Part-Time Jobs: What Women Want?

Part-time jobs are popular among partnered women in many countries. In the Netherlands the majority of partnered working women have a part-time job. Our paper investigates, from a supply-side perspective, if the current situation of abundant part-time work in the Netherlands is likely to be a transitional phase that will culminate in many women working full-time. We analyze the relationship between part-time work and life satisfaction, and between job satisfaction and preferred working hours using panel data on life and job satisfaction for a sample of partnered women and men. We also utilize time-use data to consider the distribution within the household of market work and housework, and discuss the work specialization hypothesis in this context. Our main results indicate that partnered women in part-time work have high levels of job satisfaction, a low desire to change their working hours, and live in partnerships in which household production is highly gendered. Taken together, our results suggest that part-time jobs are what most Dutch women want.


Introduction
Across OECD countries there are big differences in the share of part-time work in employment among prime age female workers. In 2007, the female part-time share of women workers aged between 25 and 54 years ranged from a high of 60% in Switzerland and 54% in the Netherlands, to a low of 9% in Greece. An interesting question is whether or not the current situation of plentiful part-time work in some countries is likely to be an intermediate stage en route to a greater proportion of women in full-time jobs.
There are two opposing views on the efficiency implications of so many women working part-time. The negative view is that part-time jobs imply wastage of resources and under-utilization of investments in human capital, since many part-time working women are highly educated. 1 The positive view is that, without the existence of part-time jobs, female labor force participation would be substantially lower since women, confronted with the choice between a full-time job and zero working hours, would opt for the latter.
Against this background, the purpose of our paper is to investigate, from a supply-side perspective, if the current situation of abundant part-time work in the Netherlands is likely to be a transitional phase culminating in many women working in full-time jobs. Our econometric analysis, using panel data on life and job satisfaction of a sample of partnered women and men, assumes that dissatisfaction with a particular work status is likely to lead to changes in working hours in the future. In addition, we utilize time-use data to consider the distribution of market work and housework within the household. We also discuss the work specialization hypothesis in this context. If the Netherlands is characterized by little gender-stereotyping about working roles, we would expect to see that, on average in our sample of partnered households, the male share of domestic work is increasing in the female partner's share of market work. If this is not the case, it suggests that there is a gendered division of household and market labor within the family unit.
Our approach differs from that in earlier studies that investigate whether or not part-time work represents a stepping stone between nonwork and full- 1 In the UK, for example, transitions into part-time work are associated with occupational downgrading (Connolly and Gregory, 2008), and part-time work is associated with a pay penalty (Manning and Petrongolo, 2008). time employment. For example, Blank (1989) used US data from the Panel Study of Income Dynamics to explore transitions between the states of fulltime, part-time or nonwork over the period 1976-1984 for a sample of women aged 18 to 60 in 1976 who were either household heads or spouses. Blank found that three out of four women over the 9 years remained predominantly in that state and that very few women use part-time work as a stepping stone from nonwork to full-time work. In Sweden, Sundström (1991) shows that part-time work has not marginalized women but instead has increased the continuity of their labor force attachment, strengthened their position in the labor market and reduced their economic dependency. Continuous part-time employment has replaced work interruptions during child rearing years. Moreover the growth in part-time work has not been followed by increasing difficulties for women working part-time to shift to full-time work (Sundström, 1991). Thus the initial increase in part-time work in Sweden might be viewed as a transitional phase leading to many Swedish women working full-time. 2 In the Netherlands, the number of part-time jobs has expanded rapidly over the past decade, due to a gradual change in policy causing barriers for part-time employment to be removed. Laws were implemented that made part-time work more attractive. In 1993 the statutory exemption of jobs of less than one-third of the normal working week from application of the legal minimum wage and related social security entitlements were abolished.
Currently, most taxes are neutral and social security benefits are usually pro rata. In 1995 unions and employers signed the first proper collective agreement for temporary workers. In 2000 a right to part-time work law was introduced. Because government introduced legislation ensuring that the rights of part-time workers are properly protected, part-time work is not limited to marginal jobs but is a feature of mainstream employment (Portegijs and Keuzenkamp, 2008). According to Portegijs et al. (2008), the part-time job in the Netherlands was born in the 1950s when, in response to shortages 2 Sweden's childcare system is also likely to have played an important role in this process. In 1999, Sweden's public expenditure on formal daycare and pre-primary education amounted to 1.9% of GDP, as compared with 0.6% in the Netherlands. The OECD average was 0.7% (see Jaumotte, 2004). Booth and Coles (2009), using a panel of OECD data, show that public expenditure on childcare is positively correlated with female participation and with years of education. of young female staff, firms began to offer part-time jobs to married women. 3 Currently, about 40% of women with part-time jobs are mothers of young children who work part-time because they either prefer this, or have no choice but to provide childcare themselves. 4 However, almost half of the part-time working women are over 40 and no longer have young children. Many women in "small" part-time jobs prefer to work longer while many women in "large" part-time jobs prefer to work shorter hours. A part-time job between 20 and 27 hours a week would be women's preferred choice (Portegijs et al., 2008).
Apart from supply-side factors, changes in labor demand may have been important too. Euwals and Hogerbrugge (2004) distinguish between dynamic flexibility -adjustment to the business cycle -and organizational flexibility -adjustment to non-standard working hours. They conclude that dynamic flexibility cannot explain the strong growth of part-time employment, but the need for organizational flexibility, related to the shift from manufacturing to services, might have contributed. Bosch et al (2008) analyze the growth of part-time work distinguishing between age, calender time and cohort effects.
They find that the incidence of part-time work has increased over successive generations at the expense of full-time and small part-time jobs. As a result, the average working hours of working women remained stable over successive cohorts. Finally, Bosch and Van der Klaauw (2009), analyzing the effects of a 2001 tax reform which made work much more financially attractive for women with a high-income partner, find that women even slightly reduced their working hours in response to receiving a higher after-tax hourly wage.
Previous studies are important in charting patterns of work mobility, which can be used as a basis for predicting future behavior using comparative static techniques. However we choose in the present paper to adopt the alter- Section 6 concludes.
As will be seen, our main results indicate that partnered women in parttime work in the Netherlands have high levels of job satisfaction, a low desire to change their working hours, and they live in partnerships in which household production is with highly gendered. Taken together, these results suggest that part-time work in the Netherlands is here to stay, at least in the near future. British women are happy about their part-time job even though this does not increase their overall life satisfaction. It is interesting that we also found that work increased partnered male life satisfaction. In this sense, the finding for female life satisfaction parallels that of men.

Part-time work across the OECD
Across the OECD there are big differences in the share of part-time work in employment among prime age female workers (25-54 years). Figure 1a gives an overview of cross-country differences in part-time and total employment rates for prime age women in 2007. 5 The part-time employment-population rate ranges from a low 10% in Finland to a high 47% in Switzerland. The total employment-population rate ranges from 60% in Italy to 83% in Switzerland.
The main suspects contributing to the observed differences in part-time work across OECD-countries are institutional arrangements and union resistance. 6 Part of the opposition to part-time jobs is based on the idea that   Table 2 In our empirical analysis we use an ordered logit model in which we introduce individual fixed effects α i and individual specific thresholds µ ij : Since partnered female life satisfaction is largely unaffected by their own hours of work, it seems that there is unlikely to be a strong desire to change working status from part-time to full-time in order to improve the quality of their lives. This is again suggestive of part-time work not being a transitory phase to full-time work.

Job satisfaction and preferred working hours
To study job satisfaction and preferred working hours, we use data from the OSA labor supply panel, a biennial panel survey of a representative sample of Dutch households. 13 The panel covers a broad range of work and life course 11 As in our previous analyses for Australia and the UK, partnered health is only significant in a cross-sectional setting. This may have to do with assortative mating or common behavior (health food, exercise et cetera). 12 We also investigated whether presence and age of children affects life satisfaction, The data contain information about job satisfaction and preferred working hours. The question about job satisfaction is specified as follows "How satisfied are you all in all with your work?" with the possible answers: 4 very satisfied, 3 satisfied, 2 dissatisfied and 1 very dissatisfied. As shown in Table 1 few men and women are in the lowest categories while more than half of the workers is in category 3.   Table 6 shows the parameter estimates. Clearly preferences to work more decrease when working hours increase and similarly preferences to work less increase with hours worked.
14 Although here too we investigated cross-partner effects, we did not find any evidence of these effects being present. The OSA data contain information about health but only since the year 2000. Therefore we did not include health status as one of the explanatory variables. Estimated over a shorter time period, good health has a positive effect on job satisfaction for both men and women. From the sample of women, we removed the 5 women who worked more than 40 hours per week. Clearly, most partnered individuals working long hours would prefer to work less, while most partnered individuals working short hours would prefer to work more. It is interesting to see the "equilibrium" hours of work, i.e. the number of hours at which there are as many individuals wanting to work fewer hours as there are individuals wanting to work longer hours. 15 To determine this "crossing point", we estimated a linear probability model of the probability of wanting to work more and the probability of wanting to work less, with the number of weekly working hours and calendar time as the independent variables. This is given as: where h denotes the actual weekly working hours and Pr(h + i ) (Pr(h − i )) denotes the probability of wanting to work more (fewer) hours. From equations (2) the "equilibrium" of preferred working hours h * t can be calculated as Because we are interested in the evolution of preferred working hours over time we estimate equation (2)  (2) using a linear probability model. Table 7 shows the parameter estimates.
As before, we find that with an increase of actual hours of work both men and women are less likely to prefer working more and more likely to prefer working less. Over time for both men and women preferences for working more and for working less go down. For men the drop in the preference of working more is larger than the drop in preference for working less. This indicates that over time the "equilibrium" hours of work goes down for men.
For women the calendar time parameter estimates for working more and for working less are about the same size. This indicates that the "equilibrium" hours of work for women hasn't changed over time. Based on the parameter 15 Note that this is an "equilibrium" at the extensive margin of expanding or reducing working hours as the number of preferred hours of work are not taken into account.
estimates presented in Table 7 we calculate that in 2005 the "equilibrium" hours of work for women would have been 21.7 hours per week, while for men it would have been 32.5 hours, coming down from 36.9 hours in 1985.
All in all, we conclude that our main results indicate that partnered women in part-time work in the Netherlands have high levels of job satisfaction, and a low desire to change their working hours. Again, this is evidence of part-time work not being a transitory phase on the route to full-time work.

Time use -a household perspective
Theories of household behavior, such as that put forward by Becker (1965), predict that partnered households will be characterized by specialization of labor, whereby in the extreme case one partner engages fully in home work and the other in market sector work. 16 Part-time jobs provide a means of combining domestic and market production, whilst maintaining workforce skills or experience capital for the future. Part-time work thereby facilitates incomplete specialization by either gender. The specialization hypothesis predicts gender differences in working hours because partners within a household specialize (completely or incompletely) in either market work or house work.
However, the prediction is symmetric: if one partner specializes in market work, the other will specialize in home production and in principle there is no a priori reason why the partner specializing in market work should be female or male.
In contrast, the gender identity hypothesis of Akerlof and Kranton (2000) is based on the idea that gender matters. Here the distribution of household work and market work is determined by gender-specific 'utility'. According to this approach, since individuals operate within society's constraints, their happiness and the gender division of labor could be powerfully affected by 16 Incomplete specialization, in which both partners perform part of the home work and and part of the market work, may arise because of non-linear production functions or because cost functions associated with skills investment are characterized by economies of scope. Non-linear production functions might arise if there is activity-specific fatigue or boredom, implying diminishing marginal productivity in each activity. Cost functions characterized by economies of scope occur if investment in market skills reduces the cost of investing in home skills (see Rosen, 1983). Under incomplete specialization there will be a monotonically declining relationship between the share of house work done by one partner and that same partner's share of market work.
social custom and conditioning. It is possible that -controlling for income -part-time jobs could make partnered women happier than either full-time work or no work, because such jobs allow them to gain esteem through working, while obtaining social and self approbation from being with and caring for their families and their homes.

Conclusions
In the Netherlands, the majority of working women have a part-time job.
There are two opposing views on the efficiency implications of so many women