Representative Bureaucracy? Promotions in Sweden’s Local Government Bureaucracy

• In Event: African Politics & Public Administration Virtual Poster Session
  In Virtual Poster Session: Public Administration: Topics in Public Administration and Bureaucracy Virtual Poster Session

Abstract

The idea of passive representation in the bureaucracy is that bureaucrats should mirror the population in terms of demographic traits like gender or race/ethnicity (Mosher 1968). Such representation holds an intrinsic value of social justice (DeHart-Davis et al. 2020). It also improves the quality of output provided to citizens from the represented groups, and increases overall bureaucratic performance (e.g. Meier and Nicholson-Crotty 2006; Park 2012; Pitts 2009). Recent research has progressed from considering whether certain groups hold bureaucratic positions to study their advancement from one bureaucratic position to another. This research takes an interest in bureaucrats’ careers, and specifically in careers preceding leadership positions. As leadership positions offer more decision-making power, under-representation of certain demographic groups on these positions are of particular interest. From a policy-making perspective, the factors that determine why certain groups progress faster or slower than others are also of obvious importance.

This paper studies career progressions by gender and immigrant status in local bureaucracies in Sweden. It makes contributions to the existing literature, many of which are made possible by uniquely detailed administrative data. This exact type of data has recently been exploited to understand which type of citizens become candidates for political office and advance in the political hierarchy. In that literature, there has been what Cirone et al. (2020) call an “administrative data revolution”, an empirical approach that this paper extends to the bureaucracy.

The first contribution is that I study a more nuanced set of demographic groups than the previous literature. Among immigrants, I can distinguish between refugee-based immigration and labor immigration, and differentiate people by regions of birth as a way of measuring racial hierarchies. This intersectional categorization improves on previous micro level studies that use administrative data (e.g. Pierskalla 2020). My intersectional approach receives high statistical precision thanks to the administrative process of collecting data. I have access to all residence permits (refugee, labor immigrant, etc.) going back to the early 1990s, and region of birth is recorded as a mandatory part of the immigration process.

A second contribution of my paper is to study more detailed career moves than previous work. My administrative data from tax records allows the identification of all employees in Sweden’s 290 municipal bureaucracies. In each municipality, bureaucrats can climb the latter to become unit head, department head, and director of the municipal bureaucracy. I can see who moves up these career steps in 13 years of annual data. To my knowledge, this makes for the most detailed study of bureaucratic career progressions to date.

Third, I offer new insights about the determinants of bureaucratic career progression for women and immigrant minorities. I focus on individual factors and estimate the extent to which observable qualifications can explain upward progression. In this framework, a situation where a bureaucratic minority advances slower than the majority while possessing the same qualifications is defined as discrimination (Cotter et al. 2001). This estimation contributes to existing research on institutional and organizational factors that shape passive representation (e.g. Newman 1994; Smith and Monaghan 2013). Compared to papers on individual factors, my approach offers a more extensive set of qualification measures and larger sample sizes (e.g. Bowling 2006; DeHart-Davis et al. 2020; Pierskalla 2020). My data includes both education length and detailed education types, as well as complete labor market histories extending more than 20 years back for every bureaucrat. For men, I also exploit data on IQ and leadership tests from Sweden’s previously mandatory military draft.

Finally, the literature on representative bureaucracy is strongly U.S. centered, and the need for studies situated in other geographical and cultural contexts is increasingly pressing (Bishu and Kennedy 2020). Sweden offers an interesting case, for example as the basis of a discussion of the applicability of the representative bureaucracy framework in countries with different administrative traditions.

Preliminary results show severe career disadvantages for refugee immigrants. They also show slower, but less severe, disadvantages for non-OECD immigrants without refugee status. These slower career progressions cannot be explained by observable qualifications. For women, I find a reversal of the pattern of career progressions over the sample period. Women have come to advance faster than men, as evidenced by a reversal of the gender balance of on top bureaucratic positions from male- to female-dominated in recent years. The paper will further probe reasons behind this reversal by describing women’s labor market background in greater detail.

Author

• Nora Anter