Why the conventional test of Thirlwall’s law is still not a ‘near-tautology’: a rejoinder to Professor Blecker

Abstract

Blecker (2021) repeats an argument, first put forward by McCombie (1981) (but later repudiated by him) and also summarised by Blecker (2016, 2021) that the traditional test of Thirlwall’s law is merely estimating what Blecker terms a ‘near-tautology’ or ‘near-identity’. This is based on an analysis concerning the arithmetical calculation of the import and export income elasticities of demand, and the circumstances under which these will equal their statistical estimates. The arithmetically calculated income elasticities are the growth of imports divided by the growth of domestic income and the growth of exports divided by the growth of world income, respectively. Blecker’s argument is that if the statistical estimates of the two income elasticities of demand equal their respective arithmetically calculated values, and are used in the testing of the law, this represents a near-tautology. This rejoinder demonstrates that Blecker’s argument concerning the near-tautology is a misinterpretation of the nature of the traditional testing of Thirlwall’s law. It is shown that Blecker and Ibarra’s (2013) alternative model, which is used to test the balance-of-payments constrained model for Mexico and which it is claimed avoids the putative near-tautology problem, is itself problematic. As such, it does not represent an improvement over the traditional model of Thirlwall’s law, but, ironically, may give approximately the same empirical result.