Capital in the Nineteenth Century. By Robert E. Gallman and Paul W. Rhode (Chicago, University of Chicago Press, 2019) 381 pp. $65.00

Abstract

Lead author Gallman died in 1998, but Rhode, at the urging of Claudia Goldin of the National Bureau of Economic Research (nber) and Harvard University, was able to put together this book from Gallman’s published works and research files, supplemented by discussions with many prominent economist historians and Gallman’s longtime assistant at the University of North Carolina. The result is a mash-up of economic history and history of economic thought that would have made a valuable contribution if published twenty years ago. Its bibliography, for example, is great, but only for sources published before the start of the third millennium.Despite its title’s play on Thomas Piketty’s Capital in the Twenty-First Century (Cambridge, Mass., 2013), the book does not systematically address issues of wealth or income disparities. It does, however, point to the importance of considering population age structure when assessing changing levels of inequality. In addition, Rhode chastises Piketty for misunderstanding nineteenth-century America: “In the areas dominated by family farmers and small business, there was widespread participation in the process of economic growth” (15).In fifteen chapters, the book covers Gallman’s methods for calculating various components of the physical capital stock of the United States between 1840 and 1900, including that utilized in agriculture, mining, and manufacturing; real estate and trade; transportation; communication and electric utilities; inventories; and consumer durables. Other chapters consider wealth in the colonial and early national periods, national product growth from 1774 until 1980, investment flows, and annual product series from 1834 until 1909. Overall, Gallman’s series help scholars to understand America’s growth and development over long intervals, but not year-to-year, as it takes no account of business cycles.Much like mentor Simon Kuznets, Gallman scoured sources for solid statistical data, to which he applied both modern and contemporary national income-accounting concepts. If numbers seemed shaky and could not be independently verified, Gallman eschewed them rather than proffer dubious estimates that could lead other researchers astray. Rhode assures readers that Gallman sought to build a solid empirical edifice for understanding the evolution of the American economy, not to publish clever theories or policy pronouncements or to apply advanced statistical techniques.The most interesting methodological issue involves Gallman’s willingness to consider national accounts from a nineteenth-century perspective (importantly, Gallman always counted slaves as people, not as equivalent to machines or livestock). For example, clearing and breaking virgin land for agriculture, barely undertaken today, was then a major source of capital accumulation that Gallman tracked along with other capital improvements like ditching, fencing, draining, and irrigating. Including those improvements showed that agriculture was more capital-intensive than manufacturing for much of the century.Gallman also tracked unimproved land, the value of which rose as a share of national wealth during the second half of the century. Other surprising findings from Gallman’s work include documentation that in the nineteenth century, structures constituted a larger portion of the capital stock than machines did, and that capital formation was a more important component of output growth than in the twentieth century.Starting with 1850, the decadal census figures provided key benchmarks for Gallman’s series. He anchored earlier estimates on the work of historian Alice Hanson Jones (1774), federal tax records (1798, 1813–1815), and early statisticians Samuel Blodget, Jr. (1806), and Ezra Seaman (1840). Gallman had to adjust for changing price levels, differing treatments of depreciation, and differing valuation models, including acquisition cost, reproduction cost, and market price. Gallman also clearly considered the effects of new technologies and techniques, like balloon framing in construction, but better interest-rate series than he used are now available. Cross-checking independently developed stock-and-flow data substantiates Rhode’s claim that Gallman’s methods produced fundamentally sound round estimates.Capital reflects Gallman’s penchant for unadorned prose and painstakingly detailed descriptions of sources and assumptions. One may question Gallman’s adjustments, but nobody can claim to be confused by them, especially after Rhode’s careful editing. Ultimately, only researchers trying to reconstruct or build upon Gallman’s series, a project that Rhode encourages, would be positioned to critique Gallman’s estimates fully.