Geography and Trade Structure: Implications for Volatility

In most of the developing world, sustained growth is a precarious achievement. The longstanding volatility of output in sub-Saharan Africa and Latin America is well known, and in the 1990s, instability extended even to some of the strong performers of East Asia. The sources of volatility remain somewhat obscure, however. There is little consensus among economists on the sources of output fluctuations even in developed countries, and since poorer countries appear to show a much wider range of volatility patterns, the intellectual challenge is a formidable one. Literature on the causes and consequences of volatility is growing by the day, however. Some of the leading explanations for output volatility include the role played by macroeconomic distortions, low levels of financial sector development and weak political institutions.1 Popular accounts of volatility in developing countries are based around the role of terms of trade fluctuations. The story is deceptively simple. Growth in a typical developing country may be more volatile by virtue of its specialization in primary commodities. Since primary commodity prices are more volatile in global markets, developing countries are more susceptible to terms of trade fluctuations—and, thereby—greater output volatility. But this is an incomplete description of growth instability in developing countries. It begs the question: why do poor economies tend to specialize in a narrow range of commodities. From the perspective of small open economies, changes in world prices can be considered as exogenous. But the effect of a given price change will depend on a country’s trade structure. And this is clearly endogenous in the long run. 1 See Malik and Temple (2009) for a more detailed review. 2 Why do some countries remain locked in primary commodity exporting, while others diversify their export structures and achieve greater specialization in manufactured exports? This paper argues that a country’s geographical characteristics can be an important determinant of its trade structure. In particular, it highlights the adverse effects of remoteness for export patterns and exposure to growth shocks resulting in high levels of volatility. Focusing on structural causes of volatility, this paper concludes that there is considerable empirical support for geography-based explanations for volatility. The effect of geography on volatility survives even after controlling for other determinants of volatility traditionally considered in the literature. The analysis in this paper is based on a forthcoming article in the Journal of Development Economics (see Malik and Temple (2009); an earlier more detailed working paper version is Malik and Temple (2006)).