Estimation and Interpretation of Capital Gains Realization Behavior: Evidence from Panel Data

摘要

The study has attempted to shed light on why empirical capital gains realization equations have produced such widely varying results and to explore the relevance of panel data to resolve some of those questions. Results suggest that a substantial part of past variance in realization elasticities may have been due to the simultaneity between marginal tax rates and capital gains realizations and to the failure of previous studies to correctly deal with sample-selection bias. In addition, the failure to deal with these sources of biases has resulted in discrepancies that have been blamed on other causes. It was also found that simulation is important for estimating the equilibrium response of individual taxpayers to changes in tax law under a progressive income tax. Some deficiences were also discussed.