June 15, 2012
By Tim Dowd (Joint Committee on Taxation), Robert McClelland (CBO), and Athiphat Muthitacharoen (CBO).
Updated on August 9, 2012, with minor corrections to equation 3, to Table 6, and to references to Table 6 in the body of the paper.
This study uses a large panel of tax returns from 1999 to 2008 to investigate how taxes affect the decision to realize gains. The study distinguishes the persistent effect of tax changes from the transitory effect. Similar to earlier studies in the literature, we use the generalized Tobit model to address the sample selection problem and the endogeneity problem in the tax variables, but we improve the identification of the tax elasticity by using the presence of carryover loss as an exclusion restriction. We also control for the financial sophistication of taxpayers because that could be an important source of omitted variable bias. The preferred persistent elasticity estimate is -0.79, and the transitory estimate is -1.2. Those estimates are statistically significant and are robust to a number of sensitivity tests. Although we focus our examination on personal capital gains, we also compare the results of our model to results from the original model applied to contemporary data, estimate our model on subperiods, and estimate our model on other types of capital gains. We find that passthrough capital gains are highly sensitive to persistent tax changes, but gains from mutual fund distributions are extremely insensitive.