An Econometric Test of Alternative Constraints on the Growth of Underdeveloped Countries
T HE of the role of foreign assistance in economic development in terms of two gaps a savings gap and a trade gap has by now become quite commonplace among development economists.' The basic notion underlying analysis is that there are at least two independent resource constraints on the growth of an underdeveloped economy. The first of these is the savings constraint: the growth of the economy is limited by the availability of savings for investment. The second is the trade constraint: the growth of the economy is also limited by the availability of foreign exchange for importing specific commodities required for current production and investment. Since foreign assistance can add both to the availability of savings and to the availability of foreign exchange, and since in a two-gap disequilibrium situation only one of the two constraints is likely to be binding at any particular time, the productivity of foreign assistance depends critically on which constraint is in fact binding. The determination of the binding constraint is also essential for the empirical implementation of two-gap planning models. Such models involve among other things aggregative functions describing the savings behavior and the import requirements of an economy. In a twogap situation, observed savings may or may not equal desired savings, and observed imports may or may not equal required imports, depending on whether the savings or the trade constraint is binding. In order to estimate savings and/or import functions from historical time series data, one must make a judgment about which constraint was in fact binding during the period used for estimation.2 Accepting the general framework of two-gap analysis, this paper is addressed to the problem of determining statistically which of the two constraints was dominant in a given country during a given historical time period. The objective of the paper is to develop and apply an econometric method that can be used systematically to classify countries according to their dominant constraint over varying periods of time. Given the importance of the problem for twogap analysis, it is surprising that very few attempts have been made to test systematically for the dominant constraint. Case studies of individual countries have often led to conjectures about binding constraints, but this author is aware of only three attempts to classify a large number of countries by dominant constraint.1 The present paper attempts to develop a more satisfactory method of classification. In
- DOI
- 10.2307/1927496
- Volume
- 54 (1)
- Pages
- 67
- Export
- BibTeX
- Sources
- openalex crossref