New research by John Landon-Lane and Peter Robertson, to be presented at the Royal Economic Society’s Annual Conference on Wednesday 9 April, compares average growth rates across 17 OECD countries in the twentieth century. They find that, except for a 30-year period around the Second World War, there are no significant differences in the average growth rates of all 17 countries. Based on this observation, the authors argue that the range of economic policies used in these countries have had little or no impact on their respective long-run growth rates.

The authors follow the standard practice of differentiating between short-term differences in growth rates, for example due to business cycles, and the long-run trend. Despite the presence of short-term differences, they find that they cannot dismiss the possibility that all 17 countries were following the same long-run growth path between 1904-31 and from 1961 onwards.

Specifically, the counter claim, that at least one country has a different long-run growth rate, has a 50% of probability of being false. Thus, the data offer little support to the notion that growth rates across countries are different, and the odds either way are equivalent to the outcome of a coin toss.

Landon-Lane and Robertson suggest that this fact has several interpretations. Most importantly, it raises the possibility that the various government policies had no effect on long-run growth rates.

This is consistent with a view that economists dub ‘old growth theory’ following independent research papers by Robert Solow and Trevor Swan in 1956. Solow and Swan argued that changes in the investment rate have only temporary effects on economic growth, and long-run growth rates depend on non-economic factors such as scientific discoveries.

So-called old growth theory has been challenged over the last 20 years. In particular, economists have attempted to determine whether there is a relationship between the economic policy environment and technological progress through, for example, scientific discoveries.

This ‘new growth theory’ suggests that policies that promote education and research and development (R&D) could propel an economy along a superior growth path. In particular, this view is endorsed by the OECD’s 2001 study of economic growth and the ‘new economy’. Landon-Lane and Robertson’s findings can be interpreted as casting doubt on the applicability of these new theories in developed market economies.

Their results also shed new light on the phenomenon of convergence. In a well-known research paper, William Baumol found that the less wealthy OECD economies at the start of the last century grew faster, so that per capita income levels across these countries were converging during the twentieth century.

Landon-Lane and Robertson show that the finding of convergence depends crucially on the differences in growth rates that occurred during wartime and recovery years. Thus, it cannot be generalised to the first or last 30 years of the twentieth century.

This in turn casts doubt on the usual interpretation, that convergence was due to a steady process of technological diffusion. To the contrary, convergence may have been linked to the fortunes of war.

ENDS

Notes for Editors: ‘Can Government Policies Increase National Long-run Growth Rates?’ by John Landon-Lane and Peter Robertson will be presented at the Royal Economic Society’s 2003 Annual Conference at the University of Warwick on Wednesday 9 April.

Landon-Lane is at Rutgers University; Robertson is in the School of Economics at the University of New South Wales, Sydney 2052, New South Wales, Australia.

For Further Information:
Before and after the conference: contact Peter Robertson on +61-9385-3367 (email: p.robertson@unsw.edu.au); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).
During the conference (7-9 April 2003): contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).