Wage dispersion and trade in Argentina, 1974-1994 (original) (raw)

A prominent argument regarding the effects of trade liberalization is presented clearly by Krueger (1990), who argues that trade liberalization in less developed countries (LDCs) will generally compress the wage gap between more and less skilled labor. This reasoning builds from the Hecksher-Ohlin and Stolper-Samuelson theorems. In these models free trade substitutes for factor mobility, and trade liberalization leads to growth in sectors where countries have comparative advantages, causing factor prices to converge internationally. We will refer to this as the "extended Hecksher-Ohlin-Samuelson" hypothesis, or "HOS-X". It argues that for LDCs comparative advantage generally lies in their stocks of unskilled labor, while protectionism distorts prices in favor of capital. Because capital and skill are complements, protectionism raises the demand for skilled versus unskilled labor. Therefore, moving from protectionism to trade liberalism should shift the composition of output and employment towards sectors intensive in unskilled labor, raise the relative demand for unskilled labor versus skilled labor, and increase the wages of unskilled workers relative to the wages of skilled workers. An opposing argument is made by the "New" trade theorists (e.g. Grossman and Helpman, 1991; Stokey, 1994). In their view, trade liberalization leads to larger markets, which in turn induces greater Research and Development (R&D), increases the stock of technological knowledge, and reallocates employment toward innovative activities requiring more education. Through these interrelated channels, they advance the opposite hypothesis that trade liberalization raises the return to human capital, driving up the wage gap between skilled and unskilled workers. We will refer to this hypothesis as Skill Enhancing Theory or "SET". Recent work by Robbins (1995a) analyzing changes in the dispersion of wages in Chile after its major trade liberalization reforms provides evidence in favor of the "New" trade theorist arguments and against the HOS-X hypothesis. Robbins showed that trade liberalization in Chile led to an increase in the wages of more skilled relative to less skilled workers (hereafter called "relative wages"). He further argued that relative wages rose because trade liberalization in Chile led to an increase in both between-and within-industry demand for more skilled workers. This paper examines how wage and employment structure in Argentina changed over the 1974-1994 period. As a by product of this analysis, this study also provides important information on schooling. Because studying trade's impact on wage dispersion requires controlling for the relative supply of skills, this paper documents the changing distribution of educational supply through time, and analyzes its impact upon wage dispersion. The rest of this paper is organized as follows. Section I summarizes the Argentine experience with trade liberalization during the mid 70s and late 80s. Section II presents the data. Section III presents the methodology and findings for a desagregated non-parametric methodology measuring relative wage and relative supply shifts. Section III also estimates a time-series of relative demand shifts, finding that relative demand became more skill intensive after 1976 and after 1986. Finally, in Section III we consider, and reject, alternative non-demand based explanations for this rise in relative wages with trade liberalization. Section IV examines the pattern and causes of relative demand shifts in two parts. First, we decompose employment shifts into "between-industry" and "within-industry" components, finding that-counter to standard trade theory-between-shifts favored more skilled workers after 1976.