Timothy Guinnane, William A. Sundstrom, Gavin Wright
2 Mar 2023
Paul David, who passed away in January 2023, was a major figure in redefining economic history; an early practitioner of cliometrics – exploring historical economies through rigorously specified models and quantitative evidence; and a strong advocate of the view that historical research should be fundamental to the economics discipline – that ‘history matters’. This column, written by a Stanford colleague and two former students, outlines some of his many contributions to our understanding of economic progress, not least the diffusion of new technologies. His account of the persistence of the QWERTY keyboard despite its technical disadvantages is one of the most cited articles in all of economics.
Paul Allan David died on 23 January 2023, at his home in Palo Alto, at the age of 87, after a long struggle with Alzheimer’s disease.
Paul was a major figure in redefining economic history, though his research interests and influence ultimately ranged far beyond his original specialties. Influenced by his father Henry David, a distinguished historian of the American labour movement, Paul studied both parent disciplines even as an undergraduate at Harvard (1952-56). Thus he became an early practitioner of what was then known as the New Economic History, now more often called cliometrics – understanding historical economies through rigorously specified models and quantitative evidence.
But Paul possessed far more historical sensibility than most cliometricians, cultivated during his postgraduate studies at the University of Cambridge (1956-58). Returning to Harvard, Paul worked with Alexander Gerschenkron, pursuing an ambitious thesis project on the economic history of Chicago. Harvard awarded him a PhD in 1973, not because he had submitted the dissertation, but because by then he had compiled such an extensive publication record that they felt he had earned the degree!
Paul came to Stanford in 1961, maintaining that affiliation throughout his career. Even without a PhD, he advanced to associate professor and then full professor of economics. Degree in hand, he served as department chair from 1979 to 1983.
Perhaps Paul’s deepest legacy for Stanford was establishing the school as a leading centre for economic history. The joint Berkeley-Stanford Colloquium, organised by Paul and Al Fishlow in the 1960s, was a focal point for Bay Area economic history for decades.
At Stanford, the patriarch was Moe Abramovitz, who remained active in both research and workshops long after his retirement in 1979. But it was Paul who recruited or helped to recruit Nate Rosenberg, Gavin Wright, Avner Greif and Ran Abramitzky – to name only those who became full professors of economics.
We claim that there are more active economic historians from Stanford than from any other programme. A much larger number of graduate students did not become history specialists but were exposed to Paul’s perspective through his teaching or that of his colleagues. We like to think that some of this historical thinking rubbed off and became something of a hallmark for a Stanford economics PhD.
Always an economic historian, Paul soon extended his horizons in diverse and seemingly disparate directions. He became a strong advocate of the view that historical research should be fundamental to the economics discipline; in brief, ‘history matters’. The essence of the argument was captured by Paul’s incisive account of the persistence of the QWERTY typewriter keyboard despite its technical disadvantages, one of the most cited articles in all of economics (David 1985).
History Matters is the title of a festschrift presented by a group of Paul’s former students in 2004, in which the editors write: “No scholar has more forcefully and influentially argued the case for making economics a truly historical social science – one that, like evolutionary biology, gives past events a central role in understanding the present” (Guinnane et al. 2003).
A continuing theme throughout Paul’s career was the diffusion of new technologies. An important early paper considered the adoption of the mechanical reaper in the American Midwest (David 1966). Invention occurred in the 1830s, yet the first wave of adoption occurred only in the 1850s. The 20-year delay, according to Paul, was explained by the fact that a minimum scale was required to cover the fixed costs of purchasing the reaper. Only when farm size passed this ‘threshold’ did mechanisation make economic sense.
Specialists have debated the specifics ever since, but the basic form of Paul’s diffusion model has been highly influential. In many respects, it formalised accounts of delayed diffusion presented by Nate Rosenberg, and thus became something of a ‘Stanford school’ of thought in this area.
Scrolling forward to 1990, the era of the Solow paradox (‘You can the computer age everywhere but in the productivity statistics’), Paul offered an analogy between the delayed productivity effects of computer technology and a similar lag in the impact of electrification between the 1880s and the 1920s. With the IT-driven productivity surge of the late 1990s, this article also attained iconic status (David 1990).
This line of inquiry led Paul into an extensive engagement with policy issues related to science and technology, such as the role of interoperability standards in the evolution of network industries. He was particularly active in Europe. Although Paul maintained his Stanford affiliation throughout his lifetime, beginning in 1993 he spent about half his time at All Souls College, moving to the Oxford Internet Institute from 2002 to 2008.
From his Oxford base, Paul was a frequent participant in European networks dealing with the promotion and regulation of information technologies. His European colleagues presented Paul with a second festschrift in 2006 (Antonelli et al. 2006).
Another of Paul’s enduring interests was demography, specifically the onset of fertility control in the 19th and early 20th centuries. Often working with his former graduate student Warren Sanderson, Paul pioneered techniques to measure the extent of fertility control from a type of data several countries collected in the early twentieth century (David and Sanderson 1986). This ‘cohort parity analysis’ led to new insights into early fertility transitions in the US, Ireland and Great Britain (David et al. 1988).
Paul also deployed his under-used talent as an economic theorist to explicate better how rational couples try to reduce completed family size when the only contraceptive methods available are costly and uncertain. His thinking on this issue induced an array of historically minded social scientists to rethink the evidence for early fertility transitions. His work in demography illustrates Paul’s belief in taking seriously the methods and ideas of other disciplines, as well as the considerable influence of his wife and occasional collaborator, Sheila Ryan Johansson (David et al 2010).
These specialties would constitute an ample intellectual legacy for a normal scholar, but for Paul, they are only one component of a much larger total. He also made significant contributions in macroeconomics and growth accounting, natural resources, slavery, migration and climate change, among other subject areas.
Paul’s academic honours are too numerous to be listed in full, but they include: Fellow of the Econometric Society, the American Academy of Arts and Sciences, and the British Academy; Pitt Professor of American History and Institutions at Cambridge; President of the Economic History Association; and President of the Western Economic Association International.
Paul married Janet M Williamson in 1958. They later divorced. He married Sheila Ryan Johansson in 1982. In addition to Sheila, Paul is survived by children: Rachel, Matthew, Elizabeth and Kenneth; and five grandchildren.
Editors’ note: An earlier version of this column appeared at the Economic History Association’s EH.net.
Antonelli, C, D Foray, B H Hall and W E Steinmueller (2006), New Frontiers in the Economics of Innovation and New Technology, Edward Elgar Publishing.
David, P A (1966), “The Mechanization of Reaping in the Ante-Bellum Midwest”, in Henry Rosovsky (ed.), Industrialization in Two Systems: Essays in Honor of Alexander Gerschenkron, John Wiley and Sons: 3-39.
David, P A (1985), “Clio and the Economics of QWERTY”, American Economic Review 75(2): 332-37.
David, P A (1990), “The Dynamo and the Computer: An Historical Perspective on the Modern Productivity Paradox”, American Economic Review (2): 355-61.
David, P A, T Mroz, W Sanderson, K Wachter and D Weir (1988), “Cohort parity analysis: Statistical estimates of the extent of fertility control”, Demography 25(2): 163-88.
David, P A, S Ryan Johansson and A Pozzi (2010), “The Demography of an Early Mortality Transition: Life Expectancy, Survival and Mortality Rates for Britain’s Royals, 1500-1799”, Oxford Economic and Social History Working Papers 083.
David, P A, and W Sanderson (1986), “Rudimentary Contraceptive Methods and the American Transition to Marital Fertility Control, 1855-1915”, in Long-Term Factors in American Economic Growth: 307-90.
Guinnane, T, W Sundstrom and W Whatley (eds) (2003), History Matters: Essays on Economic Growth, Technology, and Demographic Change, Stanford University Press.