The Ill-Posed Problem in Growth Empirics

A problem encountered in growth empirics is that the number of explanatory variables is large compared to the number of observations. This makes it impossible to condition on all regressors when determining if a variable is important. We investigate methods used to resolve this problem: Extreme bounds, Sala-i-Martin’s test, BACE, general-to-specific, minimum t-statistics, BIC and AIC. We prove that the problem in general is ill-posed and that the existing methods are inconsistent. We propose a test and apply it to determine if "good policy" increases the effectiveness of foreign aid on growth. The test rejects inference regarding good policy.AIC; BACE; BIC; extreme bounds; general-to-specific; ill-posed inverse problem; robustness