Is There a Causal Effect of Working Part‐Time on Current and Future Wages?

DOIhttp://doi.org/10.1111/sjoe.12157
AuthorMarie Paul
Date01 July 2016
Published date01 July 2016
Scand. J. of Economics 118(3), 494–523, 2016
DOI: 10.1111/sjoe.12157
Is There a Causal Effect of Working
Part-Time on Current and Future Wages?
Marie Paul
University of Duisburg-Essen, Duisburg DE-47057, Germany
marie.paul@uni-due.de
Abstract
In this paper, I study the causal effects of part-time work on current and future wages.
To estimate these effects, I use a random effects model with a wage equation capturing
the employment history and a dynamic multinomial probit component for the choice of
employment status. Exclusion restrictions from the institutional context are exploited to
support identification. The results suggest that working part-time with few hours has a large
causal effect on current wages, but more extensive part-time work does not reduce current
wages. However, both types of part-time work lead to negative long-term wage effects.
Keywords: Female employment patterns; Markov chain Monte Carlo methods; part-time
employment; random effects models
JEL classification:C11; C33; J16; J24; J31
I. Introduction
The employment rate of women has increased over the last few decades
in many countries, and this increase is often achieved through part-time
work. In the European Union, 71 percent of working-age women engage
in market work, but 27 percent of them work part-time.1On the one hand,
part-time employment might be a good way to combine market work and
family responsibilities, particularly during the child-raising years and also
when caring for elderly family members. On the other hand, part-time work
is increasingly perceived as a “trap”. The fear is that part-time work is less
paid than full-time work, that it leads to lower wages in the long run, and
that it contributes to poverty in old age and in cases of divorce.2Yet, it
is an open question whether employees really do receive lower wages due
I would like to thank Bernd Fitzenberger, Katrin Sommerfeld, two anonymous referees, and
participants of several seminars and conferences for helpful comments.
1OECD Stat Extracts, age group 25–54, year 2012.
2Evidence for this perception comes from newspaper articles such as: “Die Teilzeitfalle”
(The Part-Time Trap), Der Tagesspiegel, October 14, 2012; “Dichtung und Wahrheit ¨
uber
die Teilzeitfalle” (Fact and Fiction about the Part-Time Trap), Die Welt, May 11, 2012;
“Altersarmut ist weiblich: Frauensache Teilzeitarbeit” (Poverty among the Elderly is Female:
Part-Time Work – A Women’s Issue), S¨
uddeutsche Zeitung, April 10, 2012.
CThe editors of The Scandinavian Journal of Economics 2015.
M. Paul 495
to working part-time and whether a long-term wage effect of sequences in
part-time work exists.
This paper adds to the literature on the wage effects of part-time work
(see, for example, Ermisch and Wright, 1993; Aaronson and French,
2004; Hirsch, 2005; Booth and Woods, 2008; Manning and Petrongolo,
2008; Connolly and Gregory, 2009; Fern ´
andez-Kranz and Rodr´
ıguez-Planas,
2011). It provides evidence on the causal wage effect of current part-time
work for women, distinguishing between part-time work with few hours (5–
15 hours, PTS) and part-time work with many hours (16–34 hours, PTL).
Moreover, I investigate how PTS and PTL in the past influence current
wages, in order to understand how working part-time today influences cur-
rent as well as future wages. Furthermore, I compare typical female employ-
ment patterns, involving part-time employment, full-time employment (FT),
and periods of non-employment (NE), with uninterrupted full-time careers
and with long interruptions with regards to long-term wage effects. I use
panel data from 1984–2011 from Germany, a country in which part-time
work plays an important role, with a part-time share of 38 percent among
the working females and a high female employment rate (78 percent).3
To estimate the causal wage effect of working part-time (conditional
on the employment history), it is necessary to take into account all of
the following factors. First, unobserved heterogeneity and selection into
employment and various types of working-time arrangements have to be
considered. This is necessary because part-time workers are likely to have
different time-invariant unobservable characteristics, which are likely to
also influence wages (e.g., ambition and energy). Second, contemporaneous
endogeneity of part-time work and of employment should be taken into
consideration, because there might be exogenous shocks that influence both
the employment status as well as the wage. Third, differences in histories in
between both types of part-time and full-time workers (such as interruptions
and prior part-time work) must be taken into account because part-time
workers are likely to have a different employment history than full-time
workers, and this difference will influence wages. Fourth, endogeneity of
the history with regard to selection based on unobserved heterogeneity
and with regard to contemporaneous endogeneity must also be taken into
account, because the histories consist of past employment decisions that
have not been exogenous.4To the best of my knowledge, the present study
is the first to estimate the causal wage effect of part-time work taking
into account all these issues. The only paper in the part-time literature
3OECD Stat Extracts, own calculations, age group 25–54, year 2012.
4If one is solely interested in the wage effect of current part-time work and has a suitable
identification strategy, then the two latter points do not necessarily have to be taken into
account.
CThe editors of The Scandinavian Journal of Economics 2015.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT