory.

But reasoning on the basis of adaptation delivers the correct answer only if the trait in question is a product of selection and if we have correctly identified the relevant selection pressures. If the trait isn’t adaptive in any sense, we’ll be wrong. If we assume that the trait is a solution to one adaptive problem (such as the need for a foraging animal to maximize energy intake per unit time), when it is a solution to another adaptive problem (such as the need for a foraging animal to manage a trade-off between energy gain and predation risk), we’ll also be wrong.

That’s where Gould and Lewontin’s “Spandrels” paper comes in. They chastised some of their evolutionist colleagues for assuming that every trait must have an adaptive explanation and for accepting adaptive “just-so” stories without adequate proof. They outlined a more comprehensive approach that requires strong evidence for any given adaptationist explanation and reflects the many ways that nonadaptive traits can persist in a population. The compleat evolutionist might begin with an adaptationist hypothesis to explain a given trait, but then tests the hypothesis and modifies it as warranted, keeping both other adaptation and non-adaptation hypotheses in mind as live options. Compleat evolutionists also study proximate mechanisms, development, and phylogeny in conjunction with their focus on natural selection.

Some evolutionists complain that Gould and Lewontin created a straw man with their critique, but their portrait of “naïve adaptationism” accurately describes Friedman’s defense of neoclassical economics. He assumed that one or more selection processes (genetic, learning, or cultural) resulted in people who resemble Homo economicus as far as ultimate causation is concerned. He did not consider other adaptationist or nonadaptationist hypotheses. He did not indicate that proximate mechanisms, development, and phylogeny need to be considered along with ultimate causation. The only evidence that he provided to support his hypothesis was to claim that economic policy based on the orthodox theory was successful. His “as if” argument was evolutionary, but not evolutionary enough.

The weakness of Friedman’s article, when related to Gould and Lewontin’s article, reveals a widespread problem in the basic and applied human social sciences. All accounts of human social behavior that are not creationist strive for consilience—consistency with other branches of knowledge. An economic or social policy that ignores the way we are as a species and the way that cooperation evolves in all species is no more likely to succeed than an architectural plan that ignores the laws of physics. Yet, for complex reasons, evolutionary theory has been avoided as an explanatory framework for most branches of the social sciences since before most of the current experts were born. When theories and policies derived from the social sciences are related to modern evolutionary science, they often fail the consilience test as miserably as Friedman did in 1953.

Social scientists and policymakers need to become compleat evolutionists, no less than biologists. The bad news is that a lot of work needs to be done for our current theories and policies to pass the consilience test. The good news is that when we start to earn passing grades, our economic and social policies will start working better than they do now.

Originally published here.

2016 February 13

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!

  • Pingback: Economics vs Biology | Ted Howard NZ's Blog()

  • Jan de Jonge

    The articles about the homo economicus, or about the role of greed in economic action etc., take as a premise that economic theory would be quite different when it would describe the economic agent as for instance simply as homo sapiens. And that, moreover, economic policy would be quite different too.

    I agree with regard to the reductionist’s economic theories, the mainstream, that pretend to explain economic processes as aggregates of individual actions. Though the most recent reductionist theory, the new classical theory, tried to circumvent the problem of aggregation (and coordination) by using one representative agent, with perfect foresight, in their models.

    In contrast, macroeconomic theories are not very sensitive for the characterization of economic agents because they operate with aggregates as national income etc. Ragnar Frisch was the first to make the distinction between microeconomics and macroeconomics in 1933. Keynes adopted this distinction and defined micro-economics as the study of the allocation of a given quantity of resources, whereas macroeconomics studies the input and employment determinants. In the neo-classical and new-classical schools micro- and macroeconomics are not regarded as separate domains of inquiry. The epistemological claim of methodological individualism requirs that all economic phenomena are explained in terms of individual actions. Simplified, we could say that these microeconomic schools embrace the scarcity definition of economics whereas in macroeconomics economic processes are defined in terms of production, consumption and investment. The Keynesian theory (and its successor the neo-classical-neo-Keynesian synthesis) that was dominant till the 1970s was the only macroeconomic theory in economic science

    The lesson to be learned is that in order to get rid of the homo economicus one does not need a new economic theory but a return to macroeconomics. This would also change economic policy as it liberates us from the dominance of monetary policy and opens the possibility to consider fiscal policy.

  • Norm Roulet

    The Roulet Proof: Lead Perturbation (neurotoxin poisoning) causes Chaos, defining biological and social evolution and mutating humans since 6,500 BC, at slave levels, creating uncertainty, reduced with awareness and lead poisoning eradication, which must be factored into all economic and general systems models today, including understanding the crucifixion of Christ, the American Holocaust and Nazis, which were far from optimal market activities. An approximate “perturbation solution” is obtained by truncating the
    series, usually by keeping only the first two terms, the initial
    solution and the “first-order” lead poisoning perturbation correction. Chaos: When the lead perturbed present determines the future, but the approximate lead perturbed present does not approximately determine the future. Obviously causing uncertainty – a fundamental limit to the precision with which certain pairs of properties can be known simultaneously.

    • Deadsys

      Do we mean to continue and flourish or terminate w a bang. That is the essential question. Everything has to cooperate and associate w literally everything else in the universe. We arenot each in individual vacuums.

  • Pingback: Evonomics -Consilience – a Tale of two classic Essays | Ted Howard NZ's Blog()

  • Mitch Diamond

    Hocus-pocus and conflation. Mixing and matching cultural and biological evolution as if they obey the same principles and then applying it to economics…all hand waving and smoke and mirrors. It’s fine to suggest applying evolutionary theory to economics, but that means selection and adaptation, not the completely contrived and disingenuous spandrel paradigm.

  • Agree entirely with the general thrust of this article, and evolution is much more complex than even this article explicitly acknowledges.

    Fitness of any phenotypic trait is something of a sum over life history of the positive and negative impacts of that trait (and any related traits by their degrees of relatedness).
    In considering the term life history, one must be cognisant of all the variations in environment encountered, in both space and time.
    When considering the aspect of time, it is the time that the particular allele enabling this particular phenotype actually remains in the population. Thus an allele that has a small negative impact most of the time, but every 100 generations results in a probability of .9999 survival in some specific event that occurs on average every hundred generations, might still be strongly selected in a population. Similar numbers can apply with respect to particular habitats encountered. A population may be the only population that can successfully survive in the conditions at some particular site, and from the secure physical base, spread to surrounding areas even though in many other senses other species would normally out compete them (if not for that secure base area where they have no competition). Sometimes those specific events or contexts are not at all obvious.

    So evolution in this sense is very much like a hologram – where every bit in the image contains a little bit of information and influence from the entire object being imaged. Evolution deals with this summation of probabilities (positive and negative) across all aspects of space and time that particular organism exists in.

    Similarly in the mimetic context.

    The dimensionality of the resulting probability maps, the dynamic pressures resulting from constantly evolving n-dimensional contours, is amazing to contemplate, and even more interesting to experience as the diversity of organic, cultural and intellectual life we exist in.

    In this sense, given that exponential expansion of our ability to generate and manipulate information, coupled with our exponential expansion of ability to automate processes at ever higher levels, is leading us to the situation where the entire concept of markets and exchange is now becoming the greatest existential risk facing our species.

    When one looks at evolution from a systems complexity perspective, it is possible to characterise all major advances in complexity of living systems as the emergence of new levels of cooperation, with requisite stabilising strategies (as per Axelrod et al)

    We can relatively easily generate fully automated systems that would deliver all the goods and services required to allow any and all individuals to do whatever they responsibly choose, but producing such a system takes most of those people outside the economic system and destroys much of the economic value present in our current exchange based (scarcity based) systems. Thus our focus on exchange now threatens to produce social pressures that put everyone in danger.
    Competition alone cannot get us out of this hole, it requires massive cooperation, with all the attendant strategies to effectively prevent overrun by cheating systems (something our current economic system has failed completely to do).

  • John M Legge

    Surely the most critical flaw in Friedman’s reasoning is that the selfish short term behaviour ascribed to homo economicus is not, in fact, optimal. As observation of successful enterprises and national economies confirms, the most successful involve substantial elements of cooperation and co-respective behaviour. The real issue is that one cannot reduce a description of human society to a set of ordinary differential equations without assuming that all economic agents are monomaniacs. Drucker put it clumsily but correctly when he pointed out that even approaching real world behaviour required partial differential equations almost none of which have analytical solutions. Recognising the complexity of the real world means accepting the impossibility of making correct systematic predictions. Even compleat evolutionists can’t go much further than explaining how the past led to the present.
    Orthodox economists assume that the past doesn’t affect the future, so they no longer study economic history. That single assumption creates an unbridgeable chasm between orthodox economics and reality.

    • David Burns

      There are still plenty of economic historians. Maybe you mean that ordinary economists don’t read their work?

      • John M Legge


  • David Burns

    You claim that the “only evidence that [Friedman] provided to support his hypothesis was to claim that economic policy based on the orthodox theory was successful.” I haven’t read the article in years, but I thought he claimed that the model was only to be used to generate hypotheses, which would then be tested statistically. Wouldn’t these statistical tests count as evidence for or against the particular hypotheses? Of course, this does not explain the peculiarity of economists, that they make these claims that imply that the model is just a simplification, yet if you propose to test a hypothesis that ignores or contradicts this basic paradigm, they have a fit.

  • Ishi Crew

    My 2 cents:

    1. The Friedman quote is quite interesting. For the last few I’ve been reading alot of writing by ‘heterodox’ economists, and they make Friedman’s critique or recognition of the problems with the standard ‘neoclassical economic’ model over and over again. Many of thse points I have also read in other papers by other famous economists from years even before Frideman’s 1953 paper.
    So basically, to me it looks like they are ‘beating a dead horse’. Very few actually try to create more realistic models — though alot of current economists do in varying ways—include psychoklogy via H Simon’s bounded rationality,or behavioral economics, relax assumption of perfect information, and permit multiple equilibria or optima rather than some unique stable equilibrium.
    Models of this sort can be very mathematically complex, and that is one reason people don’t try to construct them in favor of simply pointing out limitations of current ones. Also, it seems some of this ‘failure of will and technique’ is then used to justify ‘math bashing’. —- critics say economic theory has lost its way in math, so any good—i.e. heterodox—theory will just use words.

    2. It is quite common for people in scientific disputes to sort of set up ‘two sides’—eg heterodox economics versus neoclassical and its more current variants (which possibly should not be termed neoclassical though they are derived just by modifyling some assumptions). In biology, another example is one has ‘D S Wilson (group selection) versus Richard Dawkins (or S Pinker—‘selfish genes and ‘the false allure of group selection’).

    My view is this article sort of does this in a ‘weak’ way. ‘Friedman was wrong or incomplete because he neglected proximate mechanisms’, or basically said they are irrelevant. This is done in statistical mechanics in physics—-people often don’t care what individual atoms do (or think); they just abstract away from them and use a law of large numbers and find an aggregate ‘equilibrium’ solution which describes behavior of interest—eg the direction of time.

    This is also the view of the ‘invisible hand’ basically—who cares what individuals do in an economy—all you really need to know is its aggregate behavior—levels of employment, what kinds of products are produced and consumed, whether these are stable in time or fluctuate, etc.
    Biologists sort of made the opposite mistake for a long time—which is why group selection was seen as both a false and nonsensical line of reasoning, as well as a taboo subject. People like Dawkins, George Williams, and others didn’t want to think that an ecosystem or a population actually acted in aggregate the way Friedman describes an economy as acting—as if it has an ultimate cause or purpose, and which can be mathematically descibed by some sort of optimization process. Alot of biologists it seemed and seems to me are anti-math—they want to study ‘proximate causes’—how some little animal fits into some little niche, or how some gene operates in a particular organism. This is epitomized by the current obsession with ‘widespread genetic association’ (GWAS or something, which is mathematical, but statistics not optimization). People are continually dissapointed when they can’t find a gene for autism, depression, criminality etc. They don’t want to Consider the possibility these are actually ‘group level ‘ properties (partly because that brings in politics—if diseases and social problems have a social component, rather than being due to some genetic defect in an individual, then fixing them requires social change rather than finding a drug cure).

    3. I personally don’t think there is any problem with Friedman’s view—but, as noted its just half the story. So unlike the Koch’s (who argue that actually inequality, poverty, and destroying some lives is good because that’s the way the economy funxtions—you have to break a few eggs to make an omelet which has other uses, so not all the chicks can hatch) —they basically forget about the ‘means’, only the ‘ends’ I think you have to think about proximate mechanisms too—eg child labor, living wages, environmental destruction, effect of ‘free trade’ deals on wages, etc.

    As commentators to Pinker’s edge magazine article pointed out , there is also really no conflict or problem with either pro-group selectionist and pro-selfish genes views. They are looking at different things, or from different coordinate systems.

    . There is a problem however—-in practice. This is why Friedman didn’t care much about Pinochet’s political actions in Chile, onlly the economy. A similar problem I think exists with the views in the current article. Its basically doing what I said the ‘heterodox’ economic critics of Friedman do—beat a dead horse.

    I think the whole article, basically the Lewontin-Gould view—can be summarized as saying an evolutionary process can be described by a dynamical equation with 3 kinds of solutions —one is an adaptive optimum, one is a maladaptive non-optimum (extinction) and the the third class is a ‘fitness landscape’ with multiple peaks (optima) of varying heights (fitness values) along with many trajectories which basically stay in valleys and wander around or explore the fitness landscape. They are neither ‘optima’ nor ‘extinct’. Fisher’s ‘runaway process’ would be an example and there are many more (eg ‘path dependent and nonergodic processes’ ).

    These emphasize that proximate causes do matter because they can lead to different optima, nonoptima, or just aimless wandering in the desert between peaks and dead seas. I tend to think the way both science and economics is done —the ‘proximate causes’ of current academic, research and diffusion of aquired knowledge—basically recreates the same problems they identify—the fact that the ‘ultimate cause’ or ‘purpose’ is non-optimal (ie endless, somewhat incoherent debates between heterodox and mainstream economists, group selectionist vs selfish gene types)

    The article concludes ‘social scientists and policymakers’ need to take the lesson of this article seriously. My view is that is is basically (though not completely) wrong —-thats like saying ‘lets find the gene, or expert social scientist or policy or policymaker who will fix the system’.. My view is these issues are partly social—what is really needed is for ‘social’ and other scientists to be actually more social—get out in the world—and also make everyone in a sense a scientist and policymaker, rather than assume some expert in an ivory tower or political office has all the solutions.

    The current standard approach which mostly seperates ‘macro’ and ‘ultimate’ issues from ‘micro’ and prximal’ causes to me is a kind fo schizophrenia—and is why people like George Price (who i never heard of until a few years ago, though his equation is actually the most basic comprehensive statement of group selection) dropped out of research and academia and went to serve the poor on the streets. you have to choose a side in this world. (This is one reason i decided i could never make it in academia–it would be like a disfunctional marriage, sortuh scene i grew up in—i wanted to study the basically pretty small and unknown field (though now getting now) which looks at math models which are ‘complete—both micro and macro’. People I talked to either had never heard of them or viewed them as nonsense—even people at SUNY Binghampton which i visited. . .