Studies on total import demand in the GCC countries can be classified into two groups. In one group, researchers have taken the view that in developing countries, import demand is price inelastic and specified import demand not to depend on prices of any sort. In a second group, research have adopted traditional specifications and included relative prices in their models. Nonetheless, there exist empirical evidence that models employed in these studies are incorrectly specified. The immediate criticism concerns the empirical definition of the price variable.
This paper presents a detailed examination of the empirical hypotheses implied by specifications used by previous studies on the GCC countries.
A generalized specification of import demand in which prices of imports, of tradables, and of non tradables enter as arguments is developed. Separate price indexes required for estimating such model are constructed.
Sequences of nested hypotheses of model specification were identified via the imposition of series of constraints leading to the conventional alternatives. Implied hypotheses were sequentially tested. The main inference drawn is that all of the identified prices related assumptions maintained conventional specifications are decisively rejected. An interesting result emerged that, in Kuwait, imports and nontradables are rather complements.