Kevin H. O'Rourke and Jeffrey G. Williamson, Globalization and History: The Evolution of a Nineteenth-Century Atlantic Economy. Cambridge, MA: MIT Press, 1999. xii + 343 pp. $47.95 (cloth), ISBN 0-262-15049-2.
Reviewed for EH.NET by Marvin McInnis, Department of Economics, Queen's University, Canada.
Published by EH.Net, August 2000.
For almost a decade Jeffrey Williamson, in collaboration with various other authors, has been investigating the many facets of late nineteenth century international economic integration. It was the topic he chose for his presidential address to the Economic History Association. Parts of this ongoing project have appeared in many articles; I could tally at least fifty-two. The project and many of its findings are well known to specialists in economic history, and many of these experts have closely followed Williamson's work. Why, then, this book?
This work represents an attempt at a final statement. Various pieces of research are brought together into an integrated whole. Moreover, whether or not it is the primary intention, the book speaks to a wider readership--to economists, and historians and many others beside who have not been following the reports in the journals. It is important to assess the book with that in mind. Economic historians who have been following the Williamson project will encounter few surprises. There is little in the book that avid followers of EH.NET do not already know. Nevertheless, seeing it all pulled together into a final compilation will be, for most, a worthwhile read.
For Kevin O'Rourke (University College, Dublin) and Jeffrey Williamson (Harvard) globalization means a substantial increase in the international movement of goods and also of factors of production. As economic historians are well aware, this process was clearly occurring in the late decades of the nineteenth century and was the initial inspiration for their undertaking. To relate nineteenth century globalization to today's concerns, O'Rourke and Williamson start with its alleged consequence--economic convergence. They pay some attention to convergence of real per capita output, but their asserted primary interest is in convergence of wages. Williamson and O'Rourke view this primary dependent variable in the light of its inherent interest as a measure of the economic well being of large numbers of people as well as its provision of an angle on the distributive question, which emerges as one of the central themes of the book. Trade, factor flows, and globalization all have their initial effects on factor earnings. It is the urban, unskilled wage, seen as a measure of the return to raw labor, that is focused upon as the central measure of labor earnings. Already that raises some questions about what is being put forward. Most of the economies included in the study had proportionally large numbers of agricultural workers. On the other hand, increasing numbers of workers were gaining some element of skill, so the measure of the return to urban, unskilled labor has to be interpreted with care. One might also question the claim that the wages of urban unskilled workers is more reliably measured than real per capita income, which O'Rourke and Williamson assert rather than demonstrate.
Globalization, in the O'Rourke and Williamson scheme, does not evidently encompass the international transfer of technology. The authors argue that they are able to fully account for such convergence, especially that of real wages, occurring in terms of international factor flows and the convergence of prices brought about by increased international trade. There is no residual left to be attributed to technological convergence; ergo, international flows of knowledge can be ignored. Hints that those might be lurking behind the closed door, however, appear here and there throughoutthe book. The implementation of improved transport technology is seen to play a powerful role as factors of production are attracted to natural resource rich regions, exploiting frontier opportunities that have important technological underpinnings. Connecting the primary period of convergence, 1870-1900, with a surge of technological developments that revalued natural endowments and gave greater weight to the distribution of accumulated human capital suggests that the whole story might be told from a different perspective and with a different set of prime movers.
It is difficult to give a concise resume of a book that encompasses so much, but a quick overview will allow us to highlight certain important points. The authors begin with what they conclude is evidence of substantial convergence, especially of real wages, among seventeen national economies for which statistical data are available--essentially OECD countries plus Argentina. Experiences among members of this group were quite varied, and O'Rourke and Williamson are fairly careful to show how the selection of countries alters the outcome. They emphasize the contrast between the New World and Old World economies, which is where the big change occurred. Within the European set, the evidence for convergence is much slimmer. The New World economies consist only of the resource-abundant areas settled by Europeans. Surely this conveys an inherent bias within thenarrative. It must inevitably be a tale dominated by the movement of European labor and capital to the (almost) unoccupied spaces of the earth.
The authors, assuming there was important economic convergence in the late nineteenth century, seek to relate it to the "forces of globalization." They begin with the convergence of commodity prices and in chapter 3 show that there was a lot of it (yet again mostly trans-oceanic) and that it was much more a result of greatly reduced transport costs rather than of more liberal trade policy. In chapter 4 they forge the link, following Heckscher-Ohlin, between trade-induced price convergence, factor prices, and the distribution of income. The evidence on income distribution broadly supports the predictions of Heckscher-Ohlin, but once more the main component of the effect is trans-Atlantic. In chapter 5 they turn to trade liberalization and show that the British took the lead because they realized that they were able thereby to generate large and widespread gains in real wages. The same analysis explains why there was less enthusiasm on the Continent for trade liberalization. In chapter 6 O'Rourke and Williamson turn to the backlash stirred up by price convergence and the return to protectionism. They aim to prove that this falls well within the predictive consequences of the interests affected by the grain invasion. In Britain, as expected, large numbers of (potentially voting) urban wage earners gained increases in real wages. On the Continent there commonly were larger reductions in land rents offsetting smaller wage gains, and in France real wages actually fell. Hence the backlash in many of the Continental jurisdictions is understandable and politically predictable.
Chapter 7 presents yet another explanation of the great trans-Atlantic migration. This account begins with a large wage gap, essentially exogenously introduced, that subsequently narrows as large-scale migration proceeds. In the O'Rourke and Williamson scheme the out-migration is then impelled by "demographic forces" that are not clearly explained but which result from past natural increase augmented by an "emigrant stock" effect (relatives and friends, or chain migration). The falling costs of migrant transportation and the emigration-depressing effects of industrialization in sending countries are claimed to be insignificant influences.
The effects of mass migration, which in O'Rourke and Williamson are by far the leading force for convergence, are examined in chapter 8. Attention there is on the United States as the foremost receiving economy and on Ireland and Sweden as two prominent sending economies. Large impacts are found for the United States (negative of course) and Ireland (positive), which will come as a surprise only to those who have convinced themselves that immigration did not lower wages in the United States. In Sweden, on the other hand, large-scale emigration gave wages only a modest boost. The following chapter looks at the effects of globalization on income inequality via factor prices. Once more the main result is the contrasting experience of Europe and the New World. In Europe, in the era of globalization, rising ratios of unskilled wages to land rents were associated with decreasing inequality in the distribution of income.
In chapter 10 O'Rourke and Williamson return to the political backlash against the consequences of globalization. They tell a story of increasingly restrictive immigration policy. This occurred not only in the United States, but in other receiving countries as well during an extended period of time. O'Rourke and Williamson construct an index of restrictiveness in immigration policy and, in a regression analysis, show that it moves most significantly in relation to the impact of immigration on the labor market.
The authors devote two chapters to international capital flows and the integration of world capital markets. The major claim here is that, in relative terms, capital markets were even more integrated than they are today. The big problem for O'Rourke and Williamson is that capital was not generally flowing in the opposite direction to labor. The labor abundant regions of the world were not attracting large capital flows. Capital, then, was acting as a divergent influence, substantially offsetting the convergent effects of international migration. It was the abundant land and the unexploited natural resources of the New World that attracted most of the capital. Here O'Rourke and Williamson explicitly recognize that European, especially British, capital flows were greatly attracted by foreign investment demand on the frontier. It would have been more satisfying to see that thought more effectively integrated into the analysis of labor migration (chapter 7) where capital is repeatedly described as "chasing labor." An alternative model, which O'Rourke and Williamson do not consider in their migration analysis but which is implied in their treatment of international capital flows, would have capital attracted to the New World to combine with the abundant natural resources there and with the labor drawn there by the abundance of capital.
In chapter 13 O'Rourke and Williamson directly address an issue that has been much debated in recent years--whether trade and factor flows are substitutes or complements. This is one of the fresher sections of the book, relying less on previous publications by the authors. Cases of substitutability are not to be found. There are some notable cases of complementarity between trade and factor flows, but more commonly the relationship is neutral.
In their final summing up (chapter 14), O'Rourke and Williamson reiterate their message that the globalization force most responsible for wage and even per capita income convergence was mass labor migration, given the generally perverse direction of capital movements and the more modest contribution of commodity trade and price integration. They ask whether serious wage and income convergence can be expected without large-scale international migration. A second important point is that in the late nineteenth century globalization backlash was endogenous and could well be again. O'Rourke and Williamson also offer a partial admission that they may have underplayed the role of technology and the role of its international diffusion. Throughout, they have ignored that influence and neglect it on the grounds that they can obtain substantial explanation of the phenomena they wish to account for without calling technology into play. That makes less than a wholly convincing case against it.
Throughout their book O'Rourke and Williamson acknowledge the limitations of their study. Their continuous inclination, though, is nevertheless to forge ahead. They return to some of those limitations in their final chapter. The authors, for example, admit that individual national experiences were highly varied--so varied, some readers might suspect, that generalizations cannot be made. The results reported are often from the most prominent cases that best exemplify the authors' argument. Time and again the focus is on the relationship between the United States and Europe. That is an important case, but probably sui generis and not enough of a basis for an international generalization.
This book should be widely read due to the very fact that the authors' arguments are open to debate. Professional economic historians, especially those interested in entering the debate, should see the whole account in its crystallized form. For students this book is a useful introduction to an important topic. It is also a book to be recommended to our colleagues in economics, history, or other disciplines who do not ordinarily pay attention to what is being written as economic history.
Marvin McInnis primarily studies Canadian economic development in the late nineteenth and early twentieth centuries. His most recent writing includes two chapters on Canada in Michael Haines and Richard Steckel, (eds.) Cambridge Population History of North America. New York: Cambridge University Press, 2000.
Copyright © 2000 by EH.NET. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given to the author and the list. For other permission, please contact the EH.NET Administrator (email@example.com; Telephone: 513-529-2850; Fax: 513-529-3308).
Posted: 7 August 2000