Some Speculation on Growth and Poverty over the Twenty-First Century

If there is a unifying theme to the papers in this volume, it is that when it comes to globalization and poverty, it is very difficult to make broad generalizations. So it is a bit awkward in this concluding panel to be charged not only with making broad generalizations, but in extrapolating these generalizations to the future.

From the research presented here, it is apparent that the effects of greater trade in goods and capital on the poor are extremely complex and almost always depend on domestic institutions. Overall, it is very difficult to document rigorously the effects of trade liberalization on poverty. The effect of globalization on health among the poor (and others) is similarly complex, trading off the faster spread of disease with faster spread of knowledge about treatment and other benefits. No matter how globalization is modeled, one still needs to rely heavily on innate productivity differences to explain why some countries and regions are rich while others are poor. There is great disagreement over what poverty is, with several plausible measures, depending on whether one regards poverty primarily as an absolute condition or a social condition. And it has also been shown that economic growth and happiness are not necessarily equivalent.

In this short note, I want to balance the discussion with a reminder that if the frame of reference is some absolute notion of poverty, then over the very long term, the trajectory of global growth is going to dominate all other factors. Thus one can expect that as global income inexorably expands over the next century, issues of inequality, rather than subsistence, will increasingly take center stage in the poverty debate. And at the end of this discussion, I will ponder the risks to growth, including the question of whether financial crises will continue to hold back countries as they attempt to cross the bridge from lower to upper middle income.

Suppose for a moment, one were to ignore all the many nuances stressed in this volume and hazard a guess about the evolution of global and regional incomes over the next century. What would it be? According to conventional estimates, current trend for per capita global growth appears to be about 3 percent a year (using purchasing power parity [PPP] weighted averages of individual country growth), with part of that driven by technological change in the leading edge countries (especially the United States) and part of it driven by catch-up from the rest of the world.¹ Estimates of the pace of technological change at the frontier vary, with 2 to 2.5 percent being the current consensus.² (With global population growth of just over 1 percent, this implies that medium-term trend global growth is about 4 percent.) When past growth is broken down across regions, Asian economies exhibit a period of sustained convergence to U.S. per capita income levels, while Latin America has generally failed to converge, with real income per adult remaining at about 25 percent of U.S. levels for the past fifty years.³ Europe, after a sustained period of convergence up to the early 1980s (when gross domestic product [GDP] per adult reached about 70 percent of U.S. levels), has roughly kept pace. And many African countries, of course, have been falling farther and farther behind in relative income levels since gaining independence in the 1960s and 1970s.

How quantitatively important is long-term global growth relative to these regional differentials? If global per capita growth were to remain at 3 percent for the rest of the century, world per capita income would rise seventeenfold by the year 2100. Even with productivity growth at 2 percent, global per capita income would still rise almost sevenfold. Thus, even if Latin America failed to converge at all in the coming century, its income levels in 2100 would still be four times those in the United States today under the continued 3 percent trend scenario, and a 50 percent increase compared to United States income levels under the low-growth scenario. Suppose that Europe performs catastrophically, and every prediction of its...