Working Time Mismatch and Subjective Well‐being
| DOI | http://doi.org/10.1111/j.1467-8543.2008.00705.x |
| Date | 01 March 2009 |
| Author | Robert Drago,Mark Wooden,Diana Warren |
| Published date | 01 March 2009 |
Working Time Mismatch and Subjective
Well-being
Mark Wooden, Diana Warren and Robert Drago
Abstract
This study uses nationally representative panel survey data for Australia to
identify the role played by mismatches between hours actually worked and
working time preferences in contributing to reported levels of job and life
satisfaction. Three main conclusions emerge. First, it is not the number of hours
worked that matters for subjective well-being, but working time mismatch.
Second, overemployment is a more serious problem than is underemployment.
Third, while the magnitude of the impact of overemployment may seem small in
absolute terms, relative to other variables, such as disability, the effect is quite
large.
1. Introduction
Recent decades have witnessed noticeable changes in the distribution of
working hours in many industrialized nations. In particular, while the
average length of the workweek has either continued to decline or remained
largely unchanged, variation around the average has mostly increased. In the
USA, for example, Jacobs and Gerson (2004) reported on data from the
Current Population Survey that reveal a marked decline between 1970 and
2000 in the proportion of employees reporting working a traditional 40-hour
workweek, and noticeable increases in the proportion of employees reporting
either relatively short workweeks (less than 30 hours) or relatively long
workweeks (50 hours or more). Similar trends have been documented in
Australia (Wooden 2002; Wooden and Drago 2007), Canada (Sheridan et al.
2001), Japan (Japanese Ministry of Health, Labour and Welfare 2004), New
Zealand (Callister 2005) and the UK (Green 2001). Even in continental
Europe, the variability of weekly working hours during the 1990s led the
OECD (2004: 40) to conclude that the evidence ‘is suggestive of an overall
trend towards greater diversification of weekly work schedules’.
Mark Wooden and Diana Warren are at the University of Melbourne. Robert Drago is at
Pennsylvania State University.
British Journal of Industrial Relations doi: 10.1111/j.1467-8543.2008.00705.x
47:1 March 2009 0007–1080 pp. 147–179
© Blackwell Publishing Ltd/London School of Economics 2009. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.
Such trends have been accompanied by rising levels of concern among
researchers about the impacts that working time arrangements might be
having on workers and their families. In particular, there is now a sizeable
literature documenting the effects of regular exposure to extended work
schedules on worker health and injury. Very differently, other research has
pointed to the potential for long working time regimes to adversely affect
relationships within the home, and ultimately contribute to marital break-
down and adverse outcomes for children. At the other end of the working
hours spectrum, part-time employment has also long been a subject of
research interest, with many arguing that part-time jobs often provide insuf-
ficient hours to satisfy the needs and desires of workers.
Central to the latter argument, at least, is the premise that it is not the
number of hours worked per se that matters, but whether those hours are in
line with workers’ preferences. In textbook models of labour supply, this
distinction is irrelevant; individuals are assumed to freely choose the combi-
nation of work hours and non-work hours that maximizes their personal
utility subject to time and budget constraints. Actual hours worked are thus
a direct reflection of preferences, and any mismatch should be only tempo-
rary. There is, however, a growing body of survey evidence that suggests that
work time mismatches are common and found in many countries (e.g. Bell
and Freeman 2001; Jacobs and Gerson 2004; Lee 2004; Reynolds 2004;
Reynolds and Aletraris 2006; Stier and Lewin-Epstein 2003).
Despite this evidence, relatively little research has taken working hours
preferences into account when examining relationships between hours
worked and outcome variables. This is the central feature of the study
reported on here. Specifically, panel data from the first five waves of the
Household, Income and Labour Dynamics in Australia (HILDA) Survey are
used to test for relationships between the number of usual weekly hours of
work and subjective measures of worker well-being. The HILDA Survey is
relatively unusual in that it collects information on preferred hours of work
conditional on consequences for earnings. We are therefore able to identify
workers who are underemployed (in the sense that preferred hours exceed
hours usually worked) or overemployed (usual hours exceed preferred
hours). Our hypothesis is that any adverse impacts of either short or
long hours of work should be most prominent when those working hours are
inconsistent with preferences.
A second distinguishing feature of the analysis is the use of panel data, and
hence, the ability to better control for unobserved worker heterogeneity.
Most previous research into the effects of working hours arrangements has
used cross-section data, and so, findings may be sensitive to the availability
and choice of control variables. In contrast, with panel data, we are able
to employ methods that effectively control for all worker characteristics
that are time invariant, or, at least, do not vary much over the period under
consideration.
Finally, the data are drawn from a large, nationally representative popu-
lation sample. This stands in contrast to much of the earlier research, and
148 British Journal of Industrial Relations
© Blackwell Publishing Ltd/London School of Economics 2009.
especially studies investigating the consequences of long hours, which mostly
employ data covering a small sub-group of the workforce, often taken from
a single employer or occupation.
2. Previous research
Part-time Work and Underemployment
There are a number of distinct literatures that are relevant to the study of
working time mismatch and its relationship to worker well-being. First, there
is a long-standing literature on the quality of part-time work. This body of
research has been concerned predominantly with identifying whether part-
time jobs should, on the basis of objective characteristics, be classified as
good jobs or bad jobs. Until recently, most studies appeared to support the
hypothesis that part-time jobs were generally of low quality, as reflected in
relatively poor remuneration levels, the absence of fringe benefits, low levels
of job security, and the lack of opportunities for career development (e.g.
Blank 1990; Kalleberg et al. 2000; McGovern et al. 2005; Tilly 1996).
Recent research, however, suggests that, with respect to wages at least,
much of the measured penalty for part-time employment disappears once
differences in job and worker characteristics are controlled for. Hirsch
(2005), for example, used microdata from the US Current Population Survey
and found that after controlling for measurable characteristics, and espe-
cially occupational skill requirements, much of the part-time wage differen-
tial disappeared. Indeed, for women, who account for the majority of part-
time workers, the gap is almost entirely eliminated. Manning and Petrongolo
(2004) reported very similar findings in an analysis of British Labour Force
Survey data. Furthermore, they found that the wage penalty becomes a
premium once all time-invariant worker and job characteristics are held
constant. Finally, Booth and Wood (2008) have drawn the same strong
conclusion using the same Australian panel data as used here (i.e. the
HILDA Survey).
Very differently, many studies that employ subjective measures of job
satisfaction have been unable to detect sizeable negative associations with
part-time work (e.g. Bardasi and Francesconi 2004; Blanchflower and
Oswald 1998; Booth and van Ours 2007; D’Addio et al. 2007; Manning and
Petrongolo 2004; Wooden and Warren 2004); indeed positive associations
are often found, especially among women. Such findings should not be
surprising given the widespread survey evidence that most part-time employ-
ees appear to prefer part-time hours.1Indeed, it is often argued that part-time
work is especially attractive to women, and tends to be consistent with their
own attitudes and preferences (e.g. Fortin 2005; Hakim 2000, 2002). Thus, a
positive coefficient on the part-time variable reflects the presence of many
constrained workers in full-time jobs who are unable to satisfy their prefer-
ences for fewer work hours.
Working Time Mismatch and Well-being 149
© Blackwell Publishing Ltd/London School of Economics 2009.
To continue reading
Request your trial