Choosing countries as models for industrial growth

Train station. India. Photo: © Curt Carnemark / World Bank

Shanta’s thoughtful comments on our Growth Identification and facilitation (GIF) paper are most welcome. The issues of industrialization and structural transformation are at the heart of economic development. Following comments already made by my co-author Célestin Monga on this blog, let me offer a few thoughts to this exchange.

First, the GIF approach explains the economic success of a very diverse group of countries: China (with 1.3 billion population), Japan (100 million); Taiwan-China (20 million); Korea (40 million); Singapore (5 million); or Mauritius (400.000). The framework has also been applied in large Western countries such as Germany, France, and United States, and small European countries like Sweden, Norway, Finland, and Ireland. The political systems of those economies are also very different, some are democratic and some are authoritarian.

I am therefore not too worried about the idea of choosing a federal sub-continent of 1 billion people, such as India, as a model for a centralized country of 39 million, such as Kenya. There are already successful examples in the services sector. The case of call centers is well known. After they mushroomed spontaneously in India in the 1980s, Kenyan firms discovered that they, too, had latent comparative advantages there. Kencall, a call center established by a private entrepreneur in Nairobi in the 2000s, imitated India’s call centers. Its entry/operation was made possible by subsidies from the government and World Bank support for the cost difference between satellite transmission and sea-cable transmission. The subsidies were eliminated after the sea-cable project was completed.

The major difference between a large economy and a small one is that the former can specialize in many tradable sectors while the later will only focus on a few tradable sectors. Therefore, when a small economy uses a large one as the model for its tradable sectors’ development, it must choose sectors to upgrade (or for the economy to diversify into), based on two criteria: path dependence (increasing returns based on past decisions), and market potential in the domestic and global markets.

High-tech sectors in India grew out of the legacy of the country’s capital-intensive heavy industries-oriented development strategy adopted in the 1950s. That strategy went against India’s comparative advantage. Nonetheless, it enabled India to establish some “advanced industries.” Like China and many socialist and non-socialist countries that pursued a similar strategy before the reform/transition in the 1980s, India’s economic performance under that strategy was poor. The annual GDP growth rate was 3.6% between 1950-1980, according to Angus Maddison’s estimates. Even if it chooses to emulate India, Kenya needs not to follow the wrong model in the choice of industries.

It was not surprising that India had to resort to restrictive labor regulations, which are endogenous to a development strategy is comparative advantage defying. I discussed this problem in my Marshall Lectures [Economic Development and Transition: Thought, Strategy, and Viability, Cambridge, UK: Cambridge University Press, 2009], and a working paper “Development Strategy, Viability and Economic Distortions in Developing Countries”.  While such regulations originated in misguided economic policy, they quickly became politically difficult to eliminate. A pragmatic way to deal with such types of distortions is to avoid “big-bang” changes and follow instead a gradual, dual-track approach, as done in Mauritius, China and other successful transition/reforming economies. such gradualism often entails continued government support to “nonviable” firms in priority sectors and, at the same time, liberalization of the entry of private enterprises, joint ventures, and foreign direct investment in labor-intensive sectors in which the country has a comparative advantage.

Shanta’s observation that once an industry has been identified, it may not be competitive because policies and regulations in many poor countries make certain prices immovable is inconsistent with the very premise of the GIF framework. The GIFF stresses the necessity of only identifying sectors that are consistent with the country’s (latent) comparative advantages, so that the government’s facilitation does not rely on distortions that impede competition. If a formal dual-track approach à la Mauritius is not politically feasible in South Africa, another possibility is to find other pragmatic ways to remove constraints and facilitate entry and growth of small or medium-sized owner-operated firms. Improving the infrastructure required for SMEs’ operations near shanty towns might be an example.

Finally, the suggestion of a seventh step in the GIF and dealing with broad governance issues is welcome. My paper on New Structural Economics: Rethinking Economic Development already deals with them, and I will further take them up in future research.