Wednesday, May 25, 2011

Inter-sectoral productivity gaps and structural change

Structural change happens when an economy shifts its sources of growth and employment from agriculture to non-agriculture activities. As factors of production (labor and capital) move away from agriculture into modern economic activities overall productivity rises and incomes expand. The rate at which overall productivity rises in different non-agriculture activities (inter-industry productivity gaps) plays a vital role in bringing about and sustaining this change. [Productivity is defined as the ratio of each sector’s value added to employment in that sector.]

When compared to developed countries, developing countries are characterized more by large productivity gaps among firms and plants within same industry. This is indicative of allocative inefficiencies that reduce overall labor productivity. The inter-sectoral productivity gaps (the differences in average labor productivity) are a feature of underdevelopment.

But, productivity gaps among firms in the same or different industry can be an important source of growth, argues Rodrik and McMillan (2011) in the latest working paper (Globalization, Structural Change, and Productivity Growth). The reason is that when factors of production (mostly labor and capital) move from less productive to more productive activities, the economy grows even if there is no productivity growth within sectors. The movement of labor from low-productivity to high-productivity activities raises economy-wide labor productivity. This is growth-enhancing structural change, which the high-income countries have. The main messages of their paper are:

  • In many Latin American and Sub-Saharan African countries broad patterns of structural change have served to reduce rather than increase economic growth since 1990. Factors of production (labor) are moving into less productive sectors (informal and agriculture) from more productive ones in these regions. The opposite happened and is happening in Asia. This is especially true after trade liberalization as competition forced firms to exit the market, forcing them to lay off workers who ended up in the informal sector and low paying services job. [This is happening in Nepal as well. A number of garment workers that were laid off beginning 2000 have gone into the informal sector where wages are low.] The average manufactures-agriculture productivity ratio is 2.3 in Africa, 2.8 in Latin America, and 3.9 in Asia.
  • Factors that help determine if structural change is going in the right direction (and policy intervention might help): (i) Economies with relative comparative advantage (RCA) in primary products are at a disadvantage as their large share of natural resource exports means that the scope of productivity-enhancing structural change is narrow. These sectors cannot absorb much surplus labor form agriculture. (ii) Countries that maintain competitive or undervalued currencies tend to experience more growth-enhancing structural change as there is positive effect of undervaluation on modern, tradable industries. (iii) Countries with more flexible labor markets have greater growth-enhancing structural change as rapid structural change occurs when labor mobility across firms and sectors is high. [(ii) and (iii) is true in the case of Nepal; (i) may be true if we starting producing hydroelectricity and export it to India like Bhutan is doing.]
  • Agriculture is the sector with the lowest productivity in poorest economies.
  • During economic growth the productivity gap between agricultural and non-agricultural sectors first increases and then falls, exhibiting a U-shaped pattern (ratio of agricultural to non-agricultural productivity with respect to economy-wide labor productivity). The turning point comes at an economy-wide productivity level of around $9000—a development level somewhere between that of India and China. Initially, there is no large productivity gap between agricultural and non-agricultural sectors in poor countries. As economy grows, labor begins to move from traditional to modern sectors, thus leading to convergence of productivity levels across sectors in the economy [First, labor moves from agri to non-agri sector, increasing productivity in non-agri sector and also in agri sector (due to less labor). Then, diminishing marginal returns kicks in in the non-agri sector. Eventually, productivity level converges within sectors and also within countries with same income levels.]

  • Differential pattern of structural change account for a bulk of the difference in regional growth rates.
  • Domestic convergence, just like convergence with rich countries, is not an unconditional process. Starting out with a high share of labor force in agriculture may increase the potential for structural-change induced growth (not applicable to those having strong comparative advantage in primary products), but the mechanism is clearly not automatic.

This paper contributes Hausmann’s and his work on structural change induced by “jumping monkeys”, where production is moved from one product to another having similar features and using pretty much similar (or upgraded) factors of production. It involves (structural change) moving from the production of peripheral goods to core goods in product space.