ساختار صنعت و انعطاف پذیری بازار کار در بخش تولید آفریقای جنوبی: سری زمانی و رویکرد داده های پانل

This paper presents a joint analysis of labor market flexibility and product market structure. Our investigation confirms earlier results of imperfect competition in South African manufacturing where we find an average mark-up of 50% for the period 1970 to 2004 that is without consistent trend over time. The contribution of the paper is to provide a theoretically grounded means of linking output market conduct to labor market flexibility. We infer the proportion of labor associated with rigidities in the labor market from the mark-up, and find that two thirds of total labor employed in South African manufacturing is associated with rigidities. We find that this proportion falls during the 1980s and rises during the 1990s, suggesting an increase in labor flexibility followed by a decrease.

This paper examines a new approach of empirically identifying the extent of labor market flexibility. The contribution of the paper lies in numerically estimating labor market flexibility, linking labor market characteristics to the structure of output markets.2 The empirical application of the approach adopted is to the manufacturing sector of South Africa, and provides a series of time-varying measures of the extent of inflexibility in sectoral labor markets. There is a strong labor market debate in South Africa, centered on its high rate of unemployment. A number of potential causes have been identified in the literature. Examples of these include the level of real wages (Fedderke et al., 2000), information asymmetries (Wittenberg, 2002), the impact of technology (Fedderke et al., 2000) and trade liberalization (Edwards, 2001). An important subtext of the debate is the degree of labor market flexibility. While an early study dismissed this as a potential cause of poor labor market performance,3 labor market flexibility has consistently been indicated as a source of concern in South Africa.4 To date there has been no theoretically consistent quantification of the degree of labor market flexibility in South Africa. This paper fills that gap by proposing a theoretically derived measure of flexibility. We investigate industry structure in South African manufacturing by analyzing mark-ups of prices over marginal cost. This is performed for the manufacturing sector as a whole and individually for its twenty-eight sub-sectors in an analysis that extends from 1970 to 2004. We establish average mark-ups as well as investigating the trends and changes in mark-ups for this period, using both a computational approach and an econometric approach. Both approaches follow from a methodology developed by Roeger (1995). Oliveira Martins and Scarpetta (1999) proposed an extension of Roeger's methodology that allows for the investigation of short-run dynamics in mark-ups. They considered the possible impact of downward rigidities in the labor market through the inclusion of an additional labor adjustment parameter. This parameter can be interpreted as a measure of labor flexibility. We use our mark-up results to extract estimates of this parameter. As in the analysis of mark-ups, this is done for the manufacturing sector as a whole and individually for its twenty-eight sub-sectors. We also investigate both long-run means and trends and changes of this parameter over time. This paper extends the work of Aghion et al., 2008 and Fedderke et al., 2007 by obtaining more recent estimates of mark-ups for South African manufacturing, as well as establishing the extent of labor market flexibility that derives from product market structure. The primary contribution of this paper, however, is the provision of a labor flexibility analysis in which we obtain numerical estimates of the magnitude of the flexibility and its changes. Our paper is organized as follows: Section 2 provides details on the relationship between productivity residuals and the mark-up. It also establishes the link between labor market flexibility and the level of the mark-up in output markets, concluding with an overview of relevant previous empirical results. Econometric methodology and data are presented in Section 3. Section 4 reports the results for the estimation and computation of the mark-up and labor market flexibility respectively. Our concluding comments form Section 5.

We find evidence of imperfect competition and pricing power in South African manufacturing through the presence of an aggregate mark-up of around 50% for the sector over the period from 1970 to 2004. When looking at the trend of the mark-up over time, our results suggest that the mark-up fell from a moderate base during the 1970s and 1980s before rising again after the liberalization of the South African economy in the early 1990s. The analysis of individual sectors reveals strong variation in the magnitudes and trends of mark-ups at the sectoral level. Labor market flexibility (particularly the ease of labor adjustment) is investigated through an extension of the mark-up analysis. It is found that of the total labor employed in manufacturing in South Africa, approximately two-thirds is associated with rigidities in the labor market. We also look at the trend of this proportion over time. Results indicate that there is an increase in labor flexibility in the 1970s and early 1980s (a lower proportion of labor is associated with rigidities in the labor market) followed by a decrease in labor flexibility in the 1990s. Labor legislation that was introduced in the 1990s is identified as a possible reason for this. Evidence also suggests a declining elasticity of substitution between capital and labor. An immediate implication is that the concerns regarding employment generation prospects in South African labor markets, due to rigidities in the pricing of labor (see for instance Bannerjee et al. 2008), compound. Improving flexibility of the South African labor market as a means of lowering the very high rates of unemployment becomes a core policy priority.