Although considerable attention has been directed in the last few years to the -effects of capital inflows on saving behavior in less developed countries (see, for example, Weisskopf (1972), Papanek (1972)), relatively little work has been done on the implications of this research for economic growth. Even if domestic saving is reduced by flows, one may legitimately ask, with Grinols and Bhagwati (1976, p. 416), How much does this matter? Grinols and Bhagwati's answer to this question is relatively optimistic: It is certainly possible for the usual, static, adverse impact of external capital to be eliminated quickly by a growing economy receiving the capital (p. 429). In this note I will argue that sustained adverse effects from external capital become quite likely when Grinols and Bhagwati's work is extended. With data from Puerto Rico I will show that the analytical conclusions are empirically relevant. Grinols and Bhagwati look at an interesting but limited case, that in which aid increases by a constant amount once and for all; their conclusions are weakened when the analysis is extended to growing aid streams. Their analysis can also be extended to bring out the implications of dependence on aid for return flows of profit and interest. Although it is not entirely clear from their article, Grinols and Bhagwati define aid as grants plus net capital inflows less net factor income outflows (profit, interest, royalties, etc.). In lumping together the net capital inflows and the current account factor income outflows (which inevitably follow dependence on external saving), they have removed from consideration the relationship between these flows. They have also ignored the distinction between GDP (the output produced within the country) and GNP (the income received by local residents). Net capital inflows can lead to a growing ratio of net external debt to the capital stock and therefore to a growing gap between output and income. In the note that follows, refers to grants plus the net capital inflow alone. (In fact, grants may be seen either as negligible or as fixed in proportion to the net capital inflow.) The return to aid-profit and interest-is considered separately and explicitly. The extensions to Grinols and Bhagwati's work are incorporated into a simple model which is solved analytically. The analysis leads to the conclusion that dependence can indeed be a more persistent phenomenon than Grinols and Bhagwati imply. The Model
Read full abstract