In Trade and Poverty: When the Third World Fell Behind, Jeffrey Williamson presents a critical perspective on the gains from trade. While income gaps between rich and poor countries had already emerged five centuries earlier, the divergence accelerated in the nineteenth century. The central question in this book is whether the great divergence in per capita GDP between 1830 and World War I was due to globalization. Williamson's answer is yes: Globalization led developing countries to de-industrialize and concentrate exports in a narrow set of commodities. Excessive concentration of output exacerbated income volatility. As a consequence, developing countries grew more slowly, reinforced the power of landed elites, and fell further behind.
Why would globalization lead developing countries to de-industrialize? Williamson begins by exploring why trade exploded in the nineteenth century, which he dubs the “first global century.” Falling tariffs, developments in transport such as railroads, declining shipping costs, and international peace agreements led trade as a share of global GDP to increase from 1 to nearly 8 percent, an eight-fold increase. Falling trade costs occurred simultaneously with the Industrial Revolution, which meant that industrial countries rapidly increased their demand for raw materials. As a result, the price of commodities and raw materials for the Third World boomed, and the price of manufacturing imports now made more cheaply by industrial countries fell. This enormous increase in terms of trade (TOT) led developing countries to “de-industrialize”: cut back on both the volume and share of manufacturing in GDP.
Williamson documents the reversal in industrialization that took place in emerging markets during the nineteenth century. Williamson argues that this de-industrialization led these countries to grow more slowly and concentrate production in more volatile, less dynamic commodities. Dutch disease retarded economic development, because it delayed the structural transformation out of low-productivity agriculture. Williamson draws on an existing literature that identifies industrialization and manufacturing as a source of increasing returns and economic transition.
The critical evidence to support a causal link between nineteenth century globalization, de-industrialization in the Third World and slower growth there is presented by Williamson in Chapter 11. Williamson presents econometric evidence showing that industrial and developing countries reacted quite differently to nineteenth century globalization. In the core industrial countries, improvements in terms of trade were associated with higher GDP growth and TOT volatility had no adverse impact on growth. In the periphery, TOT improvements between 1870 and 1939 were associated with GDP declines, and TOT volatility hurt growth. Williamson concludes that “up to 1870, terms of trade volatility and secular boom served to lower growth rates in the Third World relative to the European core, enough to explain perhaps three-quarters of the growth rate gap between core and periphery and the great divergence.”
One question is whether developing countries could have prevented this great divergence through protection or other industrial policies. Williamson studies countries in the periphery that did not fall behind, like Mexico and Brazil. Mexico and Brazil were able to begin the process of catch up earlier due to three factors: real exchange depreciation, an activist trade policy which raised effective rates of protection, and terms of trade deterioration which offset the disincentive effects of Dutch disease.
A natural follow-up question is why other parts of the developing world were not able to take the same path as Mexico and Brazil. Many countries could not affect their real exchange rates (many were dependent colonies) or their terms of trade. Williamson also makes the case that tariff policies in most countries were not set to conform with “optimal industrial policy.” He argues that the focus of protection outside of Mexico and Brazil was to compensate losers, not to foster industrialization. Such targeting would have had to focus on areas of latent comparative advantage or sectors with Marshallian externalities. Williamson also points out that even if protection had correctly targeted the right sectors, it would have been less desirable than other forms of industrial policy. Harrison and Rodriguez-Clare (2011) also make this point about the second best aspect of tariffs as a form of industrial policy. This last part of the book is not entirely convincing, as it is difficult to see how the pattern of protection in Mexico or Brazil was “better” without further analysis.
This book provides a compelling alternative perspective on the negative consequences of globalization. Although Williamson acknowledges that opening up to trade conveyed static gains even for the Third World, he also challenges the orthodox perspective. He argues that growth rates for the periphery were slower (and more volatile) than they would have been without globalization. This is a bold and intriguing hypothesis.
I have several additional comments on the book. The title of this book is called Trade and Poverty. This is a misnomer: the book is not about poverty at all, but about global inequality. Williamson's main idea relies critically on the assumption that industrialization carries benefits that other types of economic activity do not. Why and how does industrialization lead to economic development? What is so special about industry? This part of Williamson's argument is not clear.
While the claims in this very well written book are entirely plausible, the causal links from global shocks to terms of trade changes, de-industrialization, slow growth, and inequality are not fully fleshed out. Williamson documents a very different relationship between terms of trade and growth for industrial versus emerging markets. Yet changes within each country—such as the emergence of democracy or stronger institutions—could explain these differences. More rigor in mapping the causal links from falling transport costs to terms of trade shifts and de-industrialization in the Third World would be desirable.
I would have liked to see more space devoted to alternative policy options. Williamson's book only devotes one page to what the Third World could have done differently. The United States is itself going through a process of de-industrialization today. Exploring in more detail the policy options that could have been pursued by the periphery 100 years ago could be of immediate policy relevance for many industrial countries today.
