Jump to ContentJump to Main Navigation
Trade and PovertyWhen the Third World Fell Behind$

Jeffrey G. Williamson

Print publication date: 2011

Print ISBN-13: 9780262015158

Published to MIT Press Scholarship Online: August 2013

DOI: 10.7551/mitpress/9780262015158.001.0001

Show Summary Details
Page of

PRINTED FROM MIT PRESS SCHOLARSHIP ONLINE (www.mitpress.universitypressscholarship.com). (c) Copyright The MIT Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in MITSO for personal use (for details see http://www.mitpress.universitypressscholarship.com/page/privacy-policy). Subscriber: null; date: 21 June 2018

Better Late Than Never: Industrialization Spreads to the Poor Periphery

Better Late Than Never: Industrialization Spreads to the Poor Periphery

(p.199) 12 Better Late Than Never: Industrialization Spreads to the Poor Periphery
Trade and Poverty

Jeffrey G. Williamson

The MIT Press

Abstract and Keywords

This chapter examines the spread of industrialization to the poor periphery, focusing on Brazil and Mexico. It shows that industrialization of Brazil and Mexico was very impressive between 1870 and 1913, especially compared with the rest of the poor periphery; the industrial liftoff started well before 1890; Mexico recorded the most rapid industrialization before 1901; and after 1900, Argentina, Chile, and Brazil all recorded rapid industrial growth well in excess of their GDP growth. What makes this performance all the more striking is that it was preceded by seven decades of de-industrialization. The chapter also argues that the Latin American findings generalize to other parts of the poor periphery.

Keywords:   Mexico, Brazil, Latin American, industrialization, de-industrialization, poor periphery

12.1 When Did the Industrial Revolution Start to Penetrate the Poor Periphery?

There are some parts of Africa and Asia where modern factories are rare even today, where the majority still till the soil with primitive techniques, and where only the minority live in cities. But in some parts of the poor periphery modern industrialization started more than a century ago. Latin America had two emerging industrial leaders in the late 19th century—Brazil and Mexico; Asia had four—Bengal, Bombay, Japan, and Shanghai; and the European periphery had at least three—Catalonia, the north Italian triangle, and Russia. Why the late 19th century and why these places?

No doubt the answer is as complex as any question dealing more generally with the causes of modern economic growth, and no doubt any answer should include fundamentals like culture, geography, institutions, and good government. There is, of course, one simple explanation that would appeal to the growth theorist: As the great divergence took place, labor became increasingly expensive in the industrial core relative to the poor periphery. Thus the poor periphery became increasingly competitive in labor-intensive manufacturing. But what about the relative price of manufacturing output? What about trade and exchange rate policy? What about the timing and location of early industrialization in the poor periphery? Here, global forces have a chance to shine.

This chapter dwells on Brazil and Mexico to help identify the global-industrialization connection, but the penultimate section argues that the Latin American findings generalize to other parts of the poor periphery.

(p.200) 12.2 Measuring an Industrial Liftoff in Latin America

It has long been appreciated that industrialization started long before the 1930s in Latin America, indeed even before World War I. Ezequiel Gallo (1970) made the case for Argentina almost forty years ago when dependency theory and export-led growth were the dominant development paradigms. Warren Dean (1969) made the case for Brazil about the same time, and Henry Kirsch (1977) did so for Chile. Two decades later Stephen Haber (1989, 1990) made the case for Mexico. What is missing from this pioneering literature, however, is an explicit assessment of the timing and pace of industrialization in Latin America compared with the rest of the poor periphery, as well as a comprehensive assessment of its causes.

Since textile production dominated late 19th-century manufacturing, we start there. Imports satisfied a significant part of Latin American home textile demand in the 1870s. Chapter 8 reported that Mexico, an industrial leader in Latin America, imported 40 million square meters of cloth in 1879, compared with 60 million square meters produced domestically. Thus domestic producers were able to claim 60 percent of the local market, a fairly big number for a country that had been flooded with cheap, factory-made European textiles for almost a century. This certainly was not true for all of Latin America: indeed the local market share claimed by Colombian producers was only 10 percent in the 1870s (Gómez Galvarriato and Williamson 2009: table 2). Furthermore Mexico stacks up well against other parts of the poor periphery (table 5.3). India’s domestic textile producers claimed only 35 to 42 percent of their local market in 1887, the latter much lower than Mexico’s 60 percent in 1879, a decade earlier. Deindustrialization forces in the first three-quarters of the 19th century had been even more powerful in the Ottoman empire where the local textile industry’s market share of domestic demand was only 11 to 38 percent in the early 1870s. In Dutch Indonesia, the local textile industry’s market share of domestic demand was 38 percent in 1870, about the same as India and the Ottoman empire. Thus, despite the importance of foreign imports in the home market, the Mexican textile industry was doing fairly well by the 1870s, at least compared with the rest of the poor periphery. Mexico was poised for an industrial liftoff from a higher platform than most of the poor periphery.

More to the point, however, is that some parts of Latin America underwent significant industrialization between the 1870s and World War I. By 1907 Brazil’s domestic textile industry had carved out an impressive share of its local market, about 65 percent, while Mexico’s share had risen to (p.201) almost 78 percent in 1906 to 1908 (Gómez Galvarriato and Williamson 2009: table 2). Of course, not all of Latin America was industrializing: for example, Argentina’s textile industry claimed only 15 to 18 percent of the home textile market in 1913.1 Textiles are light industry, the sort of laborintensive manufacturing that every early industrializer exploits first, before moving farther up the skill and technology ladder. Thus it is no surprise that the local market shares for heavy (capital- and skill-intensive) industrial product groups were much smaller even among the industrial leaders. Yet even these higher tech sectors expanded at the end of the period: the domestic producers’ share of the Mexican iron and steel market rose from 6 to 28 percent between 1903 and 1911, while that for coke rose from 17 to 47 percent (Gómez Galvarriato and Williamson 2009: table 2).

These facts offer strong support for the view that the Latin American industrial leaders did indeed experience an industrialization liftoff2 before 1913, and certainly long before the 1930s and the initiation of its antiglobal and anti-market ISI policies. The harder questions, however, are exactly when the liftoff took place, where it took place, and why.

Stephen Haber dates the beginning of the Mexican industrial liftoff in the late 1880s. In the decade that followed,

the industry more than doubled in size. By 1911, the industry had grown an additional 50 percent. Estimates of total factor productivity (TFP) growth…indicate increases of between 1.5 percent…and 3.3 percent…per year. Labor productivity grew even faster…between 3.0 and 4.7 percent per year (Haber 2002: 7–8).

Such growth rates meant that Mexican textile producers had displaced most imported cloth by 1914 (Haber 2002: 11). Furthermore by about 1910 the Mexican textile industry had a machine index per worker 77 percent that of Britain, suggesting that Mexican technology had been partially catching up on the world’s light industry leaders over the previous four decades (Clark 1987: 148 and 152). Indeed Aurora Gómez Galvarriato (2007) has shown that one leading Mexican firm had achieved the labor productivity levels of similar sized textile firms in the United States and the United Kingdom at the beginning of the 20th century.

Table 12.1 and figure 12.1 report evidence on exports to Latin America from the United States and the United Kingdom that better identify Latin American industrial leaders and the timing of their industrialization experience in the decades before World War I. These manufacturing intermediates were the main suppliers of energy and capital goods to Latin American industry, so we can assume that manufacturing machinery, iron and steel, and coal imports (all in constant US dollars) are good proxies for relative (p.202)

Table 12.1 Growth rates per annum in intermediates and capital goods imports from the United States and United Kingdom, 1871 to 1911 (in US$)






Unweighted average























































Source: The import data are reported US and the UK exports of iron and steel, coal, and manufacturing machinery in 1890 to 1914 in US dollars. Gómez Galvarriato and Williamson (2008: table 3).

Better Late Than Never: Industrialization Spreads to the Poor Periphery

Figure 12.1 Index of coal, iron, and steel and of machinery imports in constant US dollars (1900 = 100).

Source: Gómez Galvarriato and Williamson (2009: fig. 3).

(p.203) industrial growth. The data in table 12.1 and figure 12.1 allow us to compare industrial progress between 1870 and 1914 within the four big Latin American countries—Argentina, Brazil, Chile, and Mexico, each of which exhibited impressive growth over the four decades. True, three of them underwent very high volatility: Argentina, Brazil, and Chile all recorded big booms in the late 1880s to early 1890s and again in the run up to World War I, as well as a big bust in the 1890s (see also chapter 10). Mexico did not undergo this volatility before 1900, but it did undergo a secular slow down from 1902 onward, up to and during the revolution (1910–1920). But the combination of imported machinery, iron, and coal as a share of Mexican gross domestic product (in 1990 US dollars) increased seven times up to 1902!

Now consider the growth rates of those intermediate and capital goods imports in table 12.1 Between 1871 and 1901 this industrial proxy grew almost twice as fast in Mexico than the average for the four (10.5 vs. 5.5 percent per annum), and it grew faster even when the run up to the revolution is included (1871–1911, 7.1 vs. 5.4 percent per annum). In addition the table shows that the liftoff started during the first two decades, not just after 1890 as Haber suggests. Note also that the Mexican industrial index grew more than twice as fast as did its GDP over those four decades up to 19113 (7.1 vs. 3.4 percent per annum), implying the dramatic structural change we have come to expect from early industrial revolutions. Over the four decades following 1871, the industrial proxy grew 1.6 times as fast as its GDP in Brazil (3.9 vs. 2.4 percent per annum), but it grew only a fifth faster than GDP in Chile (4 vs. 3.3 percent per annum), and it barely grew faster at all in Argentina (6.7 vs. 6 percent per annum).

To summarize, industrialization of the Latin American economic leaders— Brazil and Mexico—was very impressive between 1870 and 1913, especially compared with the rest of the poor periphery; the industrial liftoff started well before 1890; Mexico recorded the most rapid industrialization before 1901; and after 1900, Argentina, Chile, and Brazil all recorded rapid industrial growth well in excess of their GDP growth. What makes this performance all the more striking is that it was preceded by seven decades of de-industrialization. It looks like the 1870s recorded a big turnabout for Brazil and Mexico, from de-industrialization to re-industrialization.

12.3 Some Potential Explanations for the Latin American Industrial Liftoff

What explains the timing and the pace of the Latin American industrial liftoff? Consider five leading candidates.

(p.204) First, we now know that Latin America was far more protectionist than anywhere else in the late 19th century (Coatsworth and Williamson 2004a, b). Indeed Latin America had the highest tariffs in the world from the late 1880s onward,4 and before the late 1880s only the United States had higher average tariff rates. Still the typical average tariff rate in Latin America did not rise between 1870 and 1914, and it is changes in protection that matter.

If average levels of protection did not rise over the four decades, what about the protection of manufacturing, and what about the effective rate of protection there? Here the evidence is much more supportive of an active role for pro-industrial policy. The leader of the policy pack, Mexico, adopted far more coherent and consistent pro-industrial tariff policies under the Porfiriato regime. Policy makers in late 19th-century Latin America were certainly aware of infant industry arguments (Bulmer-Thomas 1994: 140), but tariffs were not used specifically or consciously to foster industry in Mexico, or elsewhere in Latin America, until early in the 1890s, a decade or two after the industrial liftoff began. Mexico increased its average tariff rates sharply in 1890 and graduated from having the lowest levels of protection among major Latin American countries to having levels similar to the rest. But this policy transformation did not just entail a change in average tariff levels. It also entailed a profound change in the tariff structure, a change that served to raise the effective rate of protection for manufacturing.5 Edward Beatty (2000) and Graciela Márquez (2002) have shown that the 1880s and 1890s saw the introduction of a modern pro-industrial policy in Mexico, including a rational structure of protection. Haber agrees with Márquez and Beatty:6

In 1891 Mexico was using tariffs to protect the cotton textile industry which perhaps would have otherwise been uncompetitive. This meant high tariffs on competing goods and low tariffs on inputs. The tariff on imported cloth tended to be twice that of the tariff on imported raw cotton. The result was an effective rate of protection that varied from 39 to 78 percent. (Haber 2002: 16)

This explanation helps account for the fast Mexican industrial liftoff compared with the rest of the periphery, especially given that most of Asia, Africa, and the Middle East did not have the autonomy to pursue proindustrial commercial policies. We also know that the pro-industrial Mexican policy led the rest of the autonomous Latin American republics by a decade or two: the policy was followed with a lag by Brazil and Chile a little later in the 1890s, and by Colombia in the early 1900s (Márquez 2002; Coatsworth and Williamson 2004a; b). Argentina failed to follow suit, a policy that helped put the brakes on industrialization there.7

(p.205) A more rational tariff policy clearly increased its support for local industry in Mexico, Brazil, and Chile from the 1890s onward. However, the industrial liftoff started earlier, in the 1870s. Thus we need to search for additional explanations for Latin American re-industrialization, especially for the leaders, Brazil and Mexico.

Consider four other influences that might have contributed to the industrial liftoff in Latin America, perhaps even more so than tariff policy. The first is world prices and the terms of trade. There was a big secular change in world relative prices facing the Mexican economy after the 1870s, a change that no longer penalized local manufacturing. Chapter 8 argued that a good part of the exceptionally modest Mexican deindustrialization experience in the half century before 1870 was due to an exceptionally modest terms of trade shock compared to the rest of Latin America, Asia, the Middle East, and even the European periphery. It appears likely that the same was true of the half century after 1870, but in the opposite direction: Latin America’s terms of trade fell earlier and faster than anywhere else in the periphery, especially Mexico. Since it imported manufactures, the relative price of industrial goods rose in home markets, favoring industry. Second, there was also an acceleration in total factor productivity growth in one key export activity in Latin America, mining. This served to contribute to the fall in the net barter terms of trade, but given a price elastic demand facing silver, copper, and other metals, it also served to raise the income terms of trade,8 fostering the import of relatively cheap capital goods and manufacturing intermediates (including coal), thus favoring an industrial liftoff. These productivity events were unusual for the poor periphery at that time, including much of Latin America. It gave Mexico an edge. Third, there is the depreciation of local currencies to consider. The biggest real exchange rate depreciation between 1870 and 1913 took place in Brazil and Mexico, forces that must also have contributed to their impressive industrialization performance.

Finally, economic success in the industrial core raised the cost of labor there relative to the poor periphery, eventually offering the later an industrial escape from resource specialization. Thus Latin American industry could have become more competitive to the extent that it faced weaker upward pressure on the nominal wage due to slower overall economic growth, but in addition weaker upward pressure on food prices could have strengthened their competitiveness. The latter can be explained by elastic food import supplies and domestic market integration by the railroads.

(p.206) 12.4 No More Dutch Disease in Latin America?

Most of Latin America faced a big secular change in world relative price trends over the four decades before 1913, a change that, as we have noted above, no longer penalized local manufacturing. The region’s terms of trade reached a secular peak in the mid-late 1870s, leveled off up to the early-mid 1890s, after which it fell far more dramatically up to the early 1900s than almost anywhere else in the poor periphery. Between 1870–1874 and 1909–1913, its terms of trade actually fell by 10 percent, this after rising by 174 percent over the seven decades between 1800–1804 and 1870–1874 (figure 3.6; table 3.1)! Only the European periphery underwent a similar collapse in its terms of trade, where the secular peak was even earlier, the 1850s, and the fall even steeper (figure 3.5). In contrast, the rest of the periphery underwent no net fall over the four decades: the Middle East and Southeast Asia underwent a continuous terms of trade improvement from the mid-1870s onward, while South Asia and East Asia underwent no change either way. Which parts of Latin America underwent the biggest changes in terms of trade trend? Consider first the two industrial leaders. Mexico underwent a big fall in its terms of trade after the early mid-1890s: the Mexican terms of trade was cut in half between 1890 and 1902,9 twice as big as the rest of Latin America. Furthermore the primary product price boom in the decade or so before 1913 was very modest in Mexico, and it didn’t come close to recovering the secular peak in the late 1870s and early 1880s. Indeed the Mexican terms of trade fell by more than 37 percent between 1870–1874 and 1910–1913, while it rose by 11 percent in the rest of Latin America. Mexico was exceptional compared with the Latin American average. While Brazil underwent great terms of trade volatility, it hardly underwent any secular change at all in its terms of trade over the four decades (a total rise of only 1.3 percent between 1870–1874 and 1910–1913). This secular stability certainly represented a marked change in world economic conditions after seventy years of rising terms of trade and thus falling relative prices of manufactures. In contrast with these immense price changes facing the two industrial leaders, Argentina and Chile continued its terms of trade boom up to World War I; between 1870–1874 and 1910–1913, Chile’s terms of trade rose by a huge 81 percent, while the figure for Argentina was still a big 21 percent.

In summary, a fall (or no rise) in the net barter terms of trade implied a rise (or no fall) in the relative price of imported manufactures, an event that favored (or no longer penalized) domestic industry. If a rising terms (p.207) of trade caused de-industrialization and Dutch disease in the six or seven decades before the 1870s, it follows that a falling or stable terms of trade after 1870 should have contributed greatly to the impressive Mexican and Brazilian industrialization experience up to 1913.

12.5 The Income Terms of Trade, and Export-Led Growth in Latin America

So far we have assumed that the Latin American republics had no influence over their export or import prices, and thus that their terms of trade was determined exogenously in world markets, an issue that the previous chapter already explored. While commodity exports were only a small share of world exports of any specific commodity for Argentina (maize, wheat), Columbia (coffee, gold), Cuba (sugar, tobacco), Peru (copper), Uruguay (wool, meat), and Venezuela (coffee, cacao), the shares were much bigger for Chile (nitrates, copper), Mexico (silver), and Brazil (coffee, rubber); see tables 4.2 and 11.1. Indeed Edward Beatty (2000) has argued persuasively that Mexican mineral supplies to the world market, especially silver, helped precipitate the big decline in its net barter terms of trade: by flooding the market with silver, Mexico lowered the world price and worsened its terms of trade. Beatty also argues that it was rapid productivity advance in Mexican mining that produced that result, but since the demand for minerals was price elastic, total export values and foreign exchange earnings boomed, creating export-led growth. It also reduced the relative price of imported capital goods, energy sources, and other intermediates in to manufacturing, thus fostering industrial growth.

While a very big decline in its net barter terms of trade (after a long secular boom) was consistent with a very big increase in its income terms of trade (table 12.2), Mexico appears to be the Latin American exception. That is, Beatty’s supply side argument cannot apply to mineral-producing Chile, since its net barter terms of trade did not fall but rather rose mightily. Beatty’s hypothesis also gets only weak support for Brazil where the net barter terms of trade was stable after the long and spectacular pre-1870 boom while the income terms of trade underwent only a modest increase (table 12.2).

12.6 Industrial Competitiveness: Keeping the Lid on Latin American Wages?

The share of the labor force in manufacturing never exceeded 25 percent anywhere in Latin America in 1895.10 This implies that real wages (p.208)

Table 12.2 Income terms of trade growth in Latin America, 1870 to 1913 (1900 = 100)

(1) NBTT 1870–1874

(2) 1870X volume

(3) (1)(2)/100 = INCTT 1870

(4) NBTT 1909–1913

(5) 1913X volume

(6) (4)(5)/100 = INCTT 1913

(7) INCTT growth (%)

NBTT increase




























NBTT decrease




















NBTT no change












Latin America



Sources: A 1870 population weighted average is used for Latin America. The export volume data are taken from Maddison (1989: 140) and (1995: 236). The INCTT = (NBTT)(X), or Xpx/pM, where px is export price and pM import price. The table uses the first equation. Gómez Galvarriato and Williamson (2008: table 6).

(p.209) (w/pc, where pc refers to the cost of living) were determined by labor productivity elsewhere in the economy—mining, agriculture, construction, and services—not in manufacturing. Slow-growing labor productivity in the rest of the economy would have given Latin American manufacturing the advantage of modest upward pressure on per unit wage costs, making it more competitive with North America and western Europe, where the upward pressure was much stronger. It was, of course, the own-wage that mattered to employers in manufacturing, that is, the nominal wage divided by the price of manufactures (w/pM). If the price of foodstuffs (pF) was the central determinant of the cost of living in Latin America, and if pF/pM was falling, we would have another reason to expect Latin American manufacturing to have undergone increasing wage competitiveness compared with foreign firms. Is there any reason to think that pF/pM should have fallen? Yes, and for two reasons: first, to the extent that the post-1870s grain invasion from Argentina, North America, and Ukraine flooded Latin American food deficit regions, where most Latin American manufacturing was located;11 and second, to the extent that newly built railroads brought the cheaper foreign grain to those same urban interior markets (Dobado and Marrero 2005b). The issue is how much?

Mexico recorded the biggest pF/pM decline between 1874–1878 and 1913, 34 percent, and Brazil the second, 21 percent (Gómez Galvarriato and Williamson 2009: fig. 10). Thus the grain-invasion-cum-railroads prediction is confirmed, and furthermore the two fastest industrializing countries recorded the biggest fall in pF/pM.

But did cheaper grains necessarily mean lower nominal wages and thus greater wage competitiveness for local manufacturing? Did the own-wage facing manufacturing (w/pM) fall, at least for the four countries that offer the necessary time series data—Brazil, Chile, Mexico, and Uruguay? It appears that Brazil and Uruguay had the industrial advantage on this score, since the upward pressure on the own-wage in manufacturing was much greater in Chile and Mexico (Gómez Galvarriato and Williamson 2009: fig. 11). Indeed over the period 1870 to 1913 the own-wage in Brazil grew no faster than it did in the United States, the former remaining competitive with the latter on that score at least. By the same criteria, wage competitiveness deteriorated in Mexico (but not nearly as much as Chile). Of course, there were other forces determining competitiveness, but if we are looking for explanations for precocious industrialization in Brazil and Mexico, better wage competitiveness was not one of them.

(p.210) 12.7 Did Real Currency Depreciation Increase Manufacturing Profitability in Latin America?

Depreciation of the domestic currency favors local manufacturing in primary-product exporting countries since it makes imported manufactures more expensive in the local market. Currency appreciation does the opposite. When trading partners have different rates of inflation, the nominal exchange rate must be adjusted to take account of the differential inflation rates, yielding a real exchange rate. The real exchange rate (RER) is yet another force that could have helped account for the timing and pace of industrialization in Latin America before 1913, especially given that Latin America did not rush to join the gold standard club while most European colonies did.12

Figure 12.2 plots the RER for the four big Latin American republics— Argentina, Brazil, Chile, and Mexico. The figure makes it quite clear that manufacturing in Chile must have been greatly disadvantaged by real exchange rate trends since it underwent significant real currency appreciation. Argentina underwent no secular change in its real exchange rate

Better Late Than Never: Industrialization Spreads to the Poor Periphery

Figure 12.2 Real exchange rate trends in Latin America, 1870 to 1913 (1913 = 100).

Source: Gó mez Galvarriato and Williamson (2009: 687–89).

(p.211) between 1884 and 1913, so it could not have been a source of any stimulus to import-competing manufacturing there. However, the real exchange rate did undergo secular depreciation in Brazil and Mexico, and the magnitudes appear to have been big: Mexico underwent a real currency depreciation of 137 percent between the mid-1870s and 1902, when capital goods imports surged. Mexican real currency depreciation for the whole period was an enormous 81 percent. While not quite as big, Brazil also underwent a significant real currency depreciation of 34 percent between 1870–1874 and 1913.

In short, some part of the industrial liftoff in Brazil and Mexico can indeed be explained by a real exchange rate depreciation up to 1913.

12.8 What about the Rest of the Poor Periphery?

To the extent that pro-industrial tariff and exchange rate policy mattered, it seems clear that the best place to look for early industrialization evidence elsewhere in the poor periphery would be where there was policy autonomy over tariffs and currency regimes. Unfortunately, there are few places to look in Asia and Africa since most of it was controlled by the colonists who favored keeping the new economic order in place: that is, they certainly didn’t want to foster manufacturing competition in their colonies. To the extent that world market forces mattered, it seems clear that the best place to look for early industrialization evidence elsewhere in the poor periphery would be where world market forces were driving up the relative price of manufactures in local markets. So where do we see the terms of trade for primary-product exporters dropping early and big, or where in East Asia—whose natural comparative advantage was not in primary products but in labor-intensive manufacturing—do we see the terms of trade soaring? These, after all, would be places where the relative price of manufactures was on the rise, thus favoring early industrialization. Unless these forces were too weak to offset other “fundamental” hindrances to modern economic growth, we should see—at least compared with the rest of the poor periphery—early industrialization there.

Table 12.3 inverts the changing terms of trade (pX/pM) evidence in the poor periphery between 1870–1874 and 1910–1913—except for East Asia—so as to measure changes in the relative price of imports (pM/pX) that we take to be a crude proxy for changes in the relative price of manufactures in home markets. Big negative numbers identify those regions where de-industrialization and Dutch disease forces were still powerful: Argentina, Chile, Indonesia, Egypt, and the Levant are all well above the (p.212)

Table 12.3 Percentage change in the relative price of imports in home markets, 1870–1874 up to 1910–1913

Latin America











East Asia





South Asia





Southeast Asia







Middle East







European periphery









All periphery


Source: Taken from the database underlying Williamson (2008).

Notes: 1870 population weights used to construct “all periphery.” Except for East Asia, the figures are simply the percent change in the inverse of the net barter terms of trade, or pM/pX. For East Asia, it is pX/pM.

(p.213) 1870 population weighted average for the periphery as a whole (a 9.2 percent fall in the relative price of manufactures proxy). Big positive numbers identify those regions where world market forces were giving a powerful stimulus to early industrialization: Mexico in Latin America; China and Japan in East Asia; Portugal, Russia, and Spain in the European periphery; and three Asian surprises, Ceylon, Siam, and the Philippines. The rest show no powerful world market forces one way or the other.

Since the determinants of industrialization are complex, it is hardly surprising that the correlation between changing world market conditions and industrialization in the poor periphery was imperfect. Yet four parts of the world where industrialization of the late comers was most dramatic were also those parts where the pro-industrial world market forces were most dramatic—Mexico, China (Shanghai), Japan, and the European periphery.

12.9 Decomposing the Sources of the Industrial Liftoff in the Poor Periphery

This chapter does not offer any explicit empirical decomposition of the sources of the industrial liftoff (or its absence) in the poor periphery between 1870 and World War I. At this point the historical evidence is not yet sufficient for that demanding task: rather we have been content to lay out the competing explanations, to offer some plausible support for them, and to set out an agenda. Still, when future research offers an explicit empirical decomposition of the late 19th-century industrial liftoff in some precocious locations around the poor periphery, this chapter will have shown how those decompositions are likely to have differed across countries, and where these forces are likely to have mattered most.

My guess is that changing fundamentals will end up playing a much more modest role in contributing to the industrial liftoff than the current neo-institutional literature supposes: those fundamentals may help explain where industrialization started in the poor periphery, but changing fundamentals are unlikely to explain when and by how much. I stress the word changing since it had to have been changing fundamentals raising productivity in manufacturing, or changing external terms of trade, or changing domestic wage competitiveness, or changing tariffs, or changing real exchange rates improving profitability in manufacturing. It would have been those changes that accounted for the liftoff, not levels of any of these contending forces.

(p.214) The reader must have noted that this chapter has paid little attention to rising total factor productivity in manufacturing around the poor periphery. There are two explanations for its omission. First, only rarely has manufacturing productivity growth been estimated anywhere in the poor periphery over the four decades after 1870. Second, the United States, Britain, and other competitors to local manufacturing in the poor periphery were themselves undergoing rapid productivity advance, so we’d have to find evidence of total factor productivity catching up in the periphery leaders like Catalonia, the northern Italian triangle, Russia, Japan, Shanghai, Bombay, Bengal, Brazil, and Mexico. Such evidence will be hard to find. Maybe rapid manufacturing productivity catch up played a role, but based on Latin American experience, it probably did not explain most of the industrial liftoff in the leading parts of the poor periphery. If not productivity catch-up, what was it?

Changing external terms of trade must have played a big role, as the immense pre-1870 boom in the external terms of trade switched to a post-1870 bust; that is, a pre-1870 fall in the relative price of imported manufactures reversed trend to become a post-1870 rise. What had been a sickly Dutch disease before 1870 became a healthy Dutch revival after 1870, at least in some parts of the poor periphery. These forces were strong and it is time that the literature paid more attention to them.

Changing tariff and real exchange rate policy also must have played a big role, but only in those parts of the poor periphery where there was policy autonomy. Average tariff rates in Latin America rose hardly at all over these four decades, but the effective rate of protection for manufacturing rose a lot as the tariff structure was rationalized. Since Mexico led the way, it had the first-strike advantage of effective pro-industrial policy, and this long before the early-ISI protectionist policies of the 1930s. In addition the real exchange rate depreciated for both Brazil and Mexico, offering more benefits to import-competing industry in those Latin American industrial leaders. Since South Asia, Southeast Asia, Africa, the Middle East, and the other parts of the poor periphery did not undergo the same real currency depreciation (tied as they were to gold and/or the mother country’s currency), these two leaders enjoyed a first strike pro-industrial advantage—this long before the well-known real currency depreciations of the 1930s.

It is time for us to confront trade policy in detail, and that’s the agenda for the next chapter.


(1.) It should be pointed out that these figures are inconsistent with Fernando Rocchi’s (2006) recent book that stresses industrialization in Argentina, perhaps because Rocchi does not report sector or market shares.

(2.) The term “liftoff” should not be confused with Walt Rostow’s term “takeoff in to self-sustained growth” coined nearly a half century ago (Rostow 1964). I use the term only to describe an acceleration, and, to my knowledge, it was first used by Haber (2002).

(3.) All of the GDP estimates cited here are from Maddison (2008).

(4.) The tariff rates would, of course, be even higher if we looked only at manufacturers.

(5.) The effective rate of protection includes not just the impact of the tariff on imported final goods (e.g., cotton textiles), but also on inputs (e.g., raw cotton). Before Mexican tariffs were rationalized, high tariffs on raw cotton tended to offset any benefits from high tariffs on cotton textiles. (p.248)

(6.) By 1960 Mexico had much lower tariffs on capital goods and industrial raw materials than did Argentina or Brazil (Taylor 1998; Haber 2006: table 13.8, 574).

(7.) Rocchi (2006: 208a) also notes that Argentina did not join this pro-industrial policy trend.

(8.) The income terms of trade is defined as pXX/pM or export revenue (pXX) over import price (pM), or the quantity of imports that export earnings will buy. To the extent that foreign exchange earnings were used to buy imported capital goods and intermediates, manufacturing was favored.

(9.) The real exchange rate also depreciated by about 50 percent just between 1885 and 1892 (Catão 1998: 74), and it never recovered. We return to this below.

(10.) Data on the percentage of the labor force in manufacturing come from Kirsch (1977: 41), Seminario (1965: 48), and Dorfman (1970: 310).

(11.) On the European grain invasion, see O’Rourke (1997).

(12.) Except for Argentina, Chile, Columbia, and Uruguay, the rest of Latin America did not adopt fixed metallic (gold or silver, or both) exchange rates until after 1900 or never did. Even India, Japan, and Russia waited until the 1890s, and Bulgaria, China, and Spain never did. See Meissner (2005: table 1).