“Stylized Facts in the Social Sciences”

Sociologist Daniel Hirschman writes:

Stylized facts are empirical regularities in search of theoretical, causal explanations. Stylized facts are both positive claims (about what is in the world) and normative claims (about what merits scholarly attention). Much of canonical social science research can be usefully characterized as the production or contestation of stylized facts. Beyond their value as grist for the theoretical mill of social scientists, stylized facts also travel directly into the political arena. Drawing on three recent examples, I show how stylized facts can interact with existing folk causal theories to reconstitute political debates and how tensions in the operationalization of folk concepts drive contention around stylized fact claims.

Interesting. I heard the term “stylized facts” many years ago in conversations with political scientists—but from Hirschman’s article, I learned that the expression is most commonly used in economics, and it was originally used in a 1961 article by macroeconomist Nicholas Kaldor, who wrote:

Since facts, as recorded by statisticians, are always subject to numerous snags and qualifications, and for that reason are incapable of being accurately summarized, the theorist, in my view, should be free to start off with a ‘stylized’ view of the facts—i.e. concentrate on broad tendencies, ignoring individual detail, and proceed on the ‘as if’ method, i.e. construct a hypothesis that could account for these ‘stylized’ facts, without necessarily committing himself on the historical accuracy, or sufficiency, of the facts or tendencies thus summarized.

Hirschman writes:

“Stylized fact” is a term in widespread use in economics and is increasingly used in other social sciences as well. Thus, in some important sense, this article is an attempt to theorize a “folk” concept, with the relevant folk being social scientists themselves. . . . I argue that stylized facts should be understood as simple empirical regularities in need of explanation.

To me, this seems close, but not quite right. I agree with everything about this paragraph except for the last four words. A stylized fact can get explained but I think it remains a stylized fact, even though it is no longer in need of explanation. I’d say that, in social science jargon, a stylized fact in need of explanation is called a “puzzle.” Once the puzzle is figured out, it’s still a stylized fact.

But that’s just my impression. As Hirschman says, a term is defined by its use, and maybe the mainstream use of “stylized fact” is actually restricted to what I would call a puzzle or an unexplained stylized fact.

Why ask, “Why ask why?”?

In any case, beyond being a careful treatment of an interesting topic, Hirschman’s discussion interests me because it connects to a concern that Guido Imbens and I raised a few years ago regarding the following problem that we characterize as being typical of a lot of scientific reasoning:

Some anomaly is observed and it needs to be explained. The resolution of the anomaly may be an entirely new paradigm (Kuhn, 1970) or a reformulation of the existing state of knowledge (Lakatos, 1978). . . . We argue that a question such as “Why are there so many cancers in this place?” can be viewed not directly as a question of causal inference, but rather in- directly as an identification of a problem with an existing statistical model, motivating the development of more sophisticated statistical models that can directly address causation in terms of counterfactuals and potential outcomes.

In short, we say that science often proceeds by identifying stylized facts, which, when they cause us to ask “Why?”, represent anomalies that motivate further study. But in our article, Guido and I didn’t mention the term “stylized fact.” We situated our ideas within statistics, econometrics, and the philosophy of science. Hirschman takes this all a step further by connecting it to the practice of social science.

18 thoughts on ““Stylized Facts in the Social Sciences”

  1. Not that it is any of my business, but if Dr. Hirshman gave his article to my freshman English grader, it would come back with red pencil marks all over it. (As my first essay did.) For instance, crossing out the useless phrase “in some important sense”.

    • I don’t know what field you work in. But in the health care literature, finding any article that would pass muster with a decent English teacher would be searching for a needle in a haystack. Academic culture promotes a verbose, hyper-periphrastic form of writing.

      • “hyper-periphrastic”

        Wow! Tha’s a great word! :) I love it!

        Or in the medical research industry, it could be a condition or an affliction! Hyperperiphrasiosis? Poorly understood, little recognized but widespread, there currently is no known treatment or therapy, although victims of this condition seem to thrive from its effects. :0

    • JimV should post samples of their Pulitzer level work here. As it is often said – “there’s no accounting for taste”.

      I personally like the tone and style of Mr. Hirschman’s prose. It seems inviting and conversational to me. I remember watching a writing tutorial from Kurt Vonnegut where he stressed the importance of sounding like your genuine self rather than “putting on airs” in an attempt to sound like the Queen of England or some other such haughty noble.

      • Thanks for the classic ad hominem argument (also covered in a good freshman course). The tone was okay, the style was how many people come out of high school, and is quickly cured by a freshman English course in college. The key is to write it however it comes out, then go back and cross out unnecessary stuff and consider whether what you have written is the best way of explaining what you actually mean.

        I apologize for misspelling Dr. Hirschman’s name. I should have checked my work better.

        • Ad hominem? Sorry – I don’t think so (but then again, my Latin is rusty).

          Judging the quality of prose is inherently subjective. I would say my call for examples of how it “should be” from an expert are no more “ad hominem” than the following…

          “…if Dr. Hirshman gave his article to my freshman English grader, it would come back with red pencil marks all over it.”

  2. Does the notion of a ‘stylized fact’ in economics and sociology have a statistical analogue in ‘heuristic’? If so and as has been commonly observed, today’s heuristic is tomorrow’s convention or commonsense rule of thumb, i.e., lacking any theoretical foundation. One famous statistical example of this is Fisher’s p<.05 level of significance, a rule of thumb with no theoretical basis.

    • Ian:

      Interesting thought. A stylized fact and a heuristic are different, but I see the connection . . .

      How’s this: for a heuristic H, the corresponding stylized fact is “H works” (whatever that means). For example:

      – “Report a result when p is less than 0.05” is a heuristic.

      – “The ‘p less than 0.05’ heuristic is standard in many areas of science” is a stylized fact.

      – “The ‘p less than 0.05’ heuristic leads to major problems with replicability” is another stylized fact.

      – “Why does ‘p less than 0.05’ remain standard practice, given that it leads to major problems with replicability?” is a puzzle.

      I’m not saying this puzzle is unresolvable—indeed, many answers have been proposed, such as “incentives” or “inertia” or “it’s easy to do” or “actually, the ‘p less than 0.05’ rule is not so bad” . . . My point here is just that this is a puzzle that needs to be resolved somehow.

    • An example of a stylized fact in economics is “industrial production collapsed during the Great Depression.” That’s not a heuristic, it’s a quick summary of statistics that are used to represent the real world.

      You could however combine that stylized fact with others and use them to create an economic model explaining the cause of it. That would give you a heuristic that you could then use to make predictions.

  3. The problem with stylized facts is that sometimes they are true and sometimes they are not. A good publishing tactic in economics is to find a plausible theoretical reason why a stylized fact is true – or to find that the stylized fact is really being “fooled by randomness.” Take the cancer cluster as an example – it could lead to finding a source for a particular cancer cluster, or it could be that the cluster is really, in fact, a random occurrence. So, I guess stylized facts are fertile ground for further investigation – either to confirm (consistent evidence) or to cast doubt (inconsistent evidence). Perhaps they should be called “stylized speculations.” “Facts” seems to presume the answer.

    • Yeah, I’m not so sure I like the idea of using the word “facts” (stylized or no) for what is typically found with statistical methods either. More like “stylized betting odds”.

      Snce most of what stats are doing is quantifying uncertainty, it feels like most “facts” discovered in the social sciences with statistics are just ways to tip the odds in your favor a bit. Maybe stylized speculation is a bit better haha

  4. Some dim memory tells me the concept, perhaps under a different name, goes back to Max Weber, but this could be wrong. Anyway, I think it captures two important, well, stylized facts. First, we can often identify broad empirical patterns that encompass variation across individual cases. For some purposes that variation may not be very important. Second, our theoretical apparatus is often insufficient (or too misguided) to predict or explain regularities we can recognize without them. Neither requires the additional notion of “needs to be explained”, although that’s a logical result of each factor: (a) Maybe the variation within the regularity is important after all, and we need the sort of explanation that can account for it. (b) Maybe the theoretical shortcomings are severe enough they need to be addressed.

    A famous example of a stylized fact in economics is Keynes’ observation about the relationship between income and consumption in The General Theory. If I recall correctly, he didn’t worry much about why but simply used it as a building block of his model of the determination of national income. Later economists took the absence of explanation as a challenge and constructed models in which Keynes’ stylized fact would arise. There has been a justified backlash against macroeconomic approaches that demand rigid adherence to one or another of these models of consumption behavior, even though their evidentiary base is weak.

  5. I echo Andrew’s discomfort with the last four words of the paragraph. Perhaps an addendum to his explanation:

    A stylized fact does not require a theoretical explanation.
    A necessary condition for a social science theory is consistency with at least one stylized fact.
    A puzzle is two or more stylized facts in need of a coherent theoretical explanation.

  6. “A stylized fact can get explained but I think it remains a stylized fact, even though it is no longer in need of explanation. ”

    I would argue, in the social science, there can be multiple explanations; hence it would still be in “need” of a single (or unified) explanation. Any ambitious social scientist would like to challenge any existing explanation, and hence, it would always be a stylized fact.

    I would argue that a “puzzle”, is a fact that has no or inadequate explanations. the definition of “inadequate” can be debatable, hence the difference is quite fuzzy.

  7. Thank you all for these stimulating exchanges. I would like to add a couple of things, if I may.

    First, what would a non-stylized fact be – or a collection of them? What is the boundary, or criterium defining it, between a stylized and non-stylized fact? Is the difference between the two determined by a pre-existing theory, no matter how wrong or arbitrary it might be? And how would the answer fit with the rest of the discussion?

    Second, in an article by Natalie Wolchover published in Quanta Magazine on Nov. 20, 2020, we learn that there are currently seven (7) competing theories, or rather interpretations, of what a quantum particle might be, or even if anything like it actually “exists”. And similar to what one of the earlier comments says, there is an effort among physicists to unify these interpretations and/or weed out those that should turn out to be (experimentally) wrong. A third tack is to prove that two or more of them are equivalent, thereby reducing the size of the “puzzle”, somewhat like the initially competing matrix (Heisenberg) vs. wavefunction (Schrödinger) representations were shown to be, thereby unifying them and “solving the puzzle”. Is that any different from what is done in the social/behavioral sciences?
    (One difference might be the ease or difficulty with which observations or experiments provide answers with sufficient confidence. Even in physics there is always a trade-off between confidence and accuracy: one can be very confident in an approximate answer but much less confident in a more accurate one – the the trade-off can be softened if one throws billions of dollars at it like at the CERN! Of course, total accuracy is beyond reach of any physical measurement.)

    In some sense, these two questions above are related, or represent two sides of the same coin.
    It also relates to another remark, that “stylized facts”, however defined, are dynamic: when one is “explained” (given some criteria of explanation), rather than disappearing it usually morphs into, or generates, one or several new “stylized facts”. It would be interesting to sleuth the trajectories of a couple of them throughout the history of economic analysis, for example.

    I’d also be interested to hear what others think about this.
    Thanks again for the stimulation!

Leave a Reply to rsm Cancel reply

Your email address will not be published. Required fields are marked *