Readings: February 16, 2020

National Accounting, Counterfactuals, and Imperial Collapse

What follows is a collection of summaries of provocative or interesting articles, blog posts, and academic papers that I read in the last week. The criterion of selection is how much I learned from each, not overall agreement, so I may argue vehemently with the authors. I have also provided a one-sentence summary for each summary to combat my prolixity.

The Little Divergence

by Pseudoerasmus | pseudoerasmus | 12 July 2014

An outline of two methods for long-run GDP calculation and their implications for the history of economic growth in Europe.

Here, Pseudoerasmus reviews the differences between the Malthusian and “modern” estimates of pre-industrial European productivity and wage growth. Of particular interest is his comparison of the time-series constructed by Stephen Broadberry (2010) and Gregory Clark (2010a) for English real wages, and subsequently for GDP per capita. The former finds a secular increase in real wages and GDP per capita beginning as early as the thirteenth century, while the latter—a proponent of the aforementioned Malthusian model, in which birth patterns channel all resources (and technological advances) into population increases while wages stagnate—charts trendless fluctuations that leave England as wealthy in 1350 as in 1750. Pseudoerasmus quotes Broadberry’s empirical objection to Clark’s pessimism:

“…there are good reasons to be sceptical about this interpretation of long run economic history [based on wage data], which seems to fly in the face of other evidence of rising living standards, including the growing diversity of diets (Feinstein, 1995; Woolgar, Serjeantson and Waldron, 2006), the availability of new and cheap consumer goods (Hersh and Voth, 2009), the growing wealth of testators (Overton, Whittle, Dean and Haan, 2004; de Vries, 1994), the virtual elimination of famines (Campbell and Ó Gráda, 2011), the growth of publicly funded welfare provision (Slack, 1990), increasing literacy (Houstan, 1982; Schofield, 1973), the growing diversity of occupations (Goose and Evans, 2000), the growth of urbanization and the transformation of the built environment (de Vries, 1984).”

That is, if the wage data depict unremitting stagnation, they must belie the fundamental changes documented by most standard histories of the early modern period. An outline of Broadberry’s “heroic” method for constructing GDP from the output side follows, before a rough conclusion is reached—namely, that the difference between the two series results from varied interpretations of days worked.

Normally, you compare the GDP estimates calculated with different methods, but in this case, Clark’s and Broadberry’s are very different, especially for the late Middle Ages. Where, precisely, do they differ? That is, what statistical adjustment is necessary to harmonise Broadberry’s and Clark’s estimates ? The number of days worked! In Clark’s data, the number of days worked per worker per year stays within the range of 250-280 days over the course of 550 years… What Broadberry does is impute the days worked from his output estimates. This means, he reconciles his output-based GDP with the wage-based GDP by increasing or decreasing the days worked as necessary to fit his own GDP data.

Broadberry, following Jan de Vries, posits that an “industrious revolution” (whereby workers supplied more labor) caused annual incomes to increase while day wages stagnated. By working more days, English laborers could purchase the goods and services (as well as drive productivity expansion) described above. As Pseudoerasmus notes, however, this interpretation does not rule out the Malthusian model; efficiency may remain low and stationary. The implication of all this for the little-mentioned “Little Divergence,” in any case, is in the historical depth of Northwest European growth—whether it resulted from long-term structural factors or a sudden conjuncture.

Land Tenure and Exploitation from the Roman Empire Onward: Economic Growth in Historical Perspective

by Brad Delong | Brad DeLong’s Grasping Reality | 13 February 2020

The collapse of Roman political order led to the end of legitimate landholding in Western Europe and the rise of the aristocratically-defended fief.

The last entry was longer than intended, so this one is shorter. I’m finding my range. Here, Brad Delong offers up a passage from Peter Temin’s wonderful The Roman Market Economy on the consequences of Roman decline for land use in Western Europe. As enforcement of land-use regulations and public registration (through land taxes) dwindled with local Roman authority, violence threatened all agricultural property owners, who could only plant at great risk and expense. “Title” was determined by the ability of an incumbent to defend his fields, leading to the rise of the fief, the medieval tract ruled by a lord, managed by vassals, and farmed by peasants. An extractive division cleft apart fighting and field men, with the former seizing as much as 40% of the latter’s production, until the advent of the bowman began the obsoletion of the knight that the musket would complete. Temin perhaps exaggerates the dangers of raiding and barbarism in the immediate aftermath of Rome’s descent, and probably underestimates the role of Carolingian central power in establishing the feudal system, but the excerpt—as with the rest of his book—is otherwise sharp and sensible. Given the dependence of investment and economic activity on political order in the classical world, the disintegration of that order goes some way toward explaining the end of the imperial efflorescence. Byzantium, where land taxes continued to buttress state capacity, provides a ready and supportive counterfactual.

‘What if’ is a waste of time

by Richard J. Evans | The Guardian | 13 March 2014

Counterfactual arguments, common among popular historians, are speculative, ideological, and generally poor reasoning.

Speaking of counterfactuals, eminent twentieth-century historian—and learned historiographer—Richard J. Evans declares that such thinking (see the title) is unhelpful, leading one to speculative and unprovable conclusions. Moreover, counterfactual reasoning is but one step (if that) removed from “great man” or “kings and battles” narratology, anachronistic strains of a discipline long since turned toward sociological and anthropological work. I agree insofar as the essay applies to popular writers like Max Hastings (spare a childhood hero, please!) and Niall Ferguson—though soft, easily-caricatured targets they may be—but would remind Evans that counterfactuals remain powerful analytical tools in empirical social science. The standard hypothesis test, after all, sets up a counterfactual by which to judge the plausibility of collected data. Even alternate histories can be employed rigorously, if only to isolate and gauge the effects of an institution or policy decision when geographical comparisons do not present themselves. As Yale’s John Lewis Gaddis writes in The Landscape of History (with a sympathetic allusion to Ferguson):

[Marc] Bloch argued that we should seek “the antecedent which could have been most easily avoided.” We do that, he explained, by a “bold exercise of the mind” in which historians transport themselves “to the time before the event itself, in order to gauge its chances, as they appeared upon the eve of its realization.” We move the present back into the past so that it becomes, as he put it, “a future of bygone times.”

What Bloch was suggesting here, I believe, was nothing less than the historical equivalent of laboratory experimentation in the physical sciences: using their imagination, historians were to perform procedures similar to what chemists and physicists do with their test tubes, centrifuges, and cloud chambers. They would revisit the past, varying conditions as they did so to try to see which would produce different results. They would do this by means of counterfactuals.

I’m not quite so optimistic about the procedure in general, but I do believe both that most historians practice this and that it can (and must) be done effectively.

The Decline of Spain

by Earl J. Hamilton | The Economic History Review | May 1938

A classic economic and political explanation for the decline of Spain from world-power status during the early seventeenth century.

Spain was, by the end of the sixteenth century, seemingly set for global domination—first mover in the colonial game, ruler of Italy, India, and Peru, and sovereign of Catholic Christendom. Hamilton, once a distinguished economic historian at the University of Chicago, explains how this dissipated in less than a century under the pressures of poor rulership, excessive borrowing, and exorbitant taxation. In so doing, he undermines classical political interpretations centered on successive defeats in battle against England and France. Faced with a bloated clergy, oppressive taxes, an over-grazing shepherd guild, and rampant inflation from New World silver, Spain’s once-prosperous manufacturing sectors crumbled and population declined. Agriculture dwindled and famines ensued; cities were referred to as “deserts,” and investigations launched into the puzzle of their poverty. These efforts, however, were insufficient to stem the tide of woes, and Spanish production dwindled to nothing.