Abstract
This study investigates the effect that country size has on economic growth. This analysis is prompted by the surprising result of Easterly and Kraay (2000) who find that small states (countries with populations less than one million people) actually grow faster than larger nations once location is controlled for. Easterly and Kraay's results are found to be inconclusive due to the failure of many of their models to pass simple diagnostic tests. Altering the small state dummy variable used by Easterly and Kraay provides a suggestion that there is some quadratic relationship between sze and growth. This is tested using new general growth and income levels equations specified using a simplified form of Hendry's general-to-specific methodology. This analysis shows that country size and economic growth share a quadratic relationship with a turning point at 15.5 million, which represents the population level consistent with the lowest rate of economic growth, ceteris paribus. At a population of 31 million country size begins to have a positive impact on economic growth, ceteris paribus. This finding is not robust to the inclusion of instrumental variables, although this could be due to the weakness of the instruments used for country size despite the Sargan (1964) test being passed.
Therefore, this study presents evidence which argues against the findings of Easterly
and Kraay (2000).