The Treasury

Global Navigation

Personal tools


The Challenge of Structural Change in APEC Economies - WP 07/06

2  Processes of Economic Growth and Income Convergence

2.1  Growth and convergence in neoclassical and endogenous growth theories

Economic growth models provide a framework for thinking about how economic growth processes and income convergence take place. A starting point for understanding growth processes tends to be neoclassical and endogenous growth theories.

A characteristic of traditional neoclassical growth models, as originally developed by Solow (1956) and Swan (1956), is that savings and physical and human capital investment influence the level of income per capita. In seeking the highest possible returns, capital moves from places where it is abundant to where it is scarce, bringing with it new products, processes and technologies. By operating across economies, this process of economic integration can facilitate the convergence of incomes.

As pointed out by Barro (1991), in neoclassical growth models an economy’s per capita growth rate tends to be inversely related to its starting level of income per person. Low-income economies will tend to grow faster than high-income economies, due to the movement of capital across borders, thereby promoting economic integration and convergence in the levels of per capita income across economies. It is of course possible, according to this model, that economies do not converge to the same level of per capita income because of impediments to convergence, such as structural policies that are ineffective at improving market efficiency. However, this is not to say that a one-size-fits-all approach to structural policy settings is always appropriate. As highlighted by Rodrik (2003), a range of institutional settings can bring about growth and income convergence and the most appropriate institutional settings will depend on local conditions.

In modern endogenous growth theories, by comparison with traditional models which focus on factor accumulation, the main driver of growth and convergence is ideas. It is possible to construct endogenous growth models that exhibit convergence in per capita incomes. In such models, convergence can be driven by a number of factors. Barro and Sala-i-Martin (1997), for example, develop an endogenous growth model in which long-term growth is driven by discoveries of new technology. In this model “follower” economies catch up to “leader” economies by copying technology. As the pool of un-copied ideas diminishes, the cost of imitation increases and the growth rate of followers accordingly declines.

2.2  Convergence mechanisms

Traditional and modern growth theories imply there are various factors that could bring about income convergence. Convergence in the neoclassical growth model is driven by capital flowing from places where it is abundant (high-income economies) to where it is scarce (low-income economies) to achieve the highest possible returns. In this way, economic integration can bring about growth and income convergence. However, empirical evidence suggests that capital flows from high-income to low-income economies are very modest and much less than predicted by the neoclassical growth model. Lucas (1990) canvasses theoretical explanations for these observed differences, including differences in economic fundamentals across economies that lower the rate of return on capital in low-income economies relative to high-income economies (such as technological differences, government policies and institutional structures) and capital market imperfections. Subsequent articles have attempted to test the relative importance of these explanations empirically. Alfaro, Kalemli-Ozcan and Volosovych (2005), for example, find that low institutional quality is the leading explanation for the actual capital flows between low-income and high-income economies.

Endogenous growth models emphasise the spillover of ideas and technological knowledge as a key mechanism driving growth and income convergence. The transfer of scientific knowledge may occur through foreign direct investment (FDI) in low-income economies bringing with it the skills of investors, or through international trade. Economies may “learn by exporting” by interacting with foreign customers and learning how to meet higher product standards, or through technology embodied in imports. Keller (2004) surveys the literature on the extent of international technology diffusion and the channels through which technology is spread. He concludes that there is no evidence that international learning is inevitable, or that it is easier for relatively undeveloped economies. Evidence suggests that importing is associated with technological spillovers, but evidence of benefits associated with exporting is weaker. The literature suggests that there can be spillovers from FDI but they vary between economies, regions, sectors, and firm structures. Similar conclusions are drawn in the surveys by Greenaway and Kneller (2007) and Wagner (2007).

Migration from low-wage to high-wage economies could also bring about higher economic growth and income convergence. Migration will cause labour to become scarcer in low-income economies and more abundant in high-income economies and will tend to reduce wage differentials between economies. Restrictions in the movement of persons, rigid labour markets and overly generous and/or poorly designed welfare policies will limit the extent to which this mechanism is able to operate. Sinn (2007) assesses the impact of the emergence of “Asian tigers” and ex-Communist economies in world trade on factor price convergence in Europe. He concludes that labour market rigidities and welfare policies in Western Europe are impeding adjustment and factor price convergence. These policies may be causing greater immigration and more capital-intensive exports from Western Europe than would otherwise be the case and could be contributing to unemployment and sluggish growth in Europe.

Trade and specialisation is another mechanism that could drive economic growth and income convergence. Trade between high-income and low-income economies will allow high-income economies to specialise in capital-intensive production processes, and low-income economies to specialise in labour-intensive production processes (as per their comparative advantage). Both specialisation processes tend to reduce wage differentials as the demand for unskilled labour in high-income economies falls, while it rises in low-income economies. Estimating the effects of trade on economic growth is challenging because of the joint determination of empirical measures of economic growth and international trade. Frankel and Romer (1999) use instrumental variable estimation to help overcome some of these estimation problems, and find that trade has a large, positive (although only moderately significant) effect on income.

In addition to “static” gains from trade (arising from economies exploiting their comparative advantage and economies of scale), there are also potential “dynamic gains” from trade. Dynamic gains from trade refer to trade-related improvements in an economy’s productivity growth rate that arise from increased integration in the global economy. A recent OECD (2006) study identified three interconnected channels through which trade may increase productivity: by increasing investment; aiding technological diffusion; and promoting the competitive impetus to innovate. The empirical evidence on dynamic gains from trade is mixed. Research has not established a robust link between trade policy and long-run productivity growth rates. However, at the economy-level there is strong empirical evidence of the link between openness to trade (measured by actual trade flows) and productivity levels.

Traditional and modern growth theories imply that there are potential benefits for all economies, not simply the lower-income economies, through specialisation, better allocation of skills and other resources, the dynamic interaction of learning, and the two-way spillover of knowledge. According to these models, in the Asia-Pacific region we would expect to find lower-income economies growing faster than higher-income economies, thereby bringing about the convergence of per capita incomes. The following section examines the growth and convergence performances across the APEC region. This is followed by a more detailed discussion of evidence drawn from the research literature of factors that may be impeding these processes.

Page top