Joseph Stiglitz gave today’s Jean-Jacques Laffont lecture at the Toulouse School of Economics’ Tiger forum, talking about his new book, Creating a Learning Society: A New Approach to Growth, Development and Social Progress. The two themes of the lecture were that sustained growth needs a ‘learning society’, and that markets alone can’t create this.
Stiglitz argued that the social changes around 1800 were more important than allocative efficiency or capital accumulation, consisting of the learning how to learn or diffusion of knowledge. The idea that technological change is the most important determinant of growth encompasses the diffusion of knowledge. He linked this to Schumpeter, but the idea is obviously part of endogenous growth theory too.
The lecture focused on developing countries, and the transfer of knowledge and know-how from the developed countries. This, Stiglitz said, was more important than the transfer of resources.
Market failures are pervasive in this arena, he said. Knowledge is a public good. Markets alone do not produce and spread knowledge efficiently and do not promote innovation efficiently. Government interventions are needed. He highlighted intellectual property regimes, arguing that these need to be ‘developmentally orient’ – and also that the WTO’s TRIPS regime is very much not so, imposing costs that are not outweighed by the dynamic benefits of protecting innovations.
He went on to discuss the relationship between globalization and growth, suggesting that growth drives trade as much as trade drives growth (the orthodoxy). Indeed, there is an argument that trade liberalization has reduced growth in the developing world.
The task of government is not picking winners – although he said en passant that governments in general are quite good at doing so, getting a good return on average in research and development – but in identifying externalities. however, the WTO regime has made it hard for governments to engage in industrial policy – in contrast to examples like Korea in the past. An alternative, and one that avoids the ‘picking winners’ argument too, is to reduce the exchange rate. This would tend to give them sustained current account surpluses, with the result that the poorer countries will be lending to the richer countries, but this is worthwhile because of the ‘learning’ benefits.
As soon as we are talking about innovation, he concluded, we’re in the world of ‘second best’ or Schumpeterian rather than neoclassical economics, and especially not in the world of the Washington Consensus when it comes to development. Finally, he argued that governments need to direct innovation away from labour-saving innovations; markets not only under-invest in innovation, but also are inefficient in what kind of innovation they deliver because the social and private costs differ. What is the point of allocating scarce capital and knowledge in order to create more unemployment? he asked.
I must say that the rather superficial summary of the book in the lecture left me with many questions, but it sounds like it offers an intriguing perspective on development. If you take modern growth theory (or Schumpeter) seriously, knowledge is a central issue. And Stiglitz after all is a titan when it comes to understanding the implications of information asymmetries.