Introduction: The New Comparative Economic History

Timothy J. Hatton, Kevin H. ORourke, and Alan M. Taylor

The cliometric revolution in economic history is now nearly half a century old. By applying formal economic methodology to historical questions, it marked a sea change in the way that issues such as slavery or the Industrial Revolution were debated, and thus fully deserves its revolutionary label. This books purpose is to showcase a more recent trend in economic history, which is an evolution rather than a revolution but which still represents a distinct change in the way that many economic historians view their role, do their work, and interact with the broader economics profession. As the label suggests, the New Comparative Economic History reects a belief that economic processes can best be understood by systematically comparing experiences across time, regions, and, above all, countries. As such, it diers from much traditional work in cliometric history in the nature of the questions it asks and in the nature of the answers it provides. But like cliometrics, it has a long ancestry. It draws inspiration from such giants as Simon Kuznets, W. Arthur Lewis, and Douglass North, all three Nobel laureates. For pioneers like these, comparative thinking was so central, so natural, and so ingrained that it hardly needed to be mentioned as a point of reference and warranted no special rhetorical logo. From these towering intellectual gures and their many followers emerged what we might callonly with the benet of hindsightthe old comparative economic history. In its time a ourishing new intellectual direction, this eld set out some of the big questions and the earliest pieces of evidence on which todays scholarship builds. The New Comparative Economic History has taken up some of these earlier themes, brought in new perspectives, and built on the achievements of the earlier pioneers. But the transition from the old to the new has taken time, and it needed the evolution of economic historyand just as important, of

Timothy J. Hatton, Kevin H. ORourke, and Alan M. Taylor

economicsto set the stage for a second and quite distinctive burst of comparative economic history. Cliometrics initially cut its teeth on questions that had emerged from national historiographies, and thus it had two limitations: it was national in scope, and it sought questions from the historians agenda rather than from the economists agenda. The aim of early cliometrics was to show that these questions could be better understood by using economic theory and rigorous quantication. As such, the lessons to be learned were largely of interest to other economic historians, who were usually focused on their own countrys experience. By contrast, the New Comparative Economic History is motivated by questions being posed by economists that are less nation-specic in scopequestions such as the sources of economic growth, the importance of institutions, and the impact of globalization. For example, its practitioners are less interested in the rst, British Industrial Revolution as a specic historical episode than in what it tells us about the nature of economic growth. If the latter is the motivating question, then it makes sense to study not just this episode but other transitions to modern economic growth (and just as important, failures to achieve such transitions) as well. To take another example, New Comparative Economic History practitioners will tend to be interested in the impact of specic economic, political, or cultural institutions, such as slavery, democracy, or religion, not just in and of themselves, but also in how institutions inuence economic development. As such, it makes sense to exploit the fact that history provides an immense variety of institutional arrangements that allows us to arrive at conclusions with a general validity, rather than simply improving our understanding of an institution embedded in a particular time and place. The New Comparative Economic History is motivated by current debates among academic economists and policymakers rather than following agendas set by historians. At the same time, it aims to convince the rest of the economics profession that they have a lot to learn from taking history seriously. History can inform current debates by focusing on long-run trends rather than on short-run ups and downs in economic activity. As a result, it can provide an antidote to analysis that looks only at the present or the very recent past. It also provides a powerful reminder that one model may not t all circumstances, because economic relationships that may seem immutable to economists familiar only with the present may in fact be highly time- or place-specic.

Introduction

A good example of this concerns the relationship between openness and economic growth. While many studies have shown that the two are positively correlated for the post-1950 or post-1960 period, Clemens and Williamson (2004b) nd that the correlation was reversed in earlier epochs and (to complicate matters further) that it diered across geographical regions. In turn, such ndings can help economists think more carefully about what are the underlying assumptions driving their models predictions and the changing regimes driving their econometric results. Economists are also reminded that the political, institutional, and economic environment can play an important role in shaping the relationships between policies and outcomes. This institutional sensitivity, a distinguishing feature of economic history in general, is nding a receptive audience among todays mainstream growth, development, and international economists. Three other closely related features of the New Comparative Economic History have been its emphasis on the interactions between economies, the economic mechanisms driving aggregate correlations in the data, and political economy questions. Consider, for example, the issue of whether economic growth leads to country or regional convergence (Williamson 1965; Abramovitz 1986). A vast literature has tried to answer this question by exploring regressions of country-specic growth rates on initial income, trying to uncover evidence of unconditional, or possibly conditional, convergence. This methodology implicitly assumes that each country is an island, with its growth rate determined by a series of country-specic variables. At best, openness to trade or some other indicator of integration with the rest of the world will appear in the equation as a country-specic right-hand side variable. However, economic intuition, common sense, and experience all tell us that whether poor countries catch up with rich countries will depend critically on the nature of the economic interactions between them, on the extent to which capital, more productive technology, better policies, and more ecient institutions are transferred to backward countries, or on the nature and extent of commodity trade and labor migration. In this light, no country is an island, and thus the researchers attention is drawn irresistibly to the study of international capital ows, migration, trade, technological transfer, and the diusion of good public policy. In turn, studying these ows can tell us a lot about what drove economic convergence (and divergence) between poor and rich regions at particular times and in particular parts of the world. Thus, economic convergence

Timothy J. Hatton, Kevin H. ORourke, and Alan M. Taylor

between poor and rich countries in the late nineteenth-century Atlantic economy was mostly driven by labor migrations (Hatton and Williamson 1998; ORourke and Williamson 1999; Taylor and Williamson 1997). Capital ows occasionally played a supporting role (as in the case of Sweden) but more often were a force for divergence rather than convergence. This leads to the historical questions, Why has capital never owed more systematically to poor countries (Clemens and Williamson 2004a)? And why has migration not played the pro-convergence role in the twentieth century that it played during the nineteenth (Hatton and Williamson 2005)? The latter question is obviously one that requires a political economy explanation. Indeed, policy is clearly one of the key drivers of economic integration between countries, and it can help explain cross-country variations in economic performance more generally. Why the New Comparative Economic History is concerned with political economy issues is thus easily understood. The New Comparative Economic History not only asks dierent questions from those posed by earlier cliometricians; it also diers in the type of answer it provides. One key rule is never to throw away information; history is the only laboratory we have, and it is important to exploit any variation in the data that can be found there. One important characteristic of the eld, therefore, is its constant tendency to expand the range of countries being explored, from Great Britain and the United States (Williamson 1974; 1985; 1990), whose histories inspired the rst cliometricians, to Japan (Kelley and Williamson 1974), the Atlantic economy (Hatton and Williamson 1998; ORourke and Williamson 1999; Taylor and Williamson 1997), Latin America (Coatsworth and Williamson 2004; rtola and Williamson 2006), and the wider world (Hatton and WilliamBe son 2005; Williamson 2006). In particular, the developing world is not only where most people live, but it comprises a variety of countries with dierent institutions and endowments whose economic histories are full of lessons for policymakers and growth economists. Uncovering these lessons has been and will continue to be one of the key tasks of the New Comparative Economic History. If long-run cross-country variation is a key to understanding long-run economic processes, and consequently individual country experiences, then that has implications for the methodologies that economic historians will tend to use. Cliometrics became famous for its use of counterfactual analysis, and this made sense in a eld where economists were asking country-specic questions, such as what was the impact of railways in the United States? The answer could be found by comparing the United

Introduction

States as it actually was with a counterfactual United States without the railroad. Of course, such an approach implied that the researcher had to commit to some theoretical model, because otherwise the counterfactual comparator would remain forever unobserved. A lot of cliometrics thus involved theory with numbers, calibrating models ranging from back-ofthe-envelope partial equilibrium models to full-scale general equilibrium models. By contrast, the New Comparative Economic History instinctively tries to compare actual economies, not with counterfactuals but with each othera dierent solution to the specication problem (Fogel 1967). If they wanted to explore the impact of railroads, its practitioners rst thought would be to exploit the fact that railroads were introduced at dierent times into dierent economies; they would be aware of the possibility that railroads would have had dierent eects in each because of diering geographies, dierent patterns of specialization, dierent political contexts, or other factors. Thus, comparative or counterfactual statements are closely linked to the range of historical experience. By allowing the data to speak in this manner and by avoiding theoretical assumptions that might narrow the range of possible answers, researchers can move away from the tyranny of the Harberger triangle, which always produces the answer small, or the tyranny of partial equilibrium, which ignores the interaction between markets, or any other theoretical tyranny that imposes prior restrictions on the analysis and thus the answers. We would not wish to exaggerate or oversimplify. It is not the case that the New Comparative Economic History never uses counterfactual analysis or computable general equilibrium models. The eld is methodologically catholic, as the chapters in this book illustrate. Traditional cliometrics always made wide use of econometric analysis, relying on comparisons across time, agents, or place, as its 1970s British moniker, econometric history, clearly indicates. Economic history, including cliometric history, has certainly produced comparative work in the past, such as the classic comparisons of French and British economic development (OBrien and Keyder 1978; Crouzet 1991). Clearly, we are dealing with tendencies here. Nonetheless, it seems to us that there has been a discernible shift toward more comparative work in recent years. For example, economic history research and training networks have sprung up in Europe that are devoted explicitly to exposing students to other countries experience. The move away from narrowly national economic history to broader approaches is further symbolized by the recent Cambridge Economic History of Latin America (Bulmer-Thomas, Coatsworth, and

Timothy J. Hatton, Kevin H. ORourke, and Alan M. Taylor

s-Conde 2006), which stresses a comparative approach. The conCorte tributors to the present volume have all been actively involved in comparative economic history projects over the years, ranging from comparative banking history, to comparative famine history, to even broader comparative questions about growth and development. This trend toward comparative economic history was probably inevitable. After all, cliometrics required data, and data tend to be gathered at a national level. Early cliometric eorts were thus largely devoted to collecting these data and analyzing them systematically, starting with the construction of national income aggregates. This enterprise implied, almost by denition, a nation-specic approach to economic history. The barriers to entry for would-be comparative cliometricians were for a long time prohibitively high. Now, however, a broad range of comparative data is easily available, partly as a result of the extraordinarily cooperative tendencies of the worldwide cliometrics community, and partly as a result of collaborative data-gathering projects, such as those devoted to the systematic compilation of comparable wage and price data across as many countries and time periods as possible, illustrated by projects led by Robert Allen, Jan Luiten van Zanden, and Peter Lindert. With the barriers to entry falling, it is small wonder that the number of cliometricians doing serious comparative work has been rising over time. There is a second reason for the growth in comparative work in recent years, and that is the intellectual leadership of Jerey G. Williamson. His inuence across the world has been immense, particularly in regions such as Europe or Latin America, where economic historians have had no alternative but to take seriously the open-economy thinking that has characterized so much of his work. He has been a generous supporter of attempts to build up the cliometric profession in continental Europe, Latin America, Asia, and elsewhere, attending summer schools, giving keynote lectures, collaborating with local economic historians, and providing encouragement to those academic communities. In Europe he has been a much-valued supporter of the European Historical Economics Society, and he was crucially involved in the debates that led to the setting up of the societys journal, the European Review of Economic History, on whose editorial board he has served. The immense regard in which he is held across Europe is reected in the large number of European contributions to this book. In Latin America, Je has worked with numerous scholars, inspired many students in training, and traveled extensively to help develop and encourage new networks, conferences, and research projects, and in doing so, he has played

Introduction

an invaluable role in promoting the recent growth of Latin American economic history as a eld. Through frequent visits to Australia, he has injected new vitality into a discipline that had been in retreat there. And through his insistence on the importance of Asia to global economic history, he has helped to foster comparative work in that region, too. Indeed, he plans to pursue an economic history program beginning in the Philippines. The fact that the New Comparative Economic History exploits such a wide range of experience has boosted interest in countries and regions previously considered to be peripheral to the main story. More than anyone else, Je has carried this message to all corners of the globe. Most of all, as the many references cited here indicate, Je has led by example, helping to shape the intellectual agenda that others are now following. His own career has anticipated broader trends within the profession, its focus moving from U.S. and other single-country studies to globalization and comparative economic history more generally. He has inuenced the New Comparative Economic History directly, through his work with students and collaborators, and indirectly, via his many writings. He is a giant in the eld and in economic history more generally, and this volume is gratefully dedicated to him.
References
Abramovitz, A. 1986. Catching Up, Forging Ahead, and Falling Behind. Journal of Economic History 46: 385406. rtola, L., and J. G. Williamson. 2006. Globalization in Latin America before 1940. In The Be Cambridge Economic History of Latin America. Vol. 2, The Long Twentieth Century, ed. V. s Conde. New York: Cambridge University Bulmer-Thomas, J. H. Coatsworth, and R. Corte Press. s-Conde, eds. 2006. The Cambridge Bulmer-Thomas, V., J. H. Coatsworth, and R. Corte Economic History of Latin America. New York: Cambridge University Press. Clemens, M. A., and J. G. Williamson. 2004a. Wealth Bias in the First Global Capital Market Boom, 18701913. Economic Journal 114: 311344. . 2004b. Why Did the Tari-Growth Correlation Reverse after 1950? Journal of Economic Growth 9: 546. Coatsworth, J. H., and J. G. Williamson. 2004. The Roots of Latin American Protectionism: Looking before the Great Depression. In Integrating the Americas: FTAA and Beyond, ed. A. Estevadeordal, D. Rodrik, A. Taylor, and A. Velasco. Cambridge, Mass.: Harvard University Press. Crouzet, F. 1991. Britain Ascendant: Studies in British and Franco-British Economic History. Cambridge: Cambridge University Press. Fogel, R. W. 1967. The Specication Problem in Economic History. Journal of Economic History 27: 283308. Hatton, T. J., and J. G. Williamson. 1998. The Age of Mass Migration: Causes and Economic Impact. New York: Oxford University Press.

Timothy J. Hatton, Kevin H. ORourke, and Alan M. Taylor

. 2005. Global Migration and the World Economy: Two Centuries of Policy and Performance. Cambridge, Mass.: MIT Press. Kelley, A. C., and J. G. Williamson. 1974. Lessons from Japanese Economic Development: An Analytical Economic History. Chicago: University of Chicago Press. OBrien, P. K., and C. Keyder. 1978. Economic Growth in Britain and France: Two Paths to the Twentieth Century. London: Allen and Unwin. ORourke, K. H., and J. G. Williamson. 1999. Globalization and History: The Evolution of the Nineteenth Century Atlantic Economy. Cambridge, Mass.: MIT Press. Taylor, A. M., and J. G. Williamson. 1997. Convergence in an Age of Mass Migration. European Review of Economic History 1: 2763. Williamson, J. G. 1965. Regional Inequality and the Process of National Development: A Description of the Patterns. Economic Development and Cultural Change 13: 384. . 1974. Late Nineteenth Century American Development: A General Equilibrium History. Cambridge: Cambridge University Press. . 1985. Did British Capitalism Breed Inequality? London: Allen and Unwin. . 1990. Coping with City Growth during the British Industrial Revolution. Cambridge: Cambridge University Press. . 2006. Globalization and the Poor Periphery before the Modern Era: The 2004 Ohlin Lectures. Cambridge, Mass.: MIT Press.