Are commodities fate?

A fascinating new Science article suggests that a country’s development path resembles one of gradual product diffusion from a starting base, and so is heavily influenced by its initial endowment of commodities, skills, and industries.

In the paper and in some terrific supplementary material, the authors produce some incredible network diagrams of production and export like the one below:

The authors’ basic premise is that economies grow by upgrading the products they produce and export, and that the technology, capital, institutions, and skills needed to make newer products are more easily adapted from some products than from others. Countries move through the product space by developing goods close to those they currently produce.

What is amazing is that they manage to develop this idea without reference to Harold Innis and W.A. Mackintosh, who articulated the classic staple theory of economic development. According to the staple theory, different commodities offer different ‘backward linkages’ (input or infrastructure requirements, for instance) and ‘forward linkages’ (processing, refining, etc.) that can promote the development of related industries. And it is not as if these economists were obscure: Innis was a onetime President of the American Economic Association.

Understanding such product development and diffusion is crucial. Most countries in the world for most of their history have produced just a handful of commodities (sometimes just one or two), and have often produced them for a hundred years or more. In Canada this was fish and lumber, while in Colombia it was tobacco and (later) coffee. See my article on this here. Diversified nations like the US and the UK have been the gross exception rather than the rule.

Latin American economist Carlos Diaz-Alejandro called this the commodity lottery: each country’s exportable resources, he explained, were determined in large part by geography and chance, and differences in later economic development were a consequence of the economic, political and institutional attributes of each commodity.

Who wins the commodity lottery? According to the network diagram above, it’s metal and lumber producers, not oil and cereal ones.

Of course, a commodity’s position in the product space is just one of the relevant attributes of a commodity for development. There are potentially many more.

For instance, in a recent paper my co-authors and I show that over a hundred-year space of time, the volatility of the price of a country’s initial basket of commodities is one of the strongest determinants of long-term growth, in part because volatility diminishes investment and reduces political stability.

A lot more thinking needs to go into the relevant attributes of commodities, and what differentiates them. The cross-country regression literature usually lumps “commodities” into a single index. What the above papers illustrate is that it is time for a decomposition, and a return to some of the unfashionable theories of development of the 1920s, 30s and 40s.