Goos et al. (2014) state that the middle wage segment decreases in favour of the high and low wage segment. The corresponding hypothesis of job polarization refers to the assumption that technological change allows the automation of routine-based tasks, which may be strongly found in the middle wage segment and therein pursued tasks. The IHS apparently was not very happy with these results as well as with the way they were generated. Therefore, they provide their own ones, partly presented today by Gerlinde Titelbach. According to this presentation not only the middle wage segment shrank by about 6% in favour of an increase of the high wage rate segment, but the low wage segment also shrank by about 2%. Based on these results they refuse the hypothesis about job polarization for Austria.
I am very critical about this conclusion. First, even if their results are correct, the qualification of workers displaced in the middle wage segment probably does not suffice for a large part of jobs in the high wage segment. In the first place, therefore, a decrease of labour demand in the middle wage segment still intensifies the competition among vacancies in the low wage segment. Secondly, as they look at the workload in total hours, their results do not directly refer to the number of employees affected within the individual wage segments. Assuming that part-time agreements are more common in low wage rate segments, the latter may still have grown in terms of people. Thirdly, the investigation rather focusses on occupations as a whole instead of tasks, like the elaboration on routine-based technological change would let expect. Fourthly, their data reaches back to 1994, while the increase in the technological potential for the automation of tasks rather increases over time and may reach a new peak in the course of digitalization.
Finally, it is very likely that Austria does not face job polarization in the same intensity as other countries (cf. Peneder et al. 2016, Eichhorst/Buhlmann 2015) – due to, for example, the dual education system, the comparably successful preservation of industry as well as industrial safety. However, the results presented today do not suffice to reliably refuse the hypothesis as a whole. I am looking forward to the final publication.
- Eichhorst, Werner, Buhlmann, Florian (2015): Die Zukunft der Arbeit und der Wandel der Arbeitswelt. IZA Standpunkte Nr. 77, Bonn.
- Goos, Maarten; Manning, Alan; Salomons, Anna (2014): Explaining Job Polarization: Routine-Biased Technological Change and Offshoring. The American Economic Review 104: 8.
- Peneder, Michael; Bock-Schappelwein, Julia; Firgo, Matthias; Fritz,
Oliver; Streicher, Gerhard (2016): Österreich im Wandel der Digitalisierung. WIFO, Wien.
- Titelbach, Gerlinde (2016): Job Polarisierung in Österreich?, IHS, presentation at Workshop Arbeitsmarktökonomie, Wien.