what sociologists could learn from economic reasoning


Fabio nicely articulates some key things that economists could learn from sociology; here’s a few quick, highly-stylized reflections on the reverse, namely, what sociologists could learn from economic reasoning:

  1. Many economically-oriented scholars (Arrow, Buchanan, Nozick, Sen), and rational choice-folks (Coleman), are highly interested in understanding and explicating the origins of structure and norms, rather than taking their existence for granted as much of sociology does.  There is much work left to be done in explicating these origins – where do networks, norms, structures etc come from – and economic reasoning seems to provide a fruitful tool.
  2. While the past undoubtedly plays an important role in what happens in the future (the general ethos in sociological analysis), economic reasoning suggests (and not necessarily just of the omniscient variety) that individuals are also able to reason and take a future-oriented approach to decisions, and not just myopically be linked to the past in path-dependent fashion. And, if its all the past and imitation – where does the new-new come from?
  3. Individuals do have preferences, independent of their social environment.  And, individuals also self-select into (and create) environments which creates an important confound challenging many sociological conceptions. 
  4. Human nature matters. Herbert Simon (1985) highlighted this fact (and Williamson frequently cites it), but it seems to have been lost in organizational and sociological analysis.
  5. Economic reasoning is highly attuned to the matter of costs, while extant sociological work tends to look at benefits (consider the literature in networks, knowledge, learning etc).  Analysis needs to be done with both benefits and costs in mind, or put differently, on a net-basis and at the margin.
  6. The data will lie, lie, lie. There is a strong empiricist bias in some circles which privileges data over theory.  Economics probably does not do any better (in fact, there is plenty of a-theoretical, correlational work out there), but, a scepticism of (heavily tortured) data is healthy. We’ve talked about the theory-data debate before, and, I am guessing most would side with theory.

Again, highly-stylized and quick. Of course, it seems that there are some scholars nicely working on both sides of the fence – economic sociology and sociological economics(?) – and thus it appears that some of the above conceptual holes are being addressed.

Written by teppo

March 13, 2007 at 4:34 pm

Posted in economics, sociology, teppo

10 Responses

Subscribe to comments with RSS.

  1. I like these points. They do read, however more like a tally of important problems than “lessons” to be learned from economic research. In this way, they seem to be a tall order for both disciplines as currently conceived.

    The origins of norms and institutions is the number one biggie. Agreed. However, there are two things wrapped in one there: first, there is the question: should norms and institutions be taken into account in explaining economic action? This was the institutionalist economics/economic sociology contribution. Worrying about the origin of norms and institutions assumes that we already agree that they are important. While being able to tackle “origin-type” questions is a virtue of the the type of reasoning (methodological individualism) that economics promotes, let us not forget that it was that same type or reasoning that stood in the way of the realization that they were important in the first place.

    The origins of innovation. Another biggie. It is less clear to me how economic reasoning might get us somewhere different that we are now on that one.

    I would qualify three by adding “current environment.” Talking about free-floating preferences without an account (ontogenetic or phylogenetic) of where they come from is one of the vices and not a virtue of economic reasoning.

    Human nature matters agreed. It is unclear however, whether the usual economic approach to human nature is the best. Sociologists are often guilty of ignoring human nature. Economists on the other hand are guilty of assuming the existence of one ex-armchair (which ends up being similar to that of a petit-bourgeois Englishman). In the true spirit of Herbert Simon, we must conclude that human nature is an empirical and not an armchair question. I think both sociology and economics will benefit greatly for more dialogue (and outright incorporation into) the current scientific disciplines in charge of discovering that nature (linguistics, cognitive science, developmental psychology, neuroscience, etc.). Neuro-economics is a great start. I would like to see a neurosociology (but not another ASA section). In any case, the Scottish moralists are great, but human nature is not a philosophical issue. It is a scientific issue.



    March 13, 2007 at 6:49 pm

  2. I agree with Omar about the question of human nature. Many sociologists would argue that they are much more in tune with the “real” human nature of humans because, well, they actually study them (e.g. Venkatesh), instead of just assuming that human nature is equal to an assumption about self-interest and calculative decision-making. Psychologists, of all disciplines, have cornered the market for explanations of human nature and we all know the problems that psychologists and economists have with one another.

    To your last point, yes the data can be twisted and tortured to make them agree to a certain story, BUT formal models do this even worse. They just bypass the data altogether and present a stylized story of things. So, while I agree with you that data manipulation is a problem among sociologists, it is a problem in almost any social science, including economics. And while I think formal models are good for getting at the core logic of an argument, I don’t think they alone are sufficient to produce an explanation for some observable phenomenon.



    March 13, 2007 at 7:37 pm

  3. Psychologists, especially of the evolutionary variety, would be shocked to see economics getting credit for the claim that human nature matters. Or, at least, they would say that a lesson that economics could learn from psychology is “Whatever human nature is, it’s complicated and can be expected to regularly undermine simple models of actors unless those actors are in environments that structure behavior in ways that discipline them toward acting like actors from simple models.”



    March 14, 2007 at 1:36 am

  4. Ditto the claim in #3 that economist recognize self-selection and “human creation of environments” while sociologists don’t. Quite aside from the obvious and long-standing tradition of sociological interest in socially [and self-] constructed realities, in my neck of the discipline it’s hard to find an article that doesn’t acknowledge and worry about self-selection. (It’s often folded into the catch-all category of other unobserved heterogeneity, but it’s nonetheless there.)



    March 14, 2007 at 3:09 am

  5. Much of sociology and org theory/behavior can be highly situational, and thus human nature (however it gets conceptualized, I am not necessarily arguing for self-interest or anything else specific, despite the Williamson reference) gets relegated to environmental and cultural factors, in other words, it becomes infinitely malleable, and thus, in some ways, drops out of the equation.

    What is instructive to note is where models/theories start (in econ and soc), and, by and large, sociological models inherently begin at the collective, cultural, situational etc level, while economic arguments at lower levels – not surprising. (Yes, I am painting with a broad, broad brush here, and certainly recognize a myriad of heterodox approaches where this is not the case.) The point of the post, of course, is that there can be learning both directions soc-econ-soc — and to make a related point, a good chunk of ‘economic sociology’ is still heavily sociological (institutional arguments are often now simply couched as econ sociology), and I think some reciprocal and increased influence would be very healthy (thats where questions of human nature, origins, the new-new etc come in, certainly borrowing from certain strands of psych as well).

    Jeremy – econ learning from psychology, sure, of course I agree. We perhaps need some follow-up posts with regard to psych-econ-soc and variants.



    March 14, 2007 at 3:37 am

  6. Theory will lie, lie, lie. Beautiful lies, but still lies. Actually, that is my biggest problem with economists – they typically believe in their pet theories and often confuse science with policy and social engineering.


    Dan Karreman

    March 14, 2007 at 4:35 am

  7. […] Following Fabio’s post about what economists can learn from sociologists, Teppo tells what sociologists can learn from economists; while we are on the topic, take a look at this post by Mark Thoma as to why economics is to be […]


  8. Over at Fabio’s post, there is a whole anti-mathematical modeling mob forming (pitchforks and torches and everything). So I ran to this post to actually defend mathematical modeling as certainly the (perhaps the only? perhaps the main?) thing that sociologists can learn from economic reasoning.

    My main piece of evidence that something like this can be useful for sociologists? The work of Jeremy’s (current) colleague James Montgomery, the rare (i.e. only) MIT trained economist who has sociology department written on his paycheck (does that diminish the utility of his salary you think?). Also probably (I think) the only person alive today that can claim to have published in AER, QJE, AJS and ASR (!); but I might be wrong on that one.

    In any case, James has developed formal mathematical models of job search and social networks (point #2 in Fabio’s post), and recently used other forms of non-quantitative mathematics (i.e. Boolean approaches) to model the sociological concept of Role Embeddedness, as well as making some recent incursions into market approaches to the behavior of religious organizations (and individuals).

    So formal mathematical methods are a tool that is not inherently tied to the use that economists have made of them so far, and that can be used to shed light on important sociological issues. Putting our arguments in formal terms can be useful (even if it is non-monotonic logic) since it brings clarity and sometimes unexpected logical and empirical implications, that can then be tested by messy and misleading statistical methods, using messy and misleading real-world data.



    March 14, 2007 at 12:21 pm

  9. I cover Web.20 as a tool for technology companies to communicate to the media and their communities on my blog Thank you for your ideas because they relate to the real world of economic activity. Here are some thoughts:

    1. We build brands that exploit imitation and conformity as a basic feature of human behavior. Repeat your messages in a meaningful way and just maybe you will impact individuals and communities to buy your products and services.

    2. Social networks are all about Web2.0. I’ll spare you the lecture on YouTube and MySpace.

    3. Talking to customers – getting their stories using a journalist’s ear is what PR at its best has always been about.



    Howard Oliver

    March 14, 2007 at 5:00 pm

  10. […] two posts with “quick, highly-stylized” on what economists should learn from sociologists, and what sociologists should learn from economic reasoning (via Marginal Revolution). From the first […]


Comments are closed.

%d bloggers like this: