Are certain countries doomed to remain emerging?

Globalization, structural change, and the uneven geography of convergence

Editor's note:

This report is part of a series of essays by the Global Economy and Development program—a 10th anniversary edition. The series, available here, delves into the critical issues facing all those concerned about globalization. You can join the conversation on Twitter using #11GlobalDebates.

1.1 What’s the issue?

Incomes in developed and developing countries have been converging, especially since the turn of the century, but the unevenness of that trajectory merits further examination. Beginning in the early the 2000s, the average per capita income of developing countries (adjusted for purchasing power parity) has increased substantially relative to the average per capita income of developed countries (Figure 1). Output per person in developing countries almost doubled between 2000 and 2010. The average annual rate of growth for all developing countries over the decade was 7.6 percent—4.5 percentage points higher than the growth rate in rich countries.


Global trade has been the biggest driver of convergence. Cheaper and easier transport and communications have allowed supply chains to expand across the globe. Between 1994 and 2007, trade grew at almost twice the rate of growth of the world economy, spurring much faster catch-up than in the past. In an era of “Brexit,” calls in the U.S. for retreat from the North America Free Trade Agreement and the Trans‑Pacific Partnership and in the face of rising populist nationalism in Europe, advocates of globalization can point to convergence and the decline in global poverty that it has fostered as happy outcomes of growing global connectivity. Like most happy stories, however, the reality is more complicated.

1.1.1 The uneven geography of convergence

Driven by the growth of China and India, rapid growth of income per capita in Asia has been the main source of income convergence between developing and developed countries (Milanovic, 2010; Bourguignon, 2015). Incomes per capita in Asia rose from 14 percent of the average income of developed countries in 1990 to 25 percent in 2014. Since China and India represent 37 percent of the world population and 43 percent of the population in emerging economies, their success strongly influences the global average.

Excluding Asia, income differentials between developed and developing countries remain large and have barely budged. Average per capita income in Africa in 1990 was about 12 percent of per capita income in developed countries. In 2014 it was still 12 percent, the “African Growth Miracle” notwithstanding. The income per capita of countries in Latin America and the Caribbean has remained largely unchanged since the mid-1990s, averaging approximately 36 percent of the developed countries. In Brazil, the region’s largest economy, GDP per head was 25 percent of America’s in 1998. It caught up just three percentage points over the ensuing 15 years. Clearly some parts of the developing world were converging more quickly than others.

1.2 Convergence and structural change

In a series of papers, Margaret McMillan and Dani Rodrik provide a new perspective on the uneven geography of convergence (see for example McMillan, Rodrik and Verdzuco-Gallo, 2014). They point to an enduring lesson: an economy’s overall productivity depends not only on what is happening within sectors, but also on the reallocation of resources across sectors. Structural change—the movement of labor and other resources across the economy—is often a key driver of growth in poorer countries, but it can also hold growth back.

7.6% the average yearly growth rate of developing counties between 2000 and 2010.

The most striking finding by MacMillan and Rodrik is that between 1990 and 2005, differences in structural change accounted for about 60 percent of the difference in growth of output per worker in Asia versus Africa and Latin America. In Asia, reallocation across sectors added to the gains in productivity within sectors. In many Latin American and sub-Saharan African countries, changes in economic structure since 1990 have reduced rather than increased economic growth. Seen from this perspective, the uneven geography of convergence is a product of the uneven geography of structural change.

1.3 The role of globalization

MacMillan and Rodrik argue that these different outcomes reflect the different ways in which countries and regions have integrated globally. In some cases—notably China, India, and other Asian countries—high productivity employment opportunities have expanded rapidly, mainly as a consequence of explosive growth in industrial exports, with structural change boosting overall growth. When combined with productivity growth within sectors, structural change is like running up an escalator.

By contrast, in Latin America and sub-Saharan Africa, import competition has forced manufacturing industries to become more efficient, and the least productive firms have gone out of business. While this churning has had a positive impact on productivity within manufacturing, the workers who were displaced in Latin America ended up in lower-productivity, informal employment. In Africa the formal sector has grown more slowly than the labor force, trapping workers in subsistence agriculture and informal services. Moreover, MacMillan and Rodrik’s data did not permit them to take into account unemployment. Latin America and the middle income countries of Africa would look considerably worse had they done so. In short, the increase in productivity in both regions brought about by the explosion of global trade has come at the cost of growth-reducing structural change.

1.4 Global Headwinds

Since 2010, growth rates across the developing world have slipped back toward those in advanced economies. The favorable external conditions of the past decade are gone; commodity prices have already fallen, and world interest rates look set to rise. As IMF Managing Director Christine Lagarde noted in January 2016, “On current forecasts, the emerging world will converge to advanced-economy income levels at less than two-thirds the pace we had predicted just a decade ago.”

Nevertheless, the uneven geography of convergence is likely to continue. While China’s growth is expected to cool, the most recent IMF projections still give Asia a 2.6 percentage point growth advantage over the developed countries. Latin America meanwhile is expected to grow at a slower rate than the advanced economies, and per capita incomes in Africa are projected to grow at about one percent. Africa and Latin America will need major policy changes for the developing world to achieve broader-based convergence.

1.5 What to Watch out for?

For Africa, the very low levels of productivity and industrialization across the continent offer enormous potential. But to realize it, Africans will need to grow manufacturing and tradable services. A recently completed African Development Bank, Brookings, and UNU-WIDER project offers some new thinking on how Africa can accomplish this transformation—through policies to promote industrial exports, build the capabilities of domestic firms, and foster industrial clusters (see Made in Africa: Learning to Compete in Industry, Brookings Institution Press, 2016).

60% of the difference in output growth per worker in Asia vs. Africa and Latin America between 1990 and 2005 was due to differences in structural change.

In Latin America the big challenge is to reverse the trend toward informality. Sixty-one percent of all workers in Latin America are employed in the non-tradable service sector, mainly in small firms. The Inter-American Development Bank estimates that increasing informality causes Latin American countries to lose between 0.4 and 5.2 percent of output annually (Pages, 2010). Reforms that promote the growth of more productive firms are essential. One place to start is by making labor markets more flexible—for example by ending the practice of financing social security by taxing formal labor. Investing in knowledge is also key. Despite its middle income status, Latin America invests only one-fourth of the amount in innovation that countries in the Organization for Economic Co-operation and Development do.

If, as seems likely, developing countries will catch up more slowly to advanced economies over the foreseeable future, addressing the uneven geography of convergence is essential. Where Asia stands apart has been in ensuring that its pattern of structural change contributed to the overall economic growth. Governments in Africa and Latin America need to find ways to emulate it.