Tim Harford’s latest Undercover Economist column covers some interesting research on industrial development. The paper examines the type of products that countries produce, and the way that a country’s ‘manufacturing portfolio’ changes over time. The key finding is that countries tend to develop by producing similar products to those that they already produce.

This makes intuitive sense: if a country already has infrastructure suited to the manufacture of a particular product then it will be less costly to develop similar products than to develop radically different ones. The problem arises when a poor country with limited production diversity approaches the limits of its current manufacturing processes. It is very difficult and costly to make the transition to producing a new, unrelated product type. The paper’s data confirms that this rarely happens. Notably, the authors find that:

Rich countries have larger, more diversified economies, and so produce lots of products […]. East Asian economies look very different, with a big cluster around textiles and another around electronics manufacturing […]. African countries tend to produce a few products with no great similarity to any others.

If poor countries are to make the step to producing new products then some intervention in the development process may be required. This points to a role for some government industrial policy at a national level, or structural intervention at an international level. It could mean a new justification for infant industry protection in developing countries. It also, perhaps, points to a different way of making effective use of the limited aid money available to organisations such as the IMF and World Bank.