This report, after a brief survey on the effective tax burden of firms, investigates how the corporate income tax affects the investment behaviors of firms in Korea. Using firm-level panel data (Kis-value provided by Korea Credit Evaluation and Information) between 1983 and 2004, we calculate the effective average rates of the corporate tax, which are to be combined with empirical investment equations based on Tobin's Q model and error correction model. Both types of tax-adjusted investment equations are presented in form of an unbalanced dynamic panel data model and estimated by applying the first-difference GMM and system GMM of Arellano and Bond(1991) and Blundell and Bond(1998), respectively. Since the estimated asymptotic standard errors of the efficient two-step GMM estimator can be severely downward biased in small samples, we need to put more emphasis on the statistical inference based on the simple one-step GMM or Windmeijer(2005)'s robust version of two-step GMM. For various choices of instruments, cash flow and sales growth rate turn out to have both statistically and quantitatively significant effects on investment in fixed capital. The coefficients of the tax term, although having a negative sign, are not statistically significant in most cases. Even in a limited case where statistical significance of the tax effect is confirmed, the tax elasticity is far below the level of economic importance, implying that the tax burden does not play an essential role in the decision of investment.