Investment and neutrality of capital income taxation: testing the king – fullerton framework

The King-Fullerton (1984) approach to the measurement of effective marginal rates of taxation on capital income has been applied extensively. The empirical literature confirms that the dispersion of effective tax rates, both within each country and among countries, is wide. However no test has been proposed of the impact on investment of these findings. The paper tests the conjecture proposed by King and Fullerton that standard deviation of effective tax wedges and investment or growth are inversely correlated. We find that international evidence does not support the conjecture: while investment and tax wedges are negatively correlated, as expected, the standard deviation of tax wedges seems to be positively correlated to investment. We propose the following interpretation of this result: a substantial part of the variance of effective marginal rates is explained by an implicit subsidy to equipment investment, and this policy may be justified by De Long-Summers (1992) results or by the view that this type of investment have more impact on growth than other types (such as buildings and inventories). In some cases also a partial deduction of interest rates may be rational when capital markets are rationed. Thus it may be simplistic and misleading to use King- Fullerton statistics as yardsticks for tax reforms leading to uniformity of effective marginal tax rates.