Balance of payments constrained growth model: an examination of Thirlwall's Hypothesis using McCombie's Individual Country Method

Abstract

The balance of payments constrained growth (BOP) model is tested using the McCombie's cross-country and individual country test for 90 developed and developing countries for the period 1980 to 2000, and the McCombie's test is extended using export elasticity. If in the long run, the growth of exports is equal to the growth of imports and the affect of prices are not statistically significant, the estimated import elasticity from time series regression should not be statistically different from the BOP implied import elasticity, this also implied that the estimated export elasticity, also should not be statistically different from the BOP implied export elasticity. However, the finding shows that for only about 45% of the countries under study the individual country test is accepted even though cross-country analysis support the BOP model. Furthermore, for some countries where its estimated and implied import elasticity are not statistically different, its estimated and implied export elasticity are statistically different and vice-versa.

Notes

¹ McGregor and Swales (1985) suggest that the model can be tested by regressing the actual domestic growth rate on the predicted growth rate, and test whether the regression coefficient is equal to unity and the constant term is equal to zero. McCombie criticizes McGregor and Swales on the grounds that the independent variable is itself calculated using estimated parameters and hence is subject to errors.

² Houthakker's two step procedures for foreign export price:

First a price index is constructed for each of the top 26 markets of country i using the export prices of 25 other top exporters to that market weighted by their share of exports in a particular year. Then the resulting 26 price indexes are combined with the same weights used to calculate the foreign GDP for country i. Hence each exporter has a price index comparing the exporter's price with the weighted average of the export prices of its 25 competitors in each of the 26 markets. If this procedure is used, developing countries will compete with industrial countries that are not true. Houthakker and Magee (1969) use these procedures to estimate trade elasticities for industrial countries, hence developing countries is not a big issue.

³ Hussain (1999) shows that 17 out of 29 African countries and 5 out of 11 Asian countries the implied import elasticity is not statistically different from estimated import elasticity. Using elasticity estimation from Wu (2003) it was found that 9 out of 34 the implied import elasticity is not statistically different from estimated import elasticity and 15 out of 34 the implied export elasticity is not statistically different from estimated export elasticity.

⁴ Hussain's (1999) extended model (including capital flows) accepted that implied and estimated import elasticity are not statistically different for 23 of out 29 cases for African countries and 6 out of 11 cases for Asian countries.

⁵ Using the strong form of Thirlwall's law, i.e. (Perraton and Turner, 1999) does not change the result.