Why Economic History Needs a Reformation

Time to Nail My Theses to the Church Door

The Poor Rich World blog is about to get a bit medieval. I will be posting more frequently and my posts will be focused on a critical engagement with what passes for economic history in the “top five” economics journals today. My argument, in short, is that a Reformation is needed in my discipline.

The problem is threefold.

First, the use of what D. C. M. Platt described as “Mickey Mouse numbers” has escalated due to the “credibility revolution” in economics, which has increased the demand for quantitative data. But reliable historical statistics generally do not exist. As a result, what can euphemistically be called “synthetic data” have proliferated in what would have been Platt’s nightmare.

Second, the econometric results produced using those data are extremely fragile. Timothy G. Conley and Morgan Kelly’s analysis of spatial confounding is an important example of the endemic lack of robustness: in studies that claim to find deep correlations between past and present, standard errors that ignore spatial autocorrelation can make even random spatial structure look statistically significant. Nonetheless, such “persistence” studies have become increasingly dominant in the profession.

Third, the narratives that are told in those articles are often highly simplistic due to the need to identify a clear causal effect. Good institutions make countries rich; legacies of the slave trade make Africa poor today; ancient ethnic hatreds explain the Holocaust; and so on. In reality, however, things are rarely as simple as “x causes y.”

Consequently, much of what passes for economic history today in the leading economics journals should not be taken seriously, even as it has become what economic historians who wish to write on historical issues must aspire to if they want to make it in the economics profession.

For my part, however, I really care about my discipline. I think that students should not be encouraged to follow the modalities of what is currently published in the “top five.” We should rebel, even if it means that economic historians become refugees, dependent on the kindness of business schools and history departments. They may end exiled to the wilderness or perhaps to a hill in Wales. Defending the discipline of economic history will not be easy. But, in my opinion, it must be done.

And that is why I will be posting more regularly on The Poor Rich World blog. Sellers of indulgences have been warned.

Further Reading

For an account of the decline of economic history in a major American university, see:

Peter Temin, “The Rise and Fall of Economic History at MIT,” History of Political Economy 46, annual supplement (2014): 337–350.

The rise of “persistence studies” has been a morbid symptom of that decline. See:

Martina Cioni, Giovanni Federico, and Michelangelo Vasta, “The Two Revolutions in Economic History,” in The Handbook of Historical Economics, ed. Alberto Bisin and Giovanni Federico (London: Academic Press, 2021), 17–40; “Persistence Studies: A New Kind of Economic History?” Review of Regional Research 42, no. 3 (December 2022): 227–248; and “Is Economic History Changing Its Nature? Evidence from Top Journals,” Cliometrica 17, no. 1 (January 2023): 23–48.

A critical discussion is found in:

Leticia Arroyo Abad and Noel Maurer, “History Never Really Says Goodbye: A Critical Review of the Persistence Literature,” Journal of Historical Political Economy 1, no. 1 (June 2021): 31–68.

The extent of spatial confounding is demonstrated in:

Timothy G. Conley and Morgan Kelly, “The Standard Errors of Persistence,” Journal of International Economics 153 (January 2025): 104027.

For a response to an earlier version of Kelly’s critique, see:

Hans-Joachim Voth, “Persistence – Myth and Mystery,” in Handbook of Historical Economics, ed. Alberto Bisin and Giovanni Federico (London: Academic Press, 2021), 243–267.

Notably, Voth presents results that seem to refute Kelly and claims that “code and data for regressions and simulations in this essay are freely available at http://www.jvoth.com/data/replication/handbook and at the ICSPR [sic] data repository.” Nonetheless, that link never seems to have existed and it has not been possible to locate a relevant data repository on ICPSR.

For an early warning of the rise of “synthetic data,” see:

D. C. M. Platt, Mickey Mouse Numbers in World History: The Short View (Basingstoke: Macmillan Press, 1989).

For a detailed recent example:

Joseph Francis, “Research Integrity Issues in Voigtländer and Voth (2012),” isitcredible.com, 2026.

And for case studies of some of the broader problems in economics:

Joseph Francis, “A Replication of ‘Comparative Politics and the Synthetic Control Method’ by Abadie et al. (2015),” I4R Discussion Paper Series No. 271, 2025.

Joseph Francis, “The ‘China Shock’ and the Rhetoric of Economics: A Comment on Autor, Dorn, and Hanson (2013),” 2025.

Joseph Francis, “Specification Searching in the Race between Education and Technology,” Econ Journal Watch, 23, no. 1 (March 2026): 4–26.