Tax Avoidance, Dividend Signaling and Shareholder Taxation in an Open Economy

Abstract: Essay 1: The first essay contains an approach to calculate the avoidance from the income tax and the consumption tax using National Accountings data. Using Swedish data from 1994, the empirical findings indicate avoidance from both these taxes. Cross-border shopping appears to be small, 656 million SEK or 0.07% of total consumption, while avoidance from the income taxation is more extensive, amounting to some 41 billion SEK or 5% of total income. These results imply that to lower the costs from tax avoidance, the income tax rate should be reduced and the general consumption tax rate should be raised. Essay 2: In this essay I investigate "the information content of dividends" by estimating a model to test for the information in dividend changes. The model considers the relationship between dividends and permanent earnings in firms. As firms may change the dividend level for several reasons, the model distinguishes between dividend changes combined with dividend smoothing and the dividend changes that may contain information about the future Permanent earnings. Estimating the model on Swedish firm data from 1979 to 1996, I find that some of the dividend changes undertaken by Swedish firms contain information about future changes in the firm earnings. This finding indicates the existence of asymmetric information about the future earnings between firm managers and investors. Despite several important differences between Sweden and USA, the information coefficient takes almost the same value in both countries, which indicates robustness. In addition, using Monte Carlo simulations I consider the problem with stepwise movements or stickiness in the dividend process. Essay 3: It is often argued that the pre-tax required rate of return on equity in a small and open economy like Sweden is internationally determined. Changes in domestic personal taxes on dividends and capital gains affect the after-tax rate of return of the domestic investors, but have no influence on the cost of capital or on the firms' financial behavior. This essay attempts to shed some empirical light on the on-going debate about the economic effects of taxing dividends and capital gains. I follow two somewhat tentative approaches and make use of Swedish and Nordic firm data. Taken together, the results indicate that the influence from personal taxation on market valuation and financial behavior is insignificant for the firms in our samples. Given that the combined findings from the approaches support the standard view that large firms in Sweden face an internationally determined rate of return requirement the second main question I consider is the impact of personal taxes on those firms that are confined to raising equity funds from domestic investors.