ILPC 2026

For decades, Chilean scholars claimed that number of people employed in working class occupations would inevitably shrink. According to them, the growth of the service and non-manual sectors, as well as private sector employment, brought about not only the decline of the traditional blue-collar working class, but also the expansion of “middle-class” occupations (León & Martínez, 1987, 2007; Ruiz and Boccardo, 2015). In emphasizing the effects of the structural transformation that accompanied the development of free market economics, scholars argued for the consolidation of a new middle-class society in which traditional patterns of working class identities and organization would play—both in Chile and other Latin American countries—a marginal role (Franco et al., 2011; Franco & León, 2010; Hopenhayn, 2010; Espinoza, Barozet, & Méndez 2013; Ruiz and Boccardo, 2015). While these scholars have not aimed to test the main hypothesis of the labor process theory, some of them have certainly suggested that the employees working in these “middle-class” activities differ from the traditional, manufacturing workers in the way they are subjected to different employment dynamics. They have contended, for example, that middle-class employees enjoy not only higher salaries and wages than “traditional” (blue-collar) workers, but also higher skill levels (León & Martínez, 1987, 2007; Ruiz and Boccardo, 2015). In noting the difference in skills, these researchers have essentially reproduced traditional sociological accounts according to which middle-class employees are more likely to occupy managerial roles as well as enjoy higher degrees of autonomy and control over the labor process (cf. Lockwood, 1989 [1958]; DeFrozo, 1973; Poulanztas, 1975; Goldthorpe, 1982).

This “middle-class” argument has been recently called into question from different perspectives. Recent investigations have contended the changes in the Chilean class structure do not seem to suggest the growth, let alone the consolidation, of middle-class positions. According to these investigations, these changes indicate the persistence of “traditional” patterns of class inequality defined, among other things, by the predominance of working class positions (Gayo, Mendez & Teitelboim, 2016; Perez Ahumada, Forthcoming). In line with this, recent research has also demonstrated that working in non-manual or service occupations does not produce “middle-class” identities and interests, i.e. identities and interests that differ from those upheld by manual, blue-collar workers (Perez Ahumada, 2017).  

            This paper aims to contribute to this emerging literature by drawing upon insights from the labor process theory (LPT) (cf. Braverman, 1974; Burawoy, 1979; Thompson, 1982; Newsome et al., 2015; Thompson & Smith, 2017). Through quantitative evidence (regression models predicting workers’ perceptions of control over the labor process), we test the main hypothesis of the “middle-class argument”. More specifically, we examine whether working in the “middle-class sector” (i.e. in non-manual, service, and private-sector occupations) leads to higher perceptions of control over the labor process. In an attempt to overcome the divide between class analysis and the Labor Process Theory (Carter, 1995), we also use quantitative methods to test whether workers’ class position—i.e. their location in a managerial/supervisory, expert or unskilled working-class position—shapes their perceptions of control over the labor process. In doing so, we also examine whether the effects of class position differ across economic sectors. This quantitative analysis is supplemented by a qualitative study based on interviews with more than 150 workers employed in different sectors of the economy (manufacturing, agro-industrial activities, mining, retail store, banking, clerical jobs in public sector, among others).

Based on this evidence, we show that the labor process in sectors usually described as the core of the Chilean “middle class” (for example, banking, retail store, and clerical activities in the public sector) is not substantially different, in terms of workers’ perceptions of the relations of supervision and managerial control they are subjected to, from traditional “working class” activities (e.g. forestry, agro-industrial, long-shore and port activities). In this sense, our results contradict the basic arguments of the “middle-class thesis”. The findings also suggest, again in contrast to this thesis, that working in the private sector leads to weaker perceptions of control over the labor process. Something similar occurs when we analyze the impact of class position. In line with the main hypothesis of class analysis (Crompton, 1993; Wright, 1997), workers’ class location does generate significant variations in workers’ perceptions of the control over the labor process. Yet, in contrast to the middle-class hypothesis, such an impact does not substantially change when we add an interaction term to measure whether the effect of class differ across economic sectors.

The results of this study allow us to present new evidence to explain why the growth of the service and non-manual sectors should not be interpreted as the structural basis of the alleged expansion of the middle class. They also allow us to bridge the gap between class analysis and the labor process theory (cf. Carter, 1995) by presenting findings on the case of Chile.