Quantitative Economic History

Joshua L. Rosenbloom, Joshua L. Rosenbloom

• Publisher: Routledge
• Year: 2008
• ISBN:
  9781135977849
• Edition: 1
• Topic: Commerce
• Subtopic: Commerce Général

The essays in this book use the analytical tools and theoretical framework of economics to interpret quantitative historical evidence, offering new ways to approach historical issues and suggesting entirely new types of evidence outside conventional archives. Rosenbloom has gathered together seven essays from leading quantitative economic historian

1
Editor’s introduction

The good of counting

Joshua L. Rosenbloom

BOSWELL: Sir Alexander Dick tells me, that he remembers having a thousand people in a year dine at his house: that is, reckoning each person as one, each time that he dined there.

JOHNSON: That, Sir is about three a day.

BOSWELL: How your statement lessens the idea.

JOHNSON: That, Sir is the good of counting. It brings every thing to a certainty, which before floated in the mind indefinitely.

BOSWELL: But Omne ignotum pro magnifico est: one is sorry to have this diminished.

JOHNSON: Sir, you should not allow yourself to be delighted with error.

(Boswell’s Life of Johnson (Oxford University Press ed.: Oxford 1934): IV, p. 204 (18 April 1783, Aetat. 74))

One of the principal achievements of the cliometric revolution is the approach to quantitative economic history that it spawned (McCloskey 1978). Economic historians have, of course, always relied upon quantitative evidence, but beginning in the 1950s cliometricians began to analyze quantitative data in new ways. Most importantly they explicitly acknowledged the connection between economic theory and quantitative evidence. Theory not only provided guidance about what to measure, but also how to measure it, and in turn, then, suggested entirely new types of evidence that had not previously been subjected to careful analysis. Even now, after more than half a century, the field of quantitative economic history continues to yield new insights about the past.

One important contribution of quantitative economic history has been the extension and refinement of measures of national income for periods before the formation of modern government statistical agencies. The reconstruction of these historical time series describing the pace and pattern of economic growth provides an essential foundation and consistency check for narrative accounts and more narrowly focused histories of particular sectors or industries. Not only does theory provide a framework for assembling readily available statistical information about the past, but it has proved essential in working out methods to bridge gaps in the historical record through creative reconstruction.¹

In addition to providing a basis for reconstructing the quantitative dimensions of economic growth the Cliometric approach has made accessible new kinds of historical evidence that were simply beyond the scope of traditional historians’ consideration. For example, scholars considering the role President Andrew Jackson’s “War” on the Second Bank of the United States played in the depression of 1837 had long relied on the arguments of various knowledgeable contemporaries to support their competing interpretations. By turning to data on changes in the country’s stock of monetary gold, however, quantitative economic historians were able to cut through these arguments and convincingly resolve the debate by showing that the causes of the depression arose largely from changes in international markets rather than any action of the President (Temin 1969; Rockoff 1971). In much the same, way quantitative economic historians mobilized a vast array of new data to re-examine debates about the economics of slavery. Despite the controversy that some of their work initially elicited, the profitability and economic rationality of the slave system was conclusively established through the use of a wide array of statistical sources that had not previously been examined.²

The point is not that quantitative economic history is in some sense better than other approaches to the past. Rather, by mobilizing the methods of the social sciences, quantitative economic historians have opened up new ways of looking at the past that complement and extend other more traditional approaches. By focusing on collective behavior and developing techniques to describe both central tendencies and variations around them quantitative economic historians are able to pose and answer questions that are not readily accessible to more traditional historians (Fogel 1983:29). While traditional historical approaches focus on the particular, quantitative economic history provides a way to set individual experience in a broader context, to assess the representativeness of individual experience and to explore systematic patterns of variation in data that cannot be discovered from case studies alone.

New ways of looking at the past have also encouraged the collection of new types of quantitative evidence. Since the 1950s quantitative economic historians have begun to systematically collect and analyze a variety of data sources outside the conventional archives of the historian. Beginning with the work of William Parker and Robert Gallman, who collected data from a sample of farms across the Cotton South in the 1860 census of population and linked these to data drawn from the census of agriculture, quantitative economic historians have collected and made available computer readable samples from a vast array of manuscript censuses.³ The Integrated Public Use Microdata Series based at the University of Minnesota now makes available to scholars large, nationally representative samples of every extant U.S. population census since 1850, as well as a growing collection of international data.⁴ Samples of a number of nineteenth-century manufacturing censuses are also now available. The initial work on these sources was undertaken by Fred Bateman and Thomas Weiss (1981), who collected samples from the manufacturing censuses of 1850, 1860, and 1870. This work has been continued and extended by Jeremy Atack, who has added samples from 1880.⁵

Like the field of quantitative economic history, the essays in this volume are unified by a common set of methods rather than by a particular chronological or topical focus. Many of them take advantage of large data samples to examine aspects of the past. Several describe insights derived from efforts to collect new sources of data, or to combine existing data sources to yield new insights. What all of them illustrate are the ways in which these methods continue to offer new insights about the past.

This volume begins with two essays addressing the intersection between economic and demographic history, one of the areas in which quantitative approaches have made great contributions to understanding historical forces that cannot be discerned from conventional documentary sources, as Weiss’s work on U.S. mortality rates indicates (Haines et al. 2003). In Chapter 2, John Ermisch examines how the economic environment and technology affected childbearing outside marriage in England and Wales since 1845. In it he finds that, up to World War I, higher unemployment rates discouraged marriage and increased non-marital birth rates, which is consistent with poorer labor market conditions discouraging marriages among pregnant would-be brides, thereby increasing bastardy. During the inter-war period, higher unemployment continued to produce postponement of marriages, but non-marital births were no longer linked to unemployment. Free access by unmarried women to the contraceptive pill after 1975 and the legalisation of abortion dramatically altered the environment for decisions concerning marriage and childbearing, causing the postponement of marriage and the rise in cohabiting unions. The consequent rise in the unmarried population accounts for most of the explosion in the proportion of births outside marriage, reaching 42 percent of births in 2004. Ermish argues that non-marital childbearing had been discouraged by strong social stigma against it, but a temporary change in the determinants of non-marital childbearing that raised it, like the large rise in unemployment in the early 1980s, produced rapid erosion of the stigma and a self-reinforcing rise in non-marital births.

In Chapter 3 Louis Cain and Elyce Rotella focus on deaths rather than births, exploring the factors that influenced the timing of the dramatic improvements in public health that were accomplished through investments in the provision of water and sewage treatment in U.S. cities. During the first 30 years of the twentieth century, American cities experienced dramatic declines in deaths from diseases associated with bad water and poor sewage disposal. Cities bought this reduction by spending large amounts on sanitation systems. Cain and Rotella estimate that a 1 percent increase in sanitation expenditures was associated with a 3 percent decline in the mortality rate from typhoid fever, diarrhea, and dysentery. In the nineteenth century, sanitation was largely supplied privately by the service sector initially explored by Weiss (1975). With continued population and economic growth, these critical services began to move into the public sector. This chapter examines the role that epidemics and demonstration effects played in causing cities to undertake these expenditures. Empirical examination of epidemics reveals that cities that suffered an episode of increased waterborne mortality often responded with higher spending. Demonstration effects took place in a number of ways over time ranging from prominent engineers visiting sanitation works in the middle of the nineteenth century to formal demonstrations projects constructed on a large scale in the years just before the Great Depression. Cain and Rotella argue that epidemics had their biggest impact on the demand for sanitation services, while demonstration effects had their biggest impact on the supply of sanitation services. Together these forces created the conditions whereby American cities allocated the funds that dramatically reduced death rates.

The next three chapters focus on the transformation of the American economy that took place in the late nineteenth and early twentieth centuries. In Chapter 4, Weiss’s long-time collaborators Jeremy Atack and Fred Bateman pursue a natural extension of their earlier work with Weiss on the profitability of nineteenth-century manufacturing. Using expanded samples drawn from the manuscripts of the censuses of manufactures for 1850, 1860, 1870, and 1880, they examine the relationship between firm size and profitability. Many nineteenth-century populists appear to have assumed that larger firm size was synonymous with monopoly power and associated with higher profitability, a belief that played some role in the rise of government regulation intended to offset the perceived advantages of the concentration of economic power. Nonetheless empirical evidence from the twentieth century has generally revealed an inverse relationship between firm size and profitability and Atack and Bateman demonstrate that this same relationship also held in the nineteenth century as well. Moreover, they show that firm size and organizational structure were closely intertwined, with larger firms being much more likely to be organized as corporations rather than sole-proprietorships or partnerships. If profits were decreasing with firm size, it is natural to ask why was average firm size rising at this time? Atack and Bateman argue that two factors help to rationalize this pattern; first, larger businesses enjoyed a lower cost of capital, reflected in their higher capital–labor ratios, and attributable to their access to impersonal capital markets and, second, rates of return for larger firms and corporations were far less volatile than those of smaller firms or those with different organizational structures, so they were much more likely to persist over time

In Chapter 5 Robert Margo and Michael Haines revisit a topic addressed by Weiss in a 1998 p...