Imperfect labor mobility and unemployment in LDCs: comment.

• Author: Gilbert, John
• Position: Comment on article by Parai and Beladi, Southern Economic Journal, v. 64. p. 180

1. Introduction

   In a recent paper in this journal, Parai and Beladi (1997; henceforth PB) extend the well-known Harris-Todaro model to consider imperfect labor mobility. They find unambiguous support for protection in the case of a small open economy with unemployment and imperfect labor mobility (Proposition VII). We show that the PB conclusion that the optimal tariff is always positive is incorrect. Furthermore, this note contrasts PB's model with the existing literature on the open economy with unemployment. We show that in the limiting case of perfect labor mobility, their model does not produce results that economic intuition and the existing literature would lead us to expect. We believe that their specification of the elasticity of labor mobility has no economic rationale within the context of the Harris-Todaro framework. We briefly explore a simple alternative specification.

2. The Parai-Beladi Analysis

   PB introduce a two-sector (urban/rural), two-factor (labor and capital), dual, constant-returns-to-scale economy in which capital is perfectly mobile within the domestic economy. Like Corden and Findlay (1975) and Batra and Naqvi (1987), PB also incorporate labor unemployment specific to the urban sector by means of the Harris-Todaro expected wage mechanism. Their contribution is to introduce the issue of labor mobility by utilizing a simplified version of Casas's (1984) specification of the elasticity of labor mobility. The key result of PB (Proposition vII) is that imperfect labor mobility in the presence of unemployment always implies a positive optimal tariff on the output of the urban sector (the importable) for a small economy.

   In the case of a small open economy with perfect labor and capital mobility and sector-specific unemployment of the Harris-Todaro category, it has been shown that a small import tariff on the output of the urban sector causes a welfare loss because, in addition to the usual production and consumption distortions, it also increases the rate of unemployment in the urban sector. This result was first geometrically explored in Corden and Findlay (1975) and has been rigorously proven by Batra and Naqvi (1987).

   PB state that they "extend the Casas model further by incorporating unemployment caused by sector-specific sticky wages of the Harris-Todaro variety." This means in the limiting case of perfect labor mobility ([Epsilon] = [infinity], where [Epsilon] is the elasticity of labor mobility), their model must converge to the open economy Harris-Todaro model with mobile capital. It is clear that PB believe this to be the case. Therefore, we...