Stuff I Say at School – Part IX: Indirect Effects of Trade on Poverty

This is part of the Stuff I Say at School series.

The Assignment

A critical literature review of trade openness on poverty. This post consists of part of the introduction and the section on trade and economic growth.

The Stuff I Said

As the The Economist (2013) reports, “The world’s achievement in the field of poverty reduction is, by almost any measure, impressive.” The United Nations’ “aim of halving global poverty between 1990 and 2015 was achieved five years early…The [Millennium Development Goals] may have helped marginally, by creating a yardstick for measuring progress, and by focusing minds on the evil of poverty. Most of the credit, however, must go to capitalism and free trade, for they enable economies to grow—and it was growth, principally, that has eased destitution.” This last statement is at times controversial in the popular press. In order to engage the controversy, this review will survey the academic literature on the effects of trade liberalization on poverty. This will be explored through two main channels. First, through trade’s indirect effects on poverty via economic growth. Most research on trade liberalization and poverty is focused on the relationship between trade and growth. Other possible avenues associated with trade, growth, and poverty—such as innovation1 or institutional change2 —will largely be ignored. Only work that focuses specifically on the connection between trade and growth will be reviewed in this section. The final section will mine the scant research on direct effects of trade liberalization on poverty.

Economist and trade expert Jagdish Bhagwati (2004, pg. 64) argues that “freer trade is associated with higher growth and…higher growth is associated with reduced poverty. Hence, growth reduces poverty.” However, empirically establishing this connection between growth and poverty reduction is necessary, seeing that it is theoretically possible for the benefits of economic growth to not be distributed to the poorest segments of society. Using a sample of 92 countries over a 40-year period, David Dollar and Aart Kraay (2002, pg. 219) find that economic growth on average increases “the income of the poor to the same extent that it increases the income of the other households in society.” Kraay (2006) finds that the main explanation for cross-country differences in poverty shifts over time is the growth in average incomes: 70% in the short-run and 97% in the long-run, respectively. In a follow-up study, Dollar, Kleineberg, and Kraay (2016, pg. 81) look at a dataset of 121 countries over four decades and come to the same conclusion: “Incomes of the bottom 20 percent and bottom 40 percent of income distribution generally rise equiproportionally with mean incomes as economic growth proceeds.” In a book-length treatment on the economic reforms in their home country of India, Bhagwati and Arvind Panagariya (2013) find that the growth since the 1990s has reduced poverty nationwide in both rural and urban regions alike and among socially disadvantaged groups. These studies confirm that the connection between economic growth and poverty reduction is solid.

Image result for trade

According to Bhagwati (2004), trade openness produces growth through various channels, including specialization, economies of scale, increased competition (and, consequently, decreased domestic monopolies), promotion of macroeconomic stability, and increased foreign direct investment. Much of the empirical evidence supports this view that trade openness results in growth. David Dollar’s (1992) early analysis of 95 developing countries between 1976 and 1985 concludes that trade openness (what the author calls an “outward-orientation”) and per capita GDP growth are highly correlated. Those countries in the most open quartile experienced a per capita growth rate of 2.9 percent, while those in the most closed quartile languished at -1.3 percent. Similarly, Harvard’s Jeffrey Sachs and Andrew Warner (1995, pg. 45) show in a cross-country analysis that, between 1970 and 1989, “being open to international trade has been sufficient to achieve growth in excess of 2 percent for developing countries.” However, Sachs and Warner (1995, pg. 45, fn. 61) acknowledge that their “indicators of openness are associated with other market-based reform policies, which makes it difficult to identify the precise contributions of trade as compared to other policies.” Using the portion of total trade that relies on geographical factors as an independent variable, a study by Jeffrey Frankel and David Romer (1999) finds that a one percentage point increase in the ratio of trade to GDP raises income per person between 0.5 and 2 percent. Measuring trade openness by means of tariff revenues, nontariff barriers, and other liberalization indicators, Romain Wacziarg (2001) discovers a positive effect of trade openness on economic growth in 57 countries between 1970 and 1989. Halit Yanikkaya (2003, pg. 57) provides continual support for the idea that trade stimulates growth, finding a “strong and positive relationship between trade intensity ratios and growth.” However, contrary to previous studies, Yanikkaya also finds that trade barriers can promote growth under particular conditions. William Cline (2004) questions Yanikayya’s latter findings, noting their contradiction with previous scholarship and the likelihood of his measurements either understating or misgauging the effects of protection. On the flip side, Francisco Rodriguez and Dani Rodrik (2001) argue that many of the measurements used by Sachs & Warner (1995) as well as Frankel & Romer (1999) are flawed in their openness measurements, fail to establish causality, and ignore other complementary policies necessary to promote and sustain growth. Perhaps surprisingly, T.N. Srinivasan and Bhagwati (2001) also find methodological problems with various cross-country regressions. However, in their view, this undermines many of Rodriguez and Rodrik’s criticisms due to their heavy reliance on these kinds of studies. After examining the evidence from several country-specific studies, Srinivasan and Bhagwati determine that Rodriguez and Rodrik’s criticisms fall flat and that trade and growth go hand-in-hand. Nonetheless, in a later paper, Bhagwati and Srinivasan (2002, pg. 182) acknowledge the cross-country regressions’ “interesting” findings that “practically no country that has been close to autarkic has managed to sustain a high growth performance over a sustained period.” A follow-up study by Frankel & Andrew Rose (2002) addresses many of Rodriguez and Rodrik’s concerns, controlling for small city-states, geographical distance, and institutional quality. They determine, “In every case, regardless of whether the other controls are included or not, the openness variable retains most of its magnitude and all of its statistical significance in the presence of each of the three Rodriguez-Rodrik modifications” (2002, pg. 451; italics original). On the other hand, Rodrik, Subramanian, and Trebbi (2004, pg. 135) find that when the impact of geography, global integration (international trade), and institutional quality are compared, “the quality of institutions trumps everything else.” Yet, they also find that institutions and integration positively influence each other: “A unit increase in institutional quality increases the trade share by 0.45 units, while a unit increase in trade increases institutional quality by 0.22 units” (2004, pg. 143). Conversely, Francisco Alcala and Antonio Ciccone (2004, pg. 638) control for both geography and institutional quality and measure “real openness (imports plus exports in exchange rate U.S. dollars relative to GDP in purchasing power parity US$).” Their results show that trade has a significant and robust positive (and causal) effect on productivity. Marta Noguer & Marc Siscart (2005) also control for geography and institutional quality, finding that a 1% increase in the trade share of GDP leads to a similar increase in income per capita. Dollar & Kraay (2004) look at decade-by-decade changes in trade volume across 100 countries and find that within-country changes in trade volume have a strong positive relation with changes in growth. This results in increased income for the poor. Nonetheless, Dollar and Kraay recommend complementing open trade with strong safety nets; nets that are in turn better funded by trade-induced growth.

Research over the last decade continues to support these earlier findings. Wacziarg & Welch (2008, pg. 212) find that between 1950 and 1998, “countries that liberalized their trade regimes experienced average annual growth rates that were about 1.5 percentage points higher than before liberalization.” Vlad Manole & Mariana Spatareanu (2010) use data from 131 developed and developing countries and find that reductions in trade protections lead to higher levels of income per capita. Expanding the data from Sachs & Warner (1995) and Wacziarg & Welch (2008), David Weil (2013) finds that the average growth rate of income among more open countries was significantly higher (3.1% per year) than that of closed countries (1.5% per year). Antoni Estevadeordal & Alan Taylor (2013) explore the outcomes of liberalized trade policies in countries during two time periods: 1975-1989 and 1990-2004. Those countries that liberalized during these periods had about one percentage point higher growth rates compared to non-liberalized countries.

Image result for trade

Maureen Were (2015) performs a cross-country analysis of 85 countries from 1991 to 2011. In agreement with most of the literature, she finds that trade has a positive and significant effect on economic growth. However, among the Least Developed Countries (LDCs)—most of which are in Africa—the statistical significance disappears. Nonetheless, she notes that trade’s effects on both domestic and foreign direct investment (FDI) are positive and significant. However, Markus Brueckner & Daniel Lederman’s (2015, pg. 1318) study focuses specifically on sub-Saharan Africa and finds that “a 1 percentage point increase in the ratio of exports plus imports over GDP is associated with a short-run increase in GDP per capita growth of approximately 0.5% in a given year,” while the long-run effect reaches about 2 percent. Pam Zahonogo (2016) argues for a Laffer Trade Curve among sub-Saharan Africa. He finds that for most measures, the thresholds are virtually non-existent.3 However, when imports make up for more than 33.16% of GDP, the positive effects of trade on growth begin to decline. He suggests complementary policies that promote new investments, improve institutional quality, and develop human capital. Yet, it is feasible that higher barriers on imports may harm the poor. In their analysis, Pablo Fajgelbaum & Amit Khandelwal (2016, pg. 1116) find “a propoor bias of trade in every country. On average, the real income loss from closing off trade is 63% at the 10th percentile of the income distribution and 28% for the 90th percentile.” This is due to low-income consumers spending more on traded sectors compared to high-income consumers, who spend more on non-traded services. Furthermore, Furceri, Hannan, Ostri, & Rose (2019) examine a dataset of 151 countries from 1963 to 2014. According to their results, tariff increases negatively impact output, productivity, employment, and consumption. The authors conclude, “All this seems eminently sensible and bolsters the arguments that mainstream economists make against tariffs; our results can be regarded as strong empirical evidence for the benefits of liberal trade” (2019, pg. 28). Perhaps most impressively, Arvind Panagariya (2019, pg. 98) has compiled “data on per capita incomes, good and services exports, goods and services imports, and goods and services exports as a proportion of GDP in constant 2005 U.S. dollars for more than two hundred countries over a period of fifty-four years between 1960 and 2013” (broken into three smaller periods: 1961-1975, 1976-1994, and 1995-2013). With incredible detail, he demonstrates a causal relation between trade and per capita income: those countries that experienced intensive growth in these various periods always maintained a high and/or expanding trade-to-GDP ratio.

Overall, the empirical literature seems to indicate that trade openness has a positive effect on economic growth. Growth in turn reduces poverty. Multiple literature reviews and book-length treatments have drawn similar conclusions. For example, Joshua Lewer and Hendrik Van den Berg’s (2003) review of the literature finds that, on average, studies point to a 1/5 (or more) percentage point increase in real GDP for every percentage point increase in trade. Winters, McCulloch, and McKay (2004) are slightly more cautious, but ultimately admit that the preponderance of evidence suggests that trade openness increases economic growth and income levels within countries. Alan Winters and Antonio Martuscelli (2014, pg. 498) review the more recent literature and conclude that “the evidence is very strong that greater openness is generally associated with higher levels of income and, equivalently, that trade liberalization is associated with temporary increases in growth. The relationship appears to be causal but is not absolutely invariable.” Douglas Irwin’s (2015, pg. 197) survey of the evidence finds that “greater trade openness—marked by rising trade and low or declining trade barriers—has been a feature of virtually all rapid-growth developing country experiences in the past fifty years.” Examining countries such as China, India, South Korea, Chile, and Vietnam, Irwin concludes that liberalized trade has been associated with greater growth and, consequently, declining poverty. Panagariya (2019) performs a similar analysis, dedicating extensive attention to economic “miracles” such as Hong, Kong, Singapore, Taiwan, South Korea, India, and China. He then turns his attention to other successes throughout Asia and Africa (and even moderate ones in Latin America). He writes, “I have shown that in each case, trade liberalization and expanding trade are integral parts of the success story” (2019, pg. 322).

While complementary domestic policies (e.g., improvements in institutional quality) are necessary to reap the full benefits of international trade,4 there appears to be no evidence that suggests trade has anything other than positive effects on growth. The majority of studies support the claim that trade reduces poverty through increased economic growth. Panagariya (2019, pg. 125) concludes, “Given this set of facts, any advice to the developing countries to opt for protectionist policies can only be viewed as purely ideological.”