Differentiating cyclical and long-term income elasticities of import demand

Determining how imports react to cyclical and secular (long term) factors has been a recurrent theme in the empirical trade literature. The evidence suggests that cyclical income elasticities of import demand are generally higher than long term elasticities - particularly for basic materials and semi-manufactured goods. Traditional models generally underestimate the cyclical response in imports, and overestimate the long term response. For example, estimates of income elasticity using a traditional import model average 1.4 and 1.2 respectively. The authors'model suggests a cyclical elasticity averaging 2.6. This result suggests that the two elasticities may differ by an even larger factor for developing countries. Relative prices generally are more important in determining import demand in Latin America and Asian-Pacific countries, in this model, but seem to have little effect in the African and (surprisingly) Mediterranean countries. In countries for which both cyclical and long-term income elasticities are significantly different from zero, relative price coefficients are also significantly different than in countries for which income parameters are not significantly different from zero.