Fixed-Effect Versus Random-Effects Meta-Analysis in Economics: A Study of Pass-Through Rates for Alcohol Beverage Excise Taxes

Abstract

This paper compares two methods for meta-analysis: fixed-effect models and random-effects models. Both models are applied to pass-through rates of excise taxes on alcohol beverages. Using a sample of estimates from 30 primary studies, weighted means are first reported for each method and compared against a fully-passed tax or unitary rate Dispersion and heterogeneity statistics are used to assess the performance of each method. Second, means and dispersion statistics are reported by subgroups for country source; beverage (beer, wine-spirits); and published status. Third, tests are conducted for publication selection bias using funnel plots and regression asymmetry tests. Fourth, three procedures are undertaken to reduce selection bias: trim-and-fill; cumulative meta-analysis; and meta-regressions. Based on a variety of tests and procedures, three conclusions are reached. First, a random-effects model is more appropriate for these data, reflecting diverse estimates of pass-through rates. Second, pass-through rates are approximately one regardless of beverage. Third, greater attention needs to be given to the choice of model for meta-analysis in economics.