Long-run substitutability between more and less educated workers: Evidence from U.S. States 1950-1990

We estimate the aggregate long-run elasticity of substitution between more and less educatedworkers (the slope of the demand curve for more relative to less educated workers) at theUS state level. Our data come from the (five)1950-1990 decennial censuses. Our empiricalapproach allows for state and time fixed effects and relies on time and state dependentchild labor and compulsory school attendance laws as instruments for (endogenous) changesin the relative supply of more educated workers. We find the aggregate long-run elasticity ofsubstitution between more and less educated workers to be around 1.5.

Evaluating predictive densities of U.S. output growth and inflation in a large macroeconomic data set

We evaluate conditional predictive densities for U.S. output growth and inflationusing a number of commonly used forecasting models that rely on a large number ofmacroeconomic predictors. More specifically, we evaluate how well conditional predictive densities based on the commonly used normality assumption fit actual realizationsout-of-sample. Our focus on predictive densities acknowledges the possibility that, although some predictors can improve or deteriorate point forecasts, they might have theopposite effect on higher moments. We find that normality is rejected for most modelsin some dimension according to at least one of the tests we use. Interestingly, however,combinations of predictive densities appear to be correctly approximated by a normaldensity: the simple, equal average when predicting output growth and Bayesian modelaverage when predicting inflation.