Methodological problems in cross-country analyses of economic growth

Many cross country studies have been conducted over the last twenty years to explain how various factors affect economic growth rates in the developing countries. The data in these studies - which underlie international economic comparisons and serve as the basis for economic policy recommendations - give researchers the systematic and scientific information required for their investigations. But the conclusions are often fragile and sometimes contradictory. This paper finds that research results are sensitive to the choices of components, the aim of the investigation, and the type of model used. In general, researchers need to have better statistical data, particularly on economic policy indicators, and must subject the selected sample to careful tests. Cross-country studies are particularly unreliable when it comes to estimating the economic impact of government budgetary and regulatory policies. These studies thus provide only a weak basis for developing country economic policies.