EH.NET BOOK REVIEW
Published by EH.NET (July 1998)
James C. Riley, Sick, Not Dead: The Health of British Workingmen during the Mortality Decline. Baltimore: Johns Hopkins University Press, 1997. xvii +342 pp. $58.00 (cloth), ISBN: 0-8018-5411-3.
Reviewed for EH.Net by Herb Emery, Department of Economics, University of Calgary.
With Sick, Not Dead, James Riley has written an ambitious book on the important subject of trends in the health status of nineteenth-century British workingmen. The first part of Riley’s book provides an extensive description of friendly societies in England, the primary sources of sickness and health insurance for British Workingmen in the nineteenth and early twentieth centuries. Riley also examines the extent to which friendly society members had access to medical services at a relatively low cost due to the control friendly societies had over the medical marketplace. The more important contribution of this book, however, is that suggested by its title and the subject of the second part of the book. Riley challenges the belief of many scholars that the documented decline in British mortality in the nineteenth century reflected that British workingmen were also healthier. Living longer and healthier clearly indicates a rising standard of living. In contrast, Riley’s examinations of sickness benefit claim statistics compiled by the Ancient Order of Foresters (AOF) demonstrate that the decline in mortality was not necessarily an indicator of improved health. Foresters it seems, lived longer, were sick less often, but were sicker for longer periods of time. Riley also uses the Forester claims statistics to show clear regional patterns of sickness in Britain which were stable over the period 1870 to 1910. Thus, Riley shows that it is difficult to make conclusions about health status of workingmen on a national level. If the AOF sick benefit claims statistics are representative of the health status of the British working class, then Riley has contributed an important insight into the health and standard of living of nineteenth century British workingmen.
Riley’s primary finding of a surviving but sicker British population after 1870 requires that the observed increase in sickness time in AOF Courts was due to changing health conditions of workingmen, all else equal, and was not merely an artifact of compositional changes in AOF Court memberships. Observed sickness time in AOF Courts could have increased over time because the health of workingmen was changing, or because more of the membership was older with higher sickness risk, or because members with higher sickness and injury risk occupations represented more of the membership. Other potential explanations for changing observed sickness patterns could be changes in the AOF’s rules for claiming benefits, or changing attitudes of members towards claiming sick benefits. Riley addresses, and dismisses, each of these possibilities for the observed increase in AOF sickness time claimed with the exception of changes in workingmen’s health. Essentially, the conclusion that patterns of health were changing is the explanation attached to an otherwise unexplained increase in sick time claimed over time. The reader must decide if Riley has adequately explained away (or, in his regressions, controlled for) alternative explanations for the increase in the length of time per year that Foresters claimed sick benefits.
Riley is extremely careful in letting the reader know the importance of purging the sickness benefit claim statistics of effects due to aging members to identify the underlying trend in health status of British workingmen. Riley shows that as an individual aged, his length of sickness spell increased exponentially. He does not observe each individual member’s age, but he does observe the average age of the members who are generating the claim statistics. As Riley points out, while the average age of the members is a good measure of central tendency in the claims statistics, he still needs to control for the dispersion of ages in a given membership. To see why this is the case, consider two Courts each with memberships with an average age of 30 years. All members of both Courts face identical age-specific sickness risks. In membership A, all members are 30 years old. In membership B, one-third of the members are age 20, one-third of the members are age 30 and one-third are age 40. Even with the same average age, membership B with more “older” members will generate higher observed sick claims since the increase in claims from a 40-year-old member compared to a 30-year-old member is larger than reduction in claims from a 20-year-old member compared to a 30-year-old member. Riley identifies the rate of initiations into court membership and the number of years a court had been operating as key factors influencing age clustering (or conversely, age dispersion) in the membership. Initiations brought younger members into AOF Courts and tended to slow down the aging of the membership. He includes the initiation rate (new members to existing members) and the years since the AOF Court was formed in his regressions. Riley regresses the sickness time variable on these controls and still finds an increase in sickness time claimed over time.
The question remains whether the increase in sickness time that Riley identifies is a true trend in unobserved health status or just a biased residual effect resulting from the imperfect proxy variables for controlling for the increases in sickness due to an aging membership. There is good reason to suspect it is the latter since Riley is silent on quits/secessions from Court memberships in his discussions on controlling for age in his sickness time regressions. Riley points out that when quits/secessions occurred, they “typically occurred within a few years of joining” when members were in their 20s or 30s. Thus, where initiations reduced the average age of the membership by bringing in younger members, quits or secessions accelerated the aging of the Court membership. In other words, the net of initiations and secessions is the relevant factor for controlling for the aging of Court memberships since both influence the number of members at younger ages in the membership. In the absence of controls for secessions, Riley’s maintained assumption for interpreting the trend increase in AOF sickness time claimed is that membership was a lifetime status for joiners. Unfortunately, Riley does not convey much information about the extent to which membership in a friendly society was a lifetime status for initiates. The reader will learn that Forester secession rates were higher than those of the Oddfellows in Britain but Riley does not tell the reader what Oddfellow secession rates were. Readers will not get a sense from this book how big an omission from the analysis this potentially is. While not directly comparable to the British orders, in the Independent Order of Odd Fellows in North America, the average length of membership was only around 5 years and only one quarter of members remained in the membership for 25 years. Only a minority of members did not secede from membership. Thus, one explanation for Riley’s measured trend in sickness time after 1870 is that as the number of initiations into AOF Courts slowed, the aging effect of secessions became important. Riley’s controls for aging, which only account for the rate of initiation, understate the true extent of aging in the membership. This bias arising from the exclusion of secession rates would appear as an otherwise unexplained, or residual, trend increase in sickness time claimed over time.
A frustrating element of this book is that the reader does not really know who belonged to the AOF. Riley asserts that the members were drawn from the working class, and that the AOF membership was similar to the Oddfellows membership which he shows was representative of the British population in terms of occupational distribution. Riley provides no direct evidence in support of this assertion. This shortcoming of the book is important for understanding whose health patterns we are learning about. It is critical for interpreting Riley’s analysis of regional sickness patterns since the analysis requires that the Foresters shared the circumstances and characteristics of the communities in which they lived. Riley operationalizes this point in Chapter 9 asserting that “earlier parts of this book show that AOF members as a whole closely resembled the central ranks of the working population in Britain, which implies that they did so also in most local communities” (p. 243). On the next page, however, Riley notes that for Britain as a whole, the Foresters represented 7.3 percent the male population in 1891 but for individual counties, this proportion varied from a low of 0.7 percent of the male population to a high of 20 percent. This variation seems hard to reconcile with the assumption that AOF memberships were everywhere equally representative of the local population. That AOF memberships were not always representative of the county population may explain Riley’s finding that coal mining and mining trades, occupations known to have high sickness and accident risks, were not statistically significant factors for explaining AOF sickness time or mortality. Riley’s does not entertain the possibility that the AOF dealt with high-risk occupations like coal mining by discouraging the participation of miners in the organization. If that was the case, miners were not in the AOF Courts, hence the claims statistics are not affected by the amount of mining employment in the county in which the AOF Court is located. The presence of miners would have affected the ratio of AOF members to the county population. Riley could have examined this possibility with his data by examining the correlations between the ratio of AOF members to county population and the importance of mining employment in the county.
Finally, a large focus of the early chapters of the book concerns the relationship between friendly societies and doctors. Riley documents the extent of access to doctors enjoyed by friendly society members and the extent of control over the medical market place enjoyed by the consumers. Riley has done an excellent job synthesizing various sources on this issue and providing original evidence from Forester Court minute books. Readers should be cautious however, in how they interpret sickness risks faced by workingmen from this discussion and in how they interpret what friendly societies were doing. Riley’s focus on access to direct medical care through friendly society membership obscures the more important cost of sickness and injury in the nineteenth century, lost earnings. Friendly societies like the AOF may have provided access to physicians for members and have discussed the nature of care, but the income replacement benefit was clearly more important in terms Court finances. The nineteenth-century actuarial investigations of sickness were motivated by concerns about the sustainability of the income replacement sick benefit, not concerns over revenues to finance medical care. While it is interesting to know how the fraternal medical economy worked in the nineteenth century, it is a puzzling focus for an analysis of friendly societies that were more concerned with insuring men against the loss of income due to illness or accident.
In the end, Riley has provided a book that is a substantial improvement over many of the books written on British friendly societies. He provides fresh information and updates a literature that has not seen a great deal of activity for some time. Even though I have my doubts about the usefulness of friendly society sickness claim statistics for studying the health of workingmen, it is interesting to see the patterns that emerge from the data. For any scholars considering a project similar to this one, Riley’s book should be considered the point of departure.
Herb Emery Department of Economics University of Calgary
Herb Emery has co-authored (with George Emery) the forthcoming book (1999) A Young Man’s Benefit: The Independent Order of Odd Fellows and Sickness Insurance in the United States and Canada (McGill-Queen’s University Press ).