Quantitative Economic History: The Good of Counting

Published by EH.NET (September 2008)

Joshua L. Rosenbloom, editor, Quantitative Economic History: The Good of Counting. London: Routledge, 2008. xvii + 176 pp. $130 (hardback), ISBN: 978-0-415-77349-2.

Reviewed for EH.NET by Mary Eschelbach Hansen, Department of Economics, American University.

Quantitative Economic History is a compact volume dedicated to the prolific Thomas J. Weiss. Joshua Rosenbloom (Weiss? colleague at the University of Kansas), in cooperation with contributors from among Weiss? many students and collaborators, has certainly maximized the ratio of information to ink. The volume is a worthy celebration of Weiss? accomplishments.

Throughout his career, Weiss has scrutinized the details. He is, for example, the authority on the details of the inconsistencies in census enumeration that make the construction of historical series on the size of the labor force so messy. His efforts to straighten up our statistical house make our results more robust. Rosenbloom places the volume in this context, and the contributors make the case by following his example. Each of the seven papers is based upon a detailed accounting of historical fact. Each fact is carefully considered in relation to the concept purportedly being measured.

The topics in the volume are as far-flung across the landscape of economic history as Weiss? own work. The volume begins with two studies in demography. John Ermisch (University of Essex) contributes a history of non-marital births in England and Wales since 1845. He accounts for increases in the rate of non-marital births by considering the effect of macroeconomic conditions on marriage (lower real wages inhibit marriage, even for already-pregnant women) and social interaction effects of cohabitation (cohabitation facilitated a decline in the stigma costs of non-marital births). The accounting necessary to capture the effects of social interaction is based upon demographic histories constructed from interviews in Britain. Louis Cain (Loyola and Northwestern) and Elyce Rotella (Indiana) argue that the decline in urban mortality in the early twentieth century was ultimately the result of epidemics. Epidemics increased urbanites? demand for city leaders to do something about water-borne illness. At the same time, city leaders saw a dramatic decline in the cost of obtaining information about what, in fact, should be done. Statistics on investment in city-owned infrastructure complement a careful narrative accounting of flows of information.

In the next three essays, contributors examine nineteenth and early twentieth century change in three different sectors: manufacturing, transportation, and agriculture. Jeremy Atack (Vanderbilt) and Fred Bateman (Georgia) use samples from the manuscript censuses of manufactures, an extension of Weiss? work with Bateman. They show that firms with greater capitalization had lower average profits, but more predictable profits, than smaller firms. Michael R. Haines (Colgate) and Robert A. Margo (Boston University) try to capture the treatment effects of connecting counties to rail networks between 1850 and 1860. They use a data set that links the most recent compilation of individual and county-level census data to manuscripts of the census of social statistics, as well as to geographic data on rail access in counties. Lee A. Craig (North Carolina State) and Matthew Holt (Purdue) contribute an accounting of the benefits of mechanical refrigeration to farmers and consumers. They examine a long time series of corn and hog prices and demonstrate that refrigeration reduced the seasonal swings in prices long associated with poor harvests of corn.

The final two contributions return to labor-related issues. Rebecca Holmes, Price Fishback (Arizona) and Samuel Allen (VMI) attempt to measure differences between states in overall intensity of regulation of labor markets during the Progressive Era. They use federal reports of labor regulations enacted in the states and panel state-level economic data collected from a variety of sources. Simple specifications reveal that labor law intensity was positively correlated with labor productivity, but exploratory fixed-effects models do not indicate that the laws, by themselves, affected labor productivity in a meaningful way when other differences between states are considered. Joshua Rosenbloom and Gregory W. Stutes (Minnesota State-Moorhead) examine the usefulness of the Integrated Public Use Microdata Series for 1870 as a source for the measurement of inequality of wealth. They contribute to the literature on inequality and growth by exploiting the geographic dispersion within the 1870 sample.

It is, of course, possible to criticize the papers individually. The reader will want more concrete evidence that the best place to split Ermish?s long time series on non-marital births is World War II, even though the clearest discontinuity is in the late 1970s when unemployment associated with the oil crisis coincided with improvements in the availability of contraception and abortion. The reader may wonder how the municipally-owned water systems described by Cain and Rotella compared to private systems. The reader will want Atack and Bateman to discuss the limits of their data to address important questions about both the life cycle of firms and the life cycle of capital: Do small cap firms become larger cap firm through growth or through merger? To what extent are profits in large cap firms limited by the vintage of their capital? The economic geographer might object to the simplification of von Thunen?s prediction concerning the effect of railroad expansion on agriculture in the paper by Haines and Margo. Von Thunen would have expected the crop mix, not only the value of farm output, to respond to transportation improvement. The reader will wonder what quantitative evidence on the investment in refrigeration was rejected by Craig and Holt. The reader will hope that Holmes, Fishback, and Allen plan to ask whether regulatory climate affected industrial employment and the ratio of skilled to unskilled labor. And finally, the economic historian will want Rosenbloom and Stutes to provide more context for interpreting their estimate of inequality: how different does the U.S. in 1870 look from the U.S. later or from developing countries today?

Yet, no quibbles with individual contributors will reduce the usefulness of the volume. Because of its breadth, the volume will be opened and referenced often. Future students will open it expecting to extract the details. But if they read carefully, they will take away more than the details. They will learn how to scrutinize the details and improve upon them. Professor Weiss will continue to teach about the good of counting, albeit indirectly, through the works assembled in his honor.

Mary Eschelbach Hansen is Associate Professor of Economics at American University. Hansen, with co-author Brad Hansen, has been working towards a history of bankruptcy in the first half of the twentieth century. ?The Role of Path Dependence in the Development of U.S. Bankruptcy Law, 1880-1938? appeared in the Journal of Institutional Economics in 2007; ?Religion, Social Capital, and Business Bankruptcy, 1921-1932? will soon appear in Business History; more will follow.