Water, Race, and Disease starts with a juxtaposition of components of African Americans' quality of life in the early twentieth century. There were rapid improvements in health (especially in cities), but enormous and widening racial disparities in political, educational, and economic resources. How should economic historians square these facts in telling the story of American economic development?

Werner Troesken argues that the extension of the public health infrastructure, specifically sewers and clean water, led to significant disease reductions for both blacks and whites. Moreover, he argues that blacks benefited at least as much as whites, and probably more, from the provision of clean water. This begs a second question: given that sewers and water purification were costly and that blacks had little political clout, why did whites not exclude blacks from the provision of clean water? Troesken argues that concerns about disease spillovers from blacks to whites were key in this regard, especially in the context of cities that were more racially integrated (in the geographic sense) than they were later in the century. To this, I will add one more question and discuss it below: What does this tell us about the nature of racial discrimination circa 1910?

As an interdisciplinary undertaking, the book is a great success. Troesken draws on and contributes to extensive literatures in urban history, epidemiological history, and economic history; he compiles and analyzes a wide range of data sets; he develops several illuminating case studies; and, throughout, he brings a tremendous amount of evidence to bear on the questions at hand. I strongly recommend the book to economic historians, to public, urban, and health economists, and to historians who study race, cities, and health in the United States.

Chapter 2 introduces readers to the essential epidemiological and historical context —the prevalence and transmission of water-borne diseases in a world of highly impure water. Chapter 3 characterizes the extent of sewer and public water provision in U.S. cities, drawing heavily on the Census Bureau's General Statistics of Cities volumes for 1909 and 1915. Some readers will find it surprising that, by that time, southern cities did not trail the rest of the country in sewer and water provision (on average). Data on racial differences within cities in the likelihood of having sewer and public water connections are more difficult to come by. For a smaller set of cities, Chapter 3 uses data from the Negro Mortality Project and the Children's Bureau, whereas chapter 4 makes clever use of data from the Integrated Public Use Microdata Series (IPUMS) to flesh out case studies of Memphis and Savannah. Chapter 5 combines a case study of Shaw, Mississippi, where black residents successfully sued the city for providing inadequate public services (in 1971), with a nationwide comparison of black and white access to public water and sewer systems in 1960 (using the IPUMS data). Chapters 6 and 7 provide the most detailed econometric analyses in the book and establish that water filtration really did lower disease rates (specifically, typhoid death rates) for both blacks and whites. Chapters 8 and 9 extend the econometric analysis and state the book's conclusions, respectively.

The book is packed with facts, figures, and estimates, and I have space to highlight only the most central contributions. First, public investments in sewers and water purification contributed to the great decline in mortality rates in the period under study, but they were not the only factors at work. Troesken makes a laudable statistical effort to separate the impact of water from the impact of other factors. Second, within cities, blacks were indeed less likely than whites to have access to sewers and public water, but in most cases the racial gaps were not very large. Perhaps more importantly, Troesken demonstrates repeatedly that blacks in urban areas gained at least as much as whites, in terms of disease reduction, from the introduction of better water treatment. Third, the geographical separation of blacks and whites within cities was not as pronounced in the late nineteenth and early twentieth centuries as it was later. This facilitated black access to clean water both at the time the infrastructure was built and (I would add) as blacks moved into formerly white-occupied areas.

Aside from lingering concerns with some aspects of the econometrics (for example, it is extremely difficult to estimate spillover effects and to find strong instrumental variables in this context), I found the book's evidence and arguments to be convincing. Potential extensions of Troesken's work are easy to identify, reflecting the book's rich intersection of topics. For example, based on the stories Troesken tells for urban areas, I would be eager to read a new investigation of urban disamenities, living standards, and wages in this period. It would be interesting to see whether the implicit “compensating wage differential” for urban residence (relative to rural residence) differed by race or changed greatly, though the data requirements for such an investigation are not easily met. A less far-flung suggestion would be to explore the economics of households' decisions to connect to the public water system (a cost that appears to have been borne privately), conditional on having access to it. This distinction arises in chapter 4's case studies, and it is an interesting economic problem unto itself. How did households at the time value and respond to the services being offered?

What does Troesken's story tell us about discrimination in the early twentieth century? By pointing out specific concerns about disease spillovers from blacks to whites, Troesken emphasizes that self-interest can sometimes trump racial animus, even in the Jim Crow era. Related extensions of this general principle are important to consider. For example, a large proportion of blacks worked for white employers — even in a world without disease spillovers, self-interested white employers would prefer to have healthy black workers, as they are more productive, less frequently absent, et cetera. Another variation on this theme, one that is often cited in the literature on racial disparities in school provision, is that, to the extent blacks “voted with their feet” and sought out places with more amenable public services, self-interested white employers had an incentive to provide a certain level of such services.