Tax effects on investments

Abstract

This doctoral thesis investigates empirically and theoretically the effect of tax on the composition of the optimal allocation of wealth to risky assets from various points of view. The first empirical chapter considers the effect of tax on a U.K. personal investor targeting domestic financial products. This research helps investors estimate the impact of a future tax change and maximize their portfolio return using a newly proposed optimization model and solution method. Following Bonami and Lejeune (2009), personal portfolios are constrained to meet or exceed a prescribed return threshold with a high confidence level and satisfy buy-in threshold and diversification constraints. Their model is improved by incorporating complex tax trading rules with withdrawal features that enhance those considered by Osorio et al. (2004, 2008). A solution based on Greedy methods is developed to deal with the proposed large-scale portfolio optimization problem. The empirical results report substantial non-linear tax effects on riskier assets and enhanced effects of withdrawal tax only when tax rates are high. The developed framework better enables investors to react to tax changes, and tax policy makers to quantify the influence of tax changes on private investment preferences. The second empirical chapter investigates the effect of an international transaction tax, the so-called ‘Tobin tax’, from the point of view of U.K., U.S., and E.U. personal investors targeting international financial products. This empirical research helps the policy maker to estimate the impact of Tobin tax on international capital flows and, therefore, assess the optimal way to introduce the new tax. An optimization model is proposed to maximize the expected net Sharpe ratio and find the optimal risky portfolio internationally. Complex trading and tax rules are considered. To examine the precise effects of different investment and transaction tax rules, a comparison of four tax settings is presented: source only, residence only, mixed with credit and mixed with double taxation. The experimental results show that a source only tax union has more capital transits in international markets than a residence only tax union, and its optimal market portfolio is more sensitive to regional tax policy. In a mixed tax system, double taxation between residence- and source-taxed markets significantly reduces the attraction of the latter while its attraction is maintained with the credit method. Tobin tax can reduce the volatility of the market but the effect varies with tax rate, certain market specifications (e.g., expected returns and correlations with overseas markets) and investment tax rules. It does not depend on which side of the capital flow (inflow or outflow) is subject to Tobin tax. Finally, an agreement among countries to produce a consistent Tobin tax rate globally can significantly reduce the negative effect of Tobin tax on capital flows while retaining its positive effect on market stability in comparison to heterogeneous Tobin tax rates. Finally, the third analytical chapter investigates theoretically the effect of tax from the point of view of an arbitrageur. This theoretical research addresses the condition of the existence of arbitrage opportunities on an after-tax basis, helping the policy maker improve the fairness and efficiency of markets by addressing effective tax policy. To track tax arbitrage, continuous time optimization models are developed with heterogeneous taxation between investors programmed with continuous rather than static income and capital gains (or losses). It is proved analytically that arbitrage opportunities exist for both perfectly correlated and non-perfectly correlated assets. For perfectly correlated assets, the analysis shows that tax arbitrage may exist, with the investor’s top tax rate and some static asset parameters determining the existence of arbitrage opportunities. It is also proved that many of the equilibria obtained under income tax only are not optimal if investors are also subject to capital gains tax. For non-perfectly correlated assets, however, it is the market prices of cap and floor options on asset returns that decide the existence of tax arbitrage. In the government fixed income bond market, tax arbitrage between investors is difficult to eliminate unless investors are all subject to the same tax rates. But the return from this arbitrage can be limited if the government applies the same top tax rate to all investors.