Health and Labor Force Participation over the Life Cycle: Evidence from the Past

Published by EH.NET (January 2004)

?

Dora L. Costa, editor, Health and Labor Force Participation over the Life Cycle: Evidence from the Past. Chicago: University of Chicago Press, 2003. xvi + 343 pp. $75 (cloth), ISBN: 0-226-11618-2.

Reviewed for EH.NET by John E. Murray, Department of Economics, University of Toledo.

This volume is a progress report from an ongoing research project — but what a research project! In what may be the best-funded effort in the history of economic history, scholars at Chicago, MIT, Brigham Young, and several other prominent universities have created a database consisting of records on some 35,000 Union Army veterans. The effort was led by Robert Fogel, to whom the volume is dedicated. Each man in the sample can be described by the astounding number of 15,000 possible variables. In terms of quality and quantity of individual level data, the Union Army Life-Cycle Sample (UALCS) compares well to large longitudinal samples of the present such as the PSID and NLS.

This volume offers eight substantive papers that draw on the UALCS, plus one based on a separate sample from a slightly earlier period, in addition to an introduction and summary by the editor, Dora Costa (MIT) and a reflection on the history of the UALCS’s compilation by Larry Wimmer. Some chapters succeed more fully than others, but all engage the reader. As the authors present their problems and work through the approaches made possible by the UALCS database, one can only admire the incredible effort that went into constructing it and applaud the imagination that went into analyzing it. Few analytic stones were left unturned, which the reader can see for himself — over a third of the book is given over to tables of means and regression estimates.

The volume begins with Wimmer’s memoirs on the process of building the UALCS over a period of about two decades. If you ever thought you had plunged into the research project that would never end, this essay may provide some perspective. It certainly must be more satisfying now to look back on how the UALCS project grew than it was at the time to decide how to code surgeons’ reports while keeping an eye on the shrinking grant meter.

Joseph Ferrie offers the only chapter not based on the UALCS sample. He instead examines the relationship between wealth and mortality using his own linked samples of fifty counties from the 1850 population and mortality censuses, and one of eleven counties for both 1850 and 1860 with 92,079 people from both 1850 and 1860. The latter sample includes wealth data, which allowed Ferrie to test for connections from wealth to mortality. He finds strong effects of wealth to reduce mortality, but little in the way of occupational effects. He examines potential linkage biases and draws conclusions from the regression results with great care.

Chulhee Lee provides a model essay that plays to all the UALCS’s strengths. He examines an important question, the relation between early life disease exposure and later life morbidity and mortality. He compares his findings with the full database to his earlier findings from a pilot sample (Explorations in Economic History, 1997) and finds them to be consistent with each other. This chapter exploits the longitudinal nature of the database to consider long-term effects of early exposure — the period of risk runs from the Civil War, when the subjects were young men, into the twentieth century when the survivors collected the last of their pension payments. Lee also directs his findings toward other hypotheses in the literature. If early disease exposure reduces later mortality and morbidity, as he finds, then insult-accumulation theories will not describe the veterans’ experiences.

One strength of the volume as a whole is its coherence, as several chapters address the same important issues. Daniel Scott Smith’s chapter on mortality among New York regiments nicely complements Lee’s chapter. Smith adds controls for region of residence within New York state and POW status, and like Lee finds an important role for early life seasoning in later life mortality.

Sven Wilson and Clayne Pope re-examine the issue of influences on final adult heights of Union Army soldiers with more data. Their results do not comport well with the recent findings of Michael Haines, Lee Craig, and Thomas Weiss (Journal of Economic History, June 2003), who also used the UALCS as a source of height data. For example, where Wilson and Pope find a positive and significant relationship between mortality and height, Haines, Craig, and Weiss found a more conventional negative and significant relationship. Such unexpected results deserve fuller explanations than they were given here.

Mario A. S?nchez produces one of the most engaging chapters with his study of migration, return migration, and mortality. The conventional wisdom about migration within America might be summarized by Frank Lloyd Wright’s aphorism that the continent is in a tilt and everything is sliding into California. In the best part of this chapter, S?nchez considers the opposite trend: return migration. Fifteen percent of his subsample returned to whence they came, a surprisingly large share. S?nchez proposes that a desire to reunite with family left behind is an important reason.

Because the UALCS sample provides information about particular health conditions among surviving sample members, it can be made to yield a great deal of particularly precious information on morbidity. It is hard to overstate the importance of this characteristic of the sample, and several authors put this strength to good use. Sven Wilson finds that soldiers who became ill with measles or respiratory illness while serving were likely to suffer from chronic obstructive pulmonary disease in their old age. Werner Troesken and Patricia Beeson show that veterans exposed to lead consumption were at risk of developing relatively minor conditions related to lead toxicity. Chen Song and Louis Nguyen examine hernia, which may well have influenced short term labor supply even if they did not find a relation between it and retirement decisions.

The UALCS is a grand achievement, and this book is both an important contribution to the history of American health and an impressive testament to what hard work, persistence, and imagination can accomplish. Scholars will return to the insights it provides, the methods it proposes, and the questions it provokes into late nineteenth and early twentieth century male health and labor supply for years to come. Costa and the contributors to this volume have set a high standard for future publications from this ongoing super-project.

John E. Murray is Associate Professor of Economics at the University of Toledo. His recent articles on health history have appeared in Demography, Population Studies, and Social History of Medicine.