Several attempts have been made to determine the tax differential between the corporate and non-corporate sectors of the economy, implied by the present double taxation of corporate source income. A common feature of these studies is the assumption that the retention of corporate profits gives rise to capital gains on a one-for-one basis. By this assumption, the tax burden on retained earnings is identified with the tax on capital gains. In view of the preferential tax treatment given to capital gains, it is, however, quite rational for a management to undertake investments that produce less than a dollar's worth of capital gains for the marginal dollar of retention. To establish this assertion and its implications for the firm's effective tax burden, a theoretical model of firm behaviour is introduced. Specifically, the cost of capital to a firm maximizing stockholders wealth is derived, with due adjustments to the corporation income tax, stockholders' income tax and capital gains tax. In this way, the differential tax burden on corporate source income may be determined with explicit reference to the firm's cost of capital.