Trade and Economic Growth
TRADE AND ECONOMIC GROWTHflows of trade in nineteenth-century europe: grain, machinery, and other goods
trade developments in the european north and south
trade and growth
tariffs and economic growth
internal trade: direct and indirect effects
inequality effects of globalization
Since Adam Smith (1723–1790) and David Ricardo (1772–1823), trade and economic growth have been regarded as interconnected, and "Smithian growth" has become a fixed term in economic growth analysis. The connection is obvious: if previously nonintegrated regions and countries have cost advantages in different types of products, trade will benefit both sides. For example, Polish grain farmers can produce wheat at a lower cost than English farmers, whereas English workers can produce textiles more cheaply than Polish workers. The "comparative advantages" of the English textile sector furthermore imply that even if an English worker could in principle produce grain at a lower cost than even a Polish laborer, that worker can only spend his time either with grain or with textile production. If the cost advantage in textiles is relatively larger, the worker should thus continue to produce textiles and eat imported wheat.
Smith and Ricardo were convinced that the effects of trade are positive. However, international trade has also been the target of hostile criticism, and the debates on how beneficial the effects of trade were in nineteenth-century Europe are relevant for the contemporary world. The debates focused on the following four hypotheses:
- Growing imports have always provoked countermeasures, or attempted countermeasures, from interest groups who were afraid of losing income and social status when faced with competition from imported labor or goods. For example, British landowners lobbied for and benefited from protectionism in the early nineteenth century while continental European textile manufactures flourished when Napoleon I (r. 1804–1814/15) closed the French ports to British manufactures.
- While market integration had mostly beneficial effects on purchasing power, it is less clear that the contribution to the overall growth of welfare was always positive. For example, integration might in some situations have detrimental effects on health or longevity for parts of the population, which might adversely affect living standards. The "Human Development Index" of the United Nations includes measures such as life expectancy and education, whereas others have used human height as a proxy for the so-called Biological Standard of Living. During the nineteenth century in Europe, height in some cases even declined in some previously remote and nonintegrated regions of Europe.
- It has been argued that trade not only promoted direct growth in ways that Adam Smith had predicted but that it also influenced the development of institutions. Daron Acemoglu, Simon Johnson, and James Robinson have argued that trading cities on the Atlantic coast benefited particularly from the development of institutions that transmitted information more easily and protected and defined property rights. Although their study concentrates on the early modern period, the influence of greater institutional efficiency might have continued well into the nineteenth century. Another indirect effect might also have been the formation of physical and financial capital through the reinvestment of a produced surplus, as is implied by Douglass North's theory of "export-led growth."
- The indirect growth effects of trade may influence income inequality. Eli F. Heckscher and Bertil Gotthard Ohlin have argued that an increased import of goods produced primarily with unskilled labor will reduce unskilled wages, thereby increasing inequality. The opposite holds for the increased export of goods "containing" unskilled labor inputs. Yet while increasing inequality may imply more investments by the rich (because poorer people do not invest much), inequality can also lead to a lack of schooling and human capital formation among the poor, in addition to negative health and crime effects, or even upheaval and social conflict.
These four hypotheses about the relationship between trade and economic growth will be discussed in more detail below. Before addressing them, however, the trade structure of Europe in the nineteenth century needs to be addressed: Which countries concentrated on which export staples? How large were the net exports of particular items that can inform us about the country's growth prospects? Which imports and exports restored the trade balance (taking into account that national deficits were modest in nineteenth-century Europe, at least if we interpret colonial military activities as "service exports," which is surely debatable)?
Who traded which commodities in nineteenth-century Europe? To answer this question, we must concentrate on a small number of commodities, and only the principal exporters and importers can be taken into account. A German survey published by the German Imperial Statistical Yearbook (1915) gives an overview of trade in the pre–World War I period. Two commodities, or rather commodity groups, are particularly interesting for European trade history: "grain" (more precisely: grain, rice, and flour) and machinery.
Grain had been a major trading commodity from the Middle Ages onward. It is self-evident that grain, rice, and flour are crucial for human nutrition. In the nineteenth century, grain imports were responsible for sustaining the extremely high population density of some of Europe's rapidly developing regions. The biggest exporters of grain, rice, and flour in 1913 were Russia, the United States, and British India (which included today's Pakistan, Bangladesh, and Burma). Much smaller quantities were exported in 1913 by Canada, Romania, and Australia. Especially Romania, with a population of less than one tenth of the Russian population, exported remarkable quantities per capita. Argentina was also an important grain exporter.
Not surprisingly, the biggest importer of grain was the United Kingdom. Britain was thus not only the workshop of the world but also consumed most of the traded grain. The second-largest importer was Germany, whose export of industrial goods was growing strongly at the time. However, the German East (and the Polish parts of the empire) were long-established grain exporters, so that Germany's grain-trade statistics were more balanced in effect, with internal trade playing an important role as well. We also need to take into account that the United Kingdom and Germany had the largest populations among the large grain consumers. On a per capita basis, imports were in fact very high in the Netherlands and Belgium as well. Italy imported a substantial total amount of grain, but not as much per capita. Other modest grain importers were Switzerland, France, Denmark, and Norway. Most of these importing countries were among the early industrializers, whereas most exporters tended to be late developers (except for the United States). Interestingly, there were also a number of net grain importers among the less developed countries (LDCs) in 1913, such as China, Sri Lanka ("Ceylon" at the time), and Indonesia. The same applies to Japan and the British Straits Settlement (mostly Singapore and Melaka), which counted as LDCs in 1913.
During the period following the "grain invasion," a term coined by Kevin O'Rourke, the dramatic decline of transportation costs for grain from the New World and Black Sea area caused enormous shifts in production and land revenues. The New World and Russia now accounted for a large share of the total grain trade, whereas a much larger amount had been produced locally before.
Net export values of machinery are important to examine because machinery was one of the most human capital–intensive (that is, skill-intensive) commodities being traded at large quantities during the period. The United Kingdom remained the world's export champion in the prewar period, with 6.4 million Marks of net exports. Germany had heavily reduced the United Kingdom's lead in the decade before and almost reached its competitor with 6.0 million Marks in machinery exports. In a similar vein, the United States had made its mark as a strong new force producing machinery for many different purposes. Switzerland—with its much smaller population—reached only about 5 percent of the net export value of each of the big three machinery exporters. It is also interesting to look at the principal importers of machinery because those countries were using imports to build up a physical capital stock during the period. The world's leading importer of machinery was the Russian Empire. Canada and France had approximately half of the volume of Russia's imports, making them also significant importers at the time. Next followed Australia and Austria-Hungary, whereas the net machinery imports of Italy, Mexico, and Belgium were only modest in 1913.
To summarize, the United Kingdom, Germany, and the United States were the leading exporters of machinery in the nineteenth century. In contrast, the major importers were spread more evenly among the richer and developing nations in 1913. Grain imports tended to be the mirror image of machinery exports, with the important exception of the United States, which exported both grain and machinery in substantial quantities.
This discussion has concentrated on the two commodities that are of particular interest for assessing economic development. What has not been discussed, however, is the value of the most important import and export goods. By far the most valuable export commodity in 1913 was cotton from the United States, which accounted for 23 million Marks, or about three times the value of U.S. net grain exports. In second place came British coal, with Brazilian coffee ranking third. Of course, these goods were much more dependent on natural conditions such as resources and climate than were grain and machinery, which could in principle be produced in any location that had sufficient population density and skill levels.
The major European trading nations were undoubtedly Great Britain, Germany, France, and other northwest European countries. Yet the development of trade also played an important role for countries in the far north and south, and it was only in the eastern European regions that lacked access to rivers or coasts that international trade might have been somewhat less important.
The grain exports of the Russian Empire, eastern Germany (including today's Poland), and Romania have already been discussed. In addition, wood and butter exports played an important role for the far north of Europe, whereas fruit, olive oil, and other horticultural exports came from the Mediterranean. Similar to the grain invasion that shook all of Europe, important globalization events occurred in the South, as José Morilla Critz, Alan Olmstead, and Paul W. Rhode have argued: during the 1880s Mediterranean fruit exports dominated European and U.S. markets, but fruit production in California, Florida, and similar regions of the New World started to become competitive in the following three decades. California farmers first captured large slices of their home market, aided by U.S. protectionist policies and natural disasters in Europe such as the phylloxera plague that destroyed two-thirds of the European grape production in the second half of the nineteenth century. By the end of the century, U.S. products were even competing with Mediterranean fruits in northern Europe. In the 1870s, for example, the important Spanish raisin exporting area of Málaga exported nearly 60 percent of its production to the United States. In the 1890s, this share had fallen to 10 percent. Initially, this decline was caused by phylloxera, but later on, the market opportunities that had been lost to the California competitors played the major role. Since the Málaga region also had the highest emigration rates in Spain, some authors have concluded that a causal relationship existed between lost export markets and emigration.
In contrast to raisins, citrus exports were in general a success story for Mediterranean agriculture. Oranges from Spain and lemons from Italy became famous in many parts of the world. However, American competition grew in these fruit markets as well, and U.S. protectionism posed further problems. In the case of lemons in particular, the Italians had to redirect their trade to northern Europe, after performing vast advertising campaigns there. In 1907–1913, Germany and Austria-Hungary became the most important recipients of Italian citrus fruits.
Greek exporters specialized in currants and even benefited from the phylloxera catastrophe at first, since they could sell their produce to French winemakers, who used it as a temporary substitute for their own grape production. After the recovery of grape production, however, the French government in 1892 imposed high tariffs, which had disastrous consequences for the Greek economy. The plum and prune exports of Serbia and of Bosnia and Herzegovina, as well as the fig and raisin exports of the Turkish west coast, experienced a similar development: a promising start, followed by a struggle against protectionism and competition with California and other New World regions (such as Australia).
What is particularly interesting about this phenomenon is that we would not expect high-wage areas in the New World to be able to compete successfully with the low-wage areas of southern Europe in the seemingly labor-intensive horticulture. Critz, Olmstead, and Rhode, on whose research this section draws heavily, report that one acre of land requires only 9 man-hours of labor to grow wheat, whereas 286 hours are required to grow lemons, for instance. The output value varies drastically as well, of course. The authors have calculated that the value of output per man-hour on U.S. fruit farms was in fact very similar to that of other U.S. farms in the early twentieth century. Moreover, California farmers were quite successful in using modern packing techniques, brand names, and scientific techniques to overcome blue mold and similar production problems. In this way, they were able to outbalance higher labor and transportation costs compared with their major competitors.
The effects on southern European exports were sometimes dramatic. More successful development of the fruit markets could have resulted in dynamic, export-led growth engines in many southern European regions, as well as in the reinvestment of profits into those poor regions. This remains true even when considering that the total value of production of horticultural goods was certainly lower than that of other agricultural goods (mainly for the domestic market). Another important point here is that the majority of consumers clearly benefited from the increased competition.
Nonetheless, two of the three markets discussed in more detail above share some common characteristics that suggest some answers to our initial question: Why was there opposition to trade, if trade increased welfare? In both the European grain market and the Mediterranean fruit market, decisive changes occurred in the nineteenth century. The "grain invasion" and the "fruit invasion" produced a substantial number of losers who tended to be politically well-organized. Even if on average, the growth in trade surely added to European prosperity, the fate of the fruit farmers of the Mediterranean and the eastern German grain-farming nobility gave rise to political forces that influenced the economic history of their respective countries in crucial ways.
What follows is an empirical discussion of Smith's and Ricardo's expectation that trade would trigger economic growth in nineteenth-century Europe. This issue is difficult to resolve because comparative evidence on trade shares in the early nineteenth century still contains a large margin of error. Moreover, convergence effects must be taken into account, since a country with a high initial trade share (such as the Netherlands) would probably increase its trade shares at a lower rate than any newcomer. In addition, a thorough analysis would require taking into consideration all other potential growth determinants such as human and physical capital growth, institutional design, geography, and political development as well as alternatives to commodity trade, for example, the exchange of capital (foreign investment) and population (migration). Here, our aims are more modest. We have taken Paul Bairoch's rough estimates of export shares for the early and late nineteenth century and compared them with Maddison's GDP estimates for 1820 and 1913, combined with an investigation of individual cases. In the following, we distinguish trade shares (imports plus exports per GDP) from export shares (only exports per GDP).
Even with all those caveats, some important facts emerge clearly. First, the Netherlands exported the highest share of their GDP in 1840, followed by a number of relatively small countries such as Norway and Denmark. Smaller countries often had larger trade shares because country-size is negatively correlated with the amount of production items passing borders. However, in early-nineteenth-century Europe, there were also many smaller countries with low trade shares while the United Kingdom displayed one of the highest shares.
Between the early and late nineteenth century, export shares grew for all countries. Overall, they ranged between about 1 and 9 percentage points higher in 1910 than in 1840, with typical values of 1 to 3 percent. At the same time, GDP per capita grew substantially in all countries. Even the inhabitants of slowly growing economies such as Portugal, Russia, and Greece had become richer by roughly $500–1000 per year before World War I (measured in 1990 Geary-Khamis Dollars, a standard measure for making purchasing power comparable across time). In many cases, this meant a doubling or tripling of real incomes within less than a century. Lastly, the Swiss population had acquired four times more purchasing power in 1913 than in 1820.
Based on those numbers, it is not implausible to assume that nineteenth-century growth was caused to a large extent by "Smithian" trade effects. Yet was it the case that countries with higher integration into trading networks also achieved more growth? As the example of the Netherlands, with a high export level but only slightly above-average growth rates indicates, export levels probably had no impact on subsequent growth performance. Even if additional explanatory variables were taken into account, this result would not change significantly.
It is clear that countries with disproportionately large export shares at the end of the period (i.e., irrespective of the initial level) also experienced much higher increases in national income per capita. Belgium, the United Kingdom, Denmark, Switzerland, France, and Germany were examples of the strong and positive development of both variables. On the other hand, the small export-share increases of Portugal, Russia, and Greece went hand in hand with an only modest increase in income. Again, we would like to emphasize that we are not looking at growth rates in percentages here. Moreover, this analysis cannot reveal the direction of the causality involved: Did increasing trade shares cause income growth, or did increasing per capita–production (which equals income per capita) correlate with the need to export a higher share of production? Hence, from this descriptive analysis, we can only conclude that a higher export share correlates empirically with higher additional income. Overall, both export shares and income per capita grew in all European countries over the nineteenth century, yet while the increase was most significant in Belgium and the United Kingdom, it was much less so in southern and eastern Europe.
Bairoch, O'Rourke, and Jeffrey Williamson have given excellent overviews of trade policy. They
describe in detail the history of ideas that influenced the debate about free-trade versus protectionist policies. What they found was that most countries were certainly not free-traders in the early nineteenth century, except for perhaps the Netherlands, Denmark, Switzerland, and Portugal. In England, an intensive discussion was waged before midcentury on whether the country should abolish its protectionist policies against grain imports. Over time, free-traders influenced by Ricardo and the "Manchester" liberals convinced enough political decision-makers in their favor, so that the famous protectionist Corn Laws were finally dropped in the 1840s (after having been gradually reduced earlier). Subsequent negotiations between the United Kingdom and France, as well as the general economic upturn of the 1860s led to a wave of free-trade policies in the 1860s and 1870s. However, the dramatic decline in transportation costs and the high productivity of New World farmers stimulated a rebound in protectionism in many continental European countries against grain imports while the United States acted in a strongly protectionist manner against industrial imports. However, protectionist measures were not strong enough to hinder the international integration of many markets. The fall in transportation costs simply outweighed most such measures.
The integration of commodity markets within Europe can be measured by the decline in price gaps. For example, the price gap between Odessa in Ukraine (one of the major grain exporting regions) and London decreased from 40 percent to virtually 0 percent between 1870 and 1906. Similarly, the price gap of Swedish wood between its country of origin and England fell from 155 to 70 percent. In general, O'Rourke and Williamson found that the strong increase in commerce and trade in the nineteenth century was mainly the result of declining transportation costs and only to a much lesser extent of more liberal trade policies. This stands in sharp contrast to the boom in international trade in the second half of the twentieth century, when trade policies accounted for most of the trade-generating effects, whereas the decline of transportation costs due to technological improvement was only modest.
The expansion of export shares correlated with additional per capita income in nineteenth-century Europe. Does this imply that protectionism was bad for growth? To answer this question, we must first consider that neither protectionism nor its inverse, the "openness" of countries, correlates perfectly with trade shares. For example, Germany and France became grain protectionists in the late nineteenth century, yet their export shares still continued to increase substantially, since economic forces were simply much stronger than the respective political countermeasures.
The growth effects of protection (as opposed to free trade) have been studied in great detail for the contemporary world. Free-traders expect welfare effects if the external effects (i.e., side effects) of protection are not important. However, if "infant industries" cannot develop because the industries of industrialized countries have already achieved specific knowledge and cost-efficient production methods, then the protection of those industries on the side of the newcomers could stimulate growth. Empirically, most studies for the late twentieth century concur with Jeffrey D. Sachs's conclusions that free trade rather then protectionism was a successful device for growth. However, for the late nineteenth century, O'Rourke found that protective tariffs in fact had a positive influence on growth rates, even after allowing for changes in the capital-labor ratio, land-labor ratio, initial income level, schooling, and other country-specific characteristics. His analysis included seven European countries, the United States, Australia, and Canada for 1870–1913 and indicates that the protectionist United States grew fast, whereas the free-trading United Kingdom grew only modestly during the period under consideration. In addition, O'Rourke found that Germany was not as protectionist as studies focusing particularly on the grain trade have suggested. Even if the United States and the United Kingdom are omitted as extreme cases, the positive relationship between protectionism and growth is confirmed for the nineteenth century. A possible causal mechanism, O'Rourke suggests, was that tariffs caused a declining share of the labor force to be employed in agriculture, especially in the New World, where tariffs benefited industrial production.
It should be noted, however, that our comparison of export shares and GDP growth refers to long-term growth over almost a century, whereas O'Rourke's study explores the effect of protection levels on annual growth rates in the period 1875–1913, controlling for a number of other variables as well.
So far we have concentrated mainly on international trade, since external exchange always attracts the highest attention from policy makers and the general public. More important in terms of value, however, was internal trade that did not cross national boundaries in nineteenth-century Europe. Trade within countries increased dramatically at the time because transportation facilities experienced a veritable revolution. The railway network not only grew dramatically in size, but the railway system also became ever more refined. Moreover, toward the very end of the nineteenth century, refrigeration wagons allowed the transportation of perishable goods. For the first time in human history, it became possible to provide large urban populations with food that was as healthy as the food consumed in the countryside. Milk, for example, had previously not been transportable to large urban agglomerations at reasonable costs, although urban centers were in dire need of protein in particular because of the high rates of disease. Before the
introduction of refrigeration wagons and a dense railway network in general, there had always been an "urban penalty" of health and nutritional qualities, with the result that urban dwellers often lived shorter lives than their contemporaries in the countryside. The change in the "transportability of health inputs" was perhaps the most dramatic and decisive development of nineteenth-century trade history.
In the countries that invested heavily in this new transportation technology—not least because of adequately low temperatures in their territories, as opposed to the Mediterranean region, which followed about half a century later, for example—a stagnation of life expectancies and heights that had lasted for millennia finally ended in the second half of the nineteenth century. Clearly, this development was accompanied by progress in the application of hygienic and medical knowledge, although it is not easy to disentangle the individual contribution of each. Thomas McKeown, for example, has argued that nutrition played a much stronger role in increasing health and longevity, whereas the impact of medical knowledge set in much later. McKeown did not address the issue of the tradability of perishable foods, but this is probably the heart of the matter. We would, for example, not expect a decline in hygiene and health knowledge with the advent of railway stations in villages, but agricultural workers with lower incomes experienced significant downward trends in health. This has been analyzed using anthropometric techniques that take human height as a proxy for nutritional status and health (see, for example, Komlos and Baten). Human stature is mostly determined by the quality of nutrition in the first three years of life, and by the disease environment. When a railway station was opened in the countryside nearby, for example, farmers could sell their perishable products to urban consumers with high purchasing power. In earlier times, the local poor population in the countryside had often been able to consume healthy food at very low prices or even for free. If they were involved in the production of butter, for example, they were often allowed to drink the remaining milk. Others were allowed to eat offals after local cattle were slaughtered. All such nonmarket entitlements disappeared with the arrival of the railway, so that poor rural children lost their previous height advantage while the heights of urban children increased. In the long run, the health and longevity of poor agricultural laborers increased as well. In the short run, however, there were losers from the modernization process, and their experiences should not be overlooked when examining the dramatic trade and income increase in nineteenth-century Europe.
While it is clear that landowners in western Europe lost income when vast quantities of grain arrived from the New World, the effect on workers was hotly debated at the time. O'Rourke describes the nineteenth-century debates on the influence of cheap grain imports on workers' living standards. Karl Marx argued that rural depopulation in Ireland was reinforced by the switch from tillage to pasture, as Ireland focused on butter exports in response to globalization. Since pasture was less labor-intensive, fewer agricultural workers were employed. Benjamin Disraeli (1804–1881), a leading English politician, was similarly concerned about the decreasing demand for rural workers in England caused by the arrival of cheap grain from the New World. He argued that this negative labor-demand factor was more important than the positive cost of living effect for workers. A coalition of Manchesterian liberals and socialist workers opposed this view. They believed (correctly) that the cost of living effect dominated in most industrialized nations. In western Europe, the grain invasion had mostly egalitarian effects on average, reducing the previously very high income inequality. In some grain regions, however, agricultural workers had to bear at least the cost of migrating to industrializing cities. The costs of migration were particularly high if it implied the crossing of national borders (Ireland to England or the United States, Poland to Germany, and so forth).
This essay first discussed the relationship between trade and economic growth in nineteenth-century Europe. It began with a description of the net exports of two important commodities, machinery and grain. Machinery production requires a high skill level, and countries that specialized early on in its production and export became rich countries—and continued to be so even after losing world wars and colonial empires in the twentieth century. In contrast, specialization in grain did not require large investments in skills. It has been argued that the export of primary goods led to deindustrialization and poor growth in the long run. This was not the case for all grain exporters in 1913. The United States, Canada, and Australia proved success stories in the long run. Eastern Europe and Argentina might have grown less than expected in the twentieth century, but there were also other reasons for this.
We then analyzed whether countries with strongly expanding trade shares experienced the relatively highest increases in income, an assumption that is supported by the available data. However, the question of causality cannot be resolved at the current state of research, and there is evidence that protectionism may even have been positively correlated with growth, at least in the short run from 1875 to 1913.
From the point of view of trade history, it is particularly interesting to assess the reasons for the frequent protest against, and opposition to, globalization and international trade in the nineteenth century. The unfulfilled hopes of the Mediterranean fruit farmers was a case in point, as they struggled with the growing competition from California, a region they might never have heard of before. Even the internal trade expansion of this period produced many winners but also some losers, since previously isolated farmers and farm workers lost their health advantage. Yet a dramatic health revolution occurred in the cities: for the first time in the history of mankind, the inhabitants of densely populated metropolises could live long and healthy lives, which could at least in part be accredited to the nineteenth-century trade expansion that made healthy nutrition possible.
Finally, the trade history of the nineteenth century produced some "rich losers." For example, the
British gentry lost a substantial share of their rents when grain protectionism was abolished around midcentury. The Russian nobility may have been compensated by export revenues, but the German landed aristocracy felt severely threatened by the "grain invasion," a development that may have increased their willingness to start World War I—an extreme example of resistance to of globalization, to be sure.
Acemoglu, Daron, Simon Johnson, and James Robinson. "The Rise of Europe: Atlantic Trade, Institutional Change, and Economic Growth." NBER Working Paper 9378, 2002.
Bairoch, Paul. "European Trade Policy, 1815–1914." In The Cambridge Economic History of Europe, edited by Peter Mathias and Sidney Pollard, vol. 8, pp. 1–160. Cambridge, U.K., 1989.
Critz, José Morilla, Alan Olmstead, and Paul W. Rhode. "'Horn of Plenty': The Globalization of Mediterranean Horticulture and the Economic Development of Southern Europe, 1880–1930." Journal of Economic History 59, no. 2(1999): 319–352.
Komlos, John, and Joerg Baten, eds. The Biological Standard of Living in Comparative Perspective. Stuttgart, Germany, 1998.
Maddison, Angus. The World Economy: A Millennial Perspective. Paris, 2001.
McKeown, Thomas. The Modern Rise of Population. London, 1976.
North, Douglass C. "The New Institutional Economics and Third World Development." In The New Institutional Economics and Third World Development, edited by John Harriss, Janet Hunter, and Colin M. Lewis, 17–26. London, 1995.
O'Rourke, Kevin, "Tariffs and Growth in the Late 19th Century." Economic Journal 110 (Aprtil 2000): 456–483.
O'Rourke, Kevin, and Jeffrey Williamson, Globalization and History: The Evolution of a Nineteenth-Century Atlantic Economy. Cambridge, Mass., 1999.
Sachs, Jeffrey D., and Andrew Warner. "Economic Reform and the Process of Global Integration." Brookings Papers on Economic Activity 1(1995):1–118.