EXPORT INSTABILITY, INTERNATIONAL BUSINESS CYCLES, AND COMMODITY CONCENTRATION OF LDCS (PORTFOLIO

This dissertation advances two hypotheses. The first is suggested by empirical evidence from the 1950s and the 1960s and maintains that the mean export earnings instability of LDCs is directly related to the degree of synchronization in the business cycles of industrial, capitalist countries. Additional evidence supporting the hypothesis emanates from numerous empirical studies of the 1970s, as well as from one of the two sets of independent tests that were conducted here, which covered the period between 1950 and 1982. This suggests that demand side forces may be responsible for the higher mean export instability of LDCs during periods of relatively synchronized business cycles in industrial countries and contradicts empirical studies that had attributed LDC export instability to supply side (country specific) forces.;The second hypothesis is derived from an optimal portfolio selection model that assumes risk-averse preferences. It maintains that the sign of the relationship between "proportions" diversification and mean export instability will be ambiguous, while that between "numbers" diversification and mean instability of export earnings will depend on the correlation coefficient between export receipts: it will be negative if the correlation coefficient is negative (or positive but low), positive if the correlation coefficient is highly positive. This contradicts results of earlier "portfolio" models and, furthermore, suggests that the preference for numbers diversification of exports, even though it may lead to a more unstable export portfolio, is nothing but the response of risk-averse individuals to constrained choice. Support for this hypothesis is derived mainly from the internal consistency of the mathematical model from which it emerged; empirical confirmation of the hypothesis must await the development of suitable concentration indices and reliable data on number of exports of LDCs. Export diversification will reduce LDC export instability if export stabilization programs are managed with attention to the degree of randomness and synchronization in the business cycles of industrial countries, and it should not be discouraged even though it may lead to greater export instability in periods of highly synchronized business fluctuations in industrial countries.