Wealth Taxation in a Life-Cycle Model with Entrepreneurship and Housing

Abstract

A standard heterogeneous agent, two-asset life-cycle model is developed, which includes entrepreneurship and a dual role for housing assets. Housing assets serve both as collateral for loans and as a good that generates a service flow. The life-cycle model allows me to replicate the pattern of wealth holdings that is estimated from a partially linear regression, based on data from the Survey of Consumer Finances. Results suggest that, regardless of occupation, young households hold relatively less financial wealth and most of their wealth is in the form of housing assets. The levels and trends of financial and housing asset holdings are quite different between wage earner households and business (entrepreneur) households. In the median wage earner household, the age profile of housing stock is hump-shaped, but it exhibits a flattening out pattern in the second half of lifetime. The holding of net financial assets has an S-shaped age profile. In comparison, regardless of whether the asset type is housing or net financial assets (which include business equity), the asset holding of the median business household grows steadily over the life cycle. A calibrated version of the life-cycle model is developed and used to simulate counterfactual tax policies. These simulations are used to determine if a revenue-neutral introduction of wealth taxes brings about welfare gains (losses) in the U.S. economy. Quantitative inspection of the model performance suggests that a life-cycle model with collateral-based borrowing and entrepreneurship is more successful in capturing the age profile of wealth holdings as well as the fat-tailed distribution of wealth observed in the U.S. data. The quantitative analysis shows that a wealth tax policy reform decreases the aggregate capital stock. This causes a rise in the interest rate, but a decline of the wage rate. Young households are worse off because the decrease of labor income slows their accumulation of wealth, particularly in the form of housing assets. Old households also incur losses in social welfare because the first-order effect of a wealth tax reform is to increase their tax payments and reduce their consumption. With respect to aggregate output, these results suggest that a proportional (flat-rate) capital income tax policy is superior to a progressive wealth tax policy (which has been proposed by Piketty). Changes in the Gini index of the pre-tax income or post-tax wealth distributions indicate that a progressive wealth tax accomplishes a more equal distribution of income and wealth than the proportional capital income tax. I conclude that, from a political economy perspective where people vote according to their gains (losses) of social welfare, a progressive wealth tax is not likely to be favored by a majority of the population when the alternative is a proportional capital income tax.