THIS DISSERTATION examines the short-run relationships between domestic and foreign investment of U.S. manufacturing firms and the financing of such investment. A short-run framework provides results relevant to American balance-of-payments policy; in the long run, repatriated earnings, fees and royalties, and (possibly) increased exports offset some or all of the initial balance-of-payments cost of direct investment. Thus, if direct investment outlays are recouped in the U.S. balance-ofpayments, changes in such outlays affect primarily the timing of balance-of-payments deficits. To maximize a discounted stream of worldwide profits, management of a firm makes interrelated decisions concerning domestic and foreign investment. To the extent that foreign and domestic investment are related, foreign investment, and therefore balance-of-payments flows, will be related to the level of domestic economic activity as it impinges on the firm. In addition, the financing of direct investment may be related to domestic activities of the firm. The decision variables in the study are foreign and domestic plant and equipment expenditures, dividend payments to stockholders, and outflow, defined as net capital outflow less repatriated profits. The last variable represents the immediate effect of direct investment on the balance of payments. In general, foreign investment is a function of sales changes, foreign and domestic income, foreign depreciation allowances, dividends, domestic investment, and the leverage of the firm. Specifically, inducement to invest is represented by lagged sales changes and by expected sales changes, as proxied by the firm's stock price relative to the stock market. The domestic investment equation is specified analogously. The dividend equation follows a typical partial adjustment form but allows for separate reaction to a transitory component of current income. Dividend policy is also allowed to respond to investment and leverage. Net outflow is expected to be positively related to both domestic income and foreign investment and negatively related to foreign income, dividends, and domestic investment. It may be negatively related to domestic investment even if foreign and domestic investment are not substitutes. Separate firm intercepts are used to control long-run differences between firms, and thereby to isolate short-run behavior. Appropriate scaling is made to adjust for existing interfirm differences in foreign activity. The data are then normalized by previous-year capital stock of the appropriate sector (foreign, domestic, or total) of each firm. Two-stage least squares estimation is applied to annual data for 63 large U.S. manufacturing firms. Estimation is for the period 1961-1966, a period in which direct investment was relatively free from capital restrictions of host countries and the United States. A dummy variable represents the impact of firm-level targets in the 1966 U.S. voluntary balance-of-payments program. Its positive but insignificant coefficient suggests that firms, anticipating stricter controls, may have increased their foreign investment.
Read full abstract