An unpopular, critical take on Mokyr’s Econ Nobel Prize

His story sounds nice, but how sure can we be?

Joel Mokyr, who argues culture – not material factors – was crucially responsible for the appearance of the modern world, is one of the recipients of this year’s Econ Nobel! He got the prize “for having identified the prerequisites of sustained growth through technological progress” in his tomes on the economic history of early modern Europe.

Mokyr is a world-famous economic historian whose knowledge is vast and enviable. He is now also one of the few people with a Nobel Prize. I am quite literally a nobody, so don’t think that I think my critique matters in the grand scheme of things. Still, I do have some thoughts I’d like to share – some thoughts that might make us pause a bit when congratulating ourselves on how the European “culture of growth” and the “scientific ethos” ostensibly gave us modern development.

Now, a few things are undeniable. I get that his explanation for the emergence of post-Malthusian development is intuitively appealing, and I don’t mean that in a bad way. (It’s not intuitively appealing to me, because I don’t find cultural explanations convincing in many, perhaps even most, cases, but I get that others find his explanation intuitive and I think that’s fine.) Moreover, I don’t think I can deny that, at least in part and generically speaking, the Enlightenment, the scientific, experimental mode of reasoning, and optimistic and heroic ideas about growth and improvement had something to do with the emergence and/or continuation of modern economic development. At a certain level of abstraction, that proposition almost becomes tautologically true.

Is it even testable?

With that out of the way, here are the most important reasons why I’m not all that partial to Mokyr’s story.

First, I think his explanation is likely not empirically testable, at least not in the ostensibly rigorous way that is necessary in science.¹ Obviously, natural sciences have far more rigorous ways of testing than we do, but even in social science, Mokyr doesn’t have what rightly or wrongly (I think rightly) passes for rigor on our turf. And if that’s true, then we most likely have to remain silent with respect to whether his explanation is true or not. It might be, or it might not be, but we cannot know (if it’s not really testable).

I imagine people will think of this as vulgar positivism, but we should be honest about how epistemically credible various propositions are. It’s one thing to claim, based on an assortment of various descriptive facts, anecdotes, and theory, that the minimum wage causes an increase (or decrease) in unemployment. It’s much harder to actually, rigorously test whether it does so or not. Ditto for Mokyr.

However, in the minimum-wage case, we have scores of researchers using lots of fine-grained data and different quantitative methods for causal inference battling it out over the past few decades. In Mokyr’s case – not so much. Which is understandable, given the lack of data, historical distance, and the intangibility of his thesis.

If testable, what counts as evidence for or against?

Second, to the extent that his story is (or could be) testable, I’m afraid it’s so squishy – by construction – that we either cannot convincingly show whether it passes or doesn’t pass any particular test, or (again due to squishiness) we’ll probably be able to constantly move the goalposts when evidence does come in. Here, I have in mind Robert Allen’s work on the background of 79 early British inventors and capitalists. He found only weak corroboration of Mokyr’s thesis (or a testable implication of his overarching explanation). Around half of the people studied by Allen “had some connection” to the scientific community and the Enlightenment, but the connections are sometimes quite tenuous and very industry specific. As he puts it himself:

The importance, therefore, of Enlightenment links depends on which industries one thinks were central to the Industrial Revolution. Steam, of course, was important in the long run, and in that context Enlightenment links mattered. In the late eighteenth and early nineteenth centuries, however, textiles and metals had much greater economic impact. Using that yardstick, the Industrial Enlightenment did not matter much.

What does this tell us about the veracity of Mokyr’s explanation?² Not much, perhaps, because he can fall back on the claim that he just means science and the Enlightenment in a very general sense that permeated British society, and so it wouldn’t show up explicitly in individual inventors/capitalists. But is that fair? I don’t know. And that’s the problem.

Or to give another example, it’s likely that England broke out of the Malthusian trap and set itself on the path to modern, self-sustaining growth as early as 1650, which was around 200 years before anyone else in Europe did the same thing (with the exception of the Netherlands, which is a complicated case). Now, if there was (per Mokyr) this overarching scientific revolution, the Enlightenment, and a culture of growth that started permeating Europe (Germany, France, Britain, the Low Countries, Italy) after the Middle Ages, why did only England start developing? Why did it take a two-century lag for others to get going as well? Was the Enlightenment or a culture of growth not present in France and Germany?

It’s complicated and multicausal

His answer would be (and is) that, yes, the culture of growth and science were Europe-wide phenomena. Spillovers were large because of the non-localized presence of the Republic of Letters, say. However, he would add that everything only came together in Britain because of other factors, such as institutions (Britain had the right ones, France not as much) and various political forces.

Now, this makes some sense and might be true. But then again: I hope people reading and praising Mokyr are careful enough not to take his Enlightenment/culture of growth thesis as the single or preeminent cause of modern growth. Some people who are enamored with the thesis tend to cite the culture of growth as either the only or the most important reason. Only when you then press them with the facts above do they say, “Okay, yeah, sure, there are other very important reasons as well.”

Additionally, if Mokyr’s explanation is only a partial explanation – or a necessary but not sufficient cause – then we have to somehow explain those other factors that help in his story. That is, why did Britain have proper institutions, while France lacked them ca. 1600–1700, even though both were enlightened?

What caused that? Again, we probably can’t really know, at least not if we insist on a high evidentiary standard.

So, should we nevertheless talk about these matters? I guess we should, because it’s fun, and sometimes contact is made with reality. Though I’m not sure we should be talking about non-quantitative historical explanations in the context of relatively firm, established (social) science. We should recognize that, because it’s so hard to know, these topics remain inherently speculative and unresolved if looked into at a high enough resolution.

1. The paper by Kelly et al. (2023) is a potential exception to my claim.

2. See also the paper by Kelly and Ó Gráda (2022) as a possible counter.