Occupational trajectories of working conditions in Sweden: Development trends in the workforce, 1997-2015

Abstract

Objective: This study aimed to explore the development of working conditions within and between occupations in the Swedish labor market from 1997 to 2015 and whether any polarization in working conditions concurrently occurred between occupations.

Methods: Cross-sectional data from ten waves of the Swedish Work Environment Surveys (1997-2015) were used and an aggregated occupational-level dataset was created using the Swedish Standard Classification of Occupations. To capture the patterns of change in working conditions over time (ie, growth), growth curve modeling was used to identify the starting points for 89 occupations (intercepts) as well as both the shape (functional form) and rate of growth (slope) over time.

Results: The Swedish labor market was stable overall, with some small, mainly positive, changes in job demands and resources. Different occupations developed in divergent directions, but there was no evidence of polarization.

Conclusions: The findings indicate that macro-level stability can hide highly heterogeneous patterns of change among different occupational groups. This type of analysis, taking context into account, could be valuable for decision makers intending to improve the work environment.

Figure 1

Nonlinear and linear macro trends; trajectories based on estimated parameters from Model IV and III respectively. Note that all dimensions have been coded so that a high value implies a favourable development in the working condition at hand.

Figure 2

Example of meso-level observations of the variable “decision authority – when”. (Question: Are you able to determine when various work duties are to be carried out (for example, by choosing to work a bit faster on some days and taking it easier on other days)? Response alternatives: 1 = no, not at all, 2 = mostly not, 3 = mostly, 4 = always. Left panel: Aggregated observed data for all occupations, 1997–2015. Right panel: Meso trends shown as estimated trajectories (low and high) for selected starting points (intercepts) at –1 SD, 0 SD, and +1 SD. Estimated linear macro trend shown as a bold line..