[by Cyrille Godonou]

Despite abundant documentation showing that the majority of female part-time work is chosen (two-thirds to three-quarters), it is surprising that not all authors relay this conclusion, particularly in the academic world. Involuntary part-time work corresponds to part-time work due to the failure to find full-time work, i.e. due to a problem related to the demand for work (companies). Conversely, chosen part-time work depends on the supply of work (workers).

Since involuntary part-time work is a minority, some authors have suggested that there is a grey area in voluntary part-time work, insofar as it is likely to overlap with part-time work due to family constraints, i.e. due to a lack of childcare resources. Other authors suspect that part-time work chosen for family reasons would be involuntary in the absence of family factors, due to the lack of availability of full-time employment. The case of the French state civil service, where civil servants are necessarily in full-time jobs unless they request it, may prove enlightening. Similarly, data from job seekers provides insight into the desired work quota.

In addition, we will also be looking at parental leave. While this is the case of France, which we intend to focus on, we will also provide detailed information on other Western countries.

Is part-time work for family reasons due to family constraints such as the absence or insufficiency of childcare or does it reflect a family preference and therefore a choice? What about parental leave? Measuring satisfaction makes it possible to understand whether it is chosen or imposed.

Some authors conclude that most female part-time work is involuntary, despite their own data suggesting the opposite.

Margaret Maruani and Monique Méron address the issue of involuntary part-time work, with a critical look at the choices made by INSEE and international organisations in the definition of underemployment in the ILO sense (Meron & Maruani, 2012).

In their book, in Table 3.10, we can see that 30,4% of women not working full-time were working part-time in 2010, due to not having found a full-time job, which means that among women working part-time, nearly 70% of women are working part-time for other reasons. The presence of choices in decisions appears explicitly in the questionnaire (18,2% of women give the reason for working part-time as "To have free time or to do housework" compared to 13% of men). But in their commentary, Méron and Maruani write that "most women did not choose to work part-time - or more precisely, it is under the pressure of unemployment that they "choose" part-time: underemployment in order not to remain unemployed".

Figure 1: Reasons for part-time work reported by employees in 2010

They even write that it is a lie to say that part-time work suits women (part-time), that it would be a "fable", a "myth", a "legend". These conclusions are in direct contradiction with the works mentioned below (Paull, Taylor, & Duncan, 2002; Booth & van Ours, 2010). Maruani and Méron write: "Just like unemployment, underemployment and part-time work are symptoms of the shortage of jobs. But, in the history of work in the 1960th century, part-time work poses a particular problem: a specifically female form of employment has been constructed that counteracts the trend towards the homogenization of male and female activity behavior observed since the 2012s. It was created from scratch for them - "tailor-made" in a way. But to say that it suits women is another story. This would be entering the register of fable, myth, legend. Or perhaps social lies." (Meron & Maruani, XNUMX).

However, they are not the only ones to consider that involuntary part-time work is the majority. Laurence Cocandeau-Bellanger ^[1] She also believes that part-time work “is most often involuntary and imposed by the employer or by the social context (job insecurity)” (Cocandeau-Bellanger, 2004). However, a 2001 INSEE publication states: “In 1999, again according to data from the Employment Survey, more than a third of women working part-time declared that they wished to work more hours (three-quarters of them wanting to hold a full-time job)” (Bourreau-Dubois, Guillot, & Jankeliowitch-Laval, 2001). These figures, which have remained valid since then ^[2] (Thélot, 2008), therefore indicate that the majority of women's part-time work is indeed a matter of choice (two-thirds).

Moreover, after concluding on the high satisfaction of the women interviewed, as to their life in general and more particularly as to their family life, aspiration of their youth realized (ideal of founding a family and raising children) notably thanks to the resources of the spouse (p 324), Laurence Cocandeau-Bellanger writes (p 326): "But then how to explain a feeling of satisfaction of professional life lower than that of family life? It would seem that despite this choice of the double commitment, their wishes for changes relate more to the professional aspect (having less work, more time available for the family...) than to the family aspect (having children younger or less closely together)" (Cocandeau-Bellanger, 2004).

In doing so, it is possible that Cocandeau-Bellanger arrives at a conclusion (on the apparently majority involuntary part-time work) opposite to what her own study suggests due to its limited scope: indeed, the external validity of this work is uncertain. The work carried out is based on a dual methodology: the data was collected by questionnaire and by interview, the publication being focused on the quantitative approach. Around a hundred women were interviewed who met the following profile: be aged 25 to 44, live (married or not) in a stable way with a man who himself works, have at least one child from the current union, the last of whom is under 5 years old. All women in the sample studied have a full-time or part-time job of at least three-quarters time, and have an intermediate profession or are employed. The questionnaire covers the following four themes: work, family, dual commitment and satisfactions. Despite the limitations with regard to external validity, it is remarkable that these works are consistent with those mentioned below.

Françoise Milewski also comes to this conclusion based on statistics on job seekers in 2004 (more precisely on whether they are looking for full-time or part-time jobs) on page 54:

"An indirect measure consists of looking among job seekers registered with the ANPE, those who are looking for part-time work. This is a measure of "chosen" part-time work, but only for the population of registered unemployed people. In June 2004, while 84% of part-time job applications came from women, which roughly corresponds to their share of part-time employment, only 24,3% of women looking for work wanted part-time work, both for all women and for those aged 25 to 49 (Table II.11). The share of "chosen" part-time work would therefore be, according to this measure, very low..."

(Milewski, Méda, Ponthieux, & Letablier, 2005)

However, we arrive at the opposite conclusion: the share of part-time work chosen according to this indirect measure is very high since the rate is 24,3% for a total part-time work of 30% (chosen or imposed). This would be 81% of part-time work chosen according to this indirect measure (24,3/30=81), compared to 71% with the direct measure (by the employment survey). Even by approaching job seekers, we find two thirds of chosen part-time work (71% precisely) and even more (81%), i.e. a quarter of female job seekers who are looking for a part-time job for 30% of employed women who work part-time.

24,3%: share of women requesting part-time work among job seekers in 2004 (restricted field, therefore indirect measurement) according to the report by Françoise Milewski et alii.

30%: proportion of women working part-time among all women working in 2004 (full scope, therefore direct measurement of 30,1%, see figure 3, the part of the table devoted to the proportion of active workers working part-time IP no. 1070) according to INSEE (Attal-Toubert & Lavergne, 2006).

29,1%: share of women in involuntary part-time work among all women working part-time in 2004 (full scope, therefore direct measurement, see figure 3, the part of the table devoted to underemployment among part-time jobs (%) IP no. 1070) according to INSEE (Attal-Toubert & Lavergne, 2006). So we have 70,9% (or 71% by rounding) of women who are not underemployed among women working part-time (more than two thirds of part-time work is chosen).

With more recent data in 2011, DARES provides a figure of 69% of chosen part-time work for women (direct measurement), therefore close to the 66% of INSEE in 1999, thus confirming that the indirect measurement corresponds to a greater share of chosen part-time work than the direct measurement: 81%>69%. There is therefore no overestimation of chosen part-time work when it is measured directly.

Figure 2: Population in employment in France in 2004

For more up-to-date data, it is useful to consult the tables of the French economy, INSEE references, published by INSEE, the chapter “Duration and organization of working time 4.4”, for example on page 49 of the 2019 edition:

Figure 3: Part-time work for employees (excluding apprentices) by gender in 2017

The rate of female underemployment (in other words, involuntary part-time work) was 8,5% in 2004 (Table 3 of the INSEE publication) for a total female part-time of 29,9%, hence the deduction of the share of chosen part-time work (Durier, Gonzalez, Macario-Rat, & Thélot, 2007): (29,9%-8,5%)/29,9%=71,6%. We have 81%>71,6%! The indirect measurement is much higher than the direct measurement, which indicates that the chosen part-time work measured directly is not overestimated.

In France (Baradji, Davie, & Duval, 2016), with comparable characteristics (age, marital status, presence of children aged 6 or over, socio-professional category, type of employment contract, sector of activity), the probability for a man aged 16 to 65 working in the private sector to be involuntarily part-time is 2% for the years 2010 to 2012 (cumulative of the employment survey). For all other variables in the model equal (unordered polytomous model), a woman has a probability of being involuntarily part-time higher by 5,8 points or 7,8%. Women in the private sector are thus, with given characteristics, four times more likely to be involuntarily part-time than men, taking into account the characteristics of the model.

For men in the public sector, the probability of being on involuntary part-time is also 2%, still with given characteristics, while for women in the public sector it is 3 points more, or 5%. For men in the public or private sector, the probability of being on involuntary part-time is also 2%, still with given characteristics, while for women in the public or private sector it is 6,6 points more, or 8,6%. Based on simple descriptive statistics, it appears that in the public sector, 6% of women are on involuntary part-time (2% of men), 20% of them are on voluntary part-time (4% of men). In the private sector, 10% of women are on involuntary part-time (compared to 2% of men), while 21% of women employed in the private sector work on voluntary part-time (3% of men). While it is clear that women are more exposed to involuntary part-time than men, the majority of part-time work is voluntary. It is interesting to note that for all sexes, the prevalence of involuntary part-time work is 6% in the private sector and 4% in the public sector (0% in the state civil service and the hospital civil service but 5% in the territorial civil service).

To only mention involuntary part-time work is to forget that some studies suggest that some people would like to work less, as illustrated by the 2010 data: "A third of people with a job would like to spend less time working and a tenth would like to have more time to work." (Ricroch & Roumier, 2011). INSEE indicates that in 2007, 2,6% of employed women wanted to work less compared to 1,5% of employed men (Thélot, 2008): in 2007, out of 25,6 million people with a job as defined by the International Labor Office, "5,4 million people wanted to change their working hours, or 21,1% of people with a job: 4,9 million wanted to work more hours, with part-time workers being over-represented, and more than 500 people wanted to work fewer hours."

The Natural Experiment of State Civil Service

One of the arguments for the alleged mass part-time work by women is to say that women who say they work part-time for family reasons would work part-time if they were available for full-time work. One way to verify this hypothesis is to observe the holders of the State civil service who offer full-time jobs. It is therefore at the request of the staff that part-time work can be granted. Within the State civil service, we distinguish national education, whose working time (in person) is less, taking into account teachers with hourly obligations lower than the legal working time (around half the duration).

It must be noted that 17,3% of women in the State civil service were working part-time in 1998 (compared to 2,5% for men), without it being possible to invoke involuntary part-time work (Brenot Ouldali, 2007).

Figure 4: Part-time and full-time in the state civil service in 1998

If we look at female civil servants (outside national education), we see that 33% of them work part-time, compared to 2% of men.

Figure 5: Part-time and full-time in the State civil service excluding national education in 1998

We find orders of magnitude from the private sector, which tends to invalidate the hypothesis of involuntary part-time work “masked” by part-time work for family reasons.

Even women civil servants in national education are part-time in 11,3% of cases in 1998. It should be noted that among them, there is a high proportion of female teachers who benefit in a way from the equivalent of "teleworking" or at least from greater flexibility, since copies can be corrected at home and lessons can be prepared at home.

Figure 6: Part-time and full-time in the national education state civil service in 1998

As for male civil servants in the state, outside or including or exclusively in national education, 2% to 3% of them were part-time in 1998, a figure comparable to that found in part-time employment in the private sector.

On involuntary part-time work: in OECD countries, part-time work is mostly chosen, with women constituting the majority of part-time workers.

The OECD conducted a study in over thirty countries and concluded that involuntary part-time work never affects more than 25% of part-time employees (OECD, 2010). On average, 17% of part-time employment in 2007 was involuntary or unwanted (page 240) ^{[3]. The methodology, based on surveys (national employment surveys),} that "the range of possible reasons offered by the surveyors extends from school or family obligations to the impossibility of finding full-time employment, and only those citing the latter reason for their part-time work are considered involuntary part-time workers." The scope is broad, and the representativeness of the surveys allows for a robust conclusion with broad validity.

Voluntary part-time work is thus by far the most common (over 80% of cases). Seven out of ten part-time workers are women (three-quarters of part-time workers in the European Union are women) ^[4] . Women choosing part-time work are also satisfied with their pay and the resulting work-life balance ^[5] The divergence in expectations between men and women in the labor market, particularly regarding pay, remains incomprehensible if family dynamics are not taken into account, given the interactions between family life and professional life.

Figure 7: List of countries studied by the OECD

1 Germany
2 Australia
3 Austria
4 Belgium
5 Canada
6 Korea
7 Denmark
8 Spain
9 Estonia
10 United States
11 Federation de Russie
12 Finlande
13 France
14 Greece
15 Hungary
16 Irish
17 Iceland
18 Israel
19 Italy
20 Japan
21 Luxembourg
22 Mexico
23 Norway
24 Nouvelle Zelande
25 Pays Bas
26 Poland
27 Portugal
28 Slovak Republic
29 Republique tchèque
30 United Kingdom
31 Slovenia
32 Suede
33 Switzerland
34 Turkey

In 2017, 26,4% of part-time workers in the European Union were involuntarily part-time workers (EUROSTAT, 2018). Three-quarters of part-time workers chose to do so.

The paradox of part-time work

A comparison of European Union countries, using the European Working Conditions Survey of 2000, shows that the more part-time work there is, the more voluntary part-time work there is (Burchell, 2012).

“If we compare rates of involuntary part-time work across countries, the results are quite paradoxical. We might expect that where the number of part-time jobs is low (for example, in Greece), these positions would be filled by employees with a strong preference for part-time work, but that where the share of part-time employees is high, we might expect more women to be forced to accept part-time jobs due to a lack of choice. In fact, the opposite seems to be true: high rates of part-time work are associated with high levels of voluntary part-time work. For example, at both extremes, involuntary part-time work is almost non-existent in the Netherlands (1,9%) but very high in Greece (42%). Figure 3 shows that there is a very strong linear relationship between these two extremes. ^[6] »

(Burchell, 2012)

Satisfaction and choice to reduce working hours (intensive margin) or not to work at all (extensive margin): a vast majority of mothers not working full time are nevertheless satisfied

The supply of daycare has an impact on the activity of mothers (Fontaine, 2018) but the magnitude of this “childcare supply” effect is likely to be more or less significant.

In Germany, the effect of introducing one day of free childcare for children aged 2 to 6 (in the western German states between 2000 and 2015) on women's labor supply was studied (Busse & Gathmann, 2018): no effect was observed on either labor market participation (extensive margin) or labor volume (intensive margin) ^[7] Despite some success with the program, which resulted in an increase in the number of children aged 2 or 3 years, especially from low-income families, female employment is nevertheless declining ^[8] For children aged 4 or 5 years, the implementation of free daycare leads to a modest decline in women's participation in the labor market ^[9] It was therefore the income effect that prevailed: free childcare through subsidies can be analyzed as an increase in income, which did not encourage people to work more but rather to take more free (or rather, parental) time. Thus, in Germany, the use of group childcare is not closely linked to mothers' participation in the labor market ^[10] : while almost 100% of children aged 5 to 6 are in collective care, only 54% of their mothers work.

Methodologically, the variety of policies implemented in the different states (Länder) within Germany allows us to develop an identification strategy, with a control group and a treatment group (difference in differences and fixed effects on years and states). The dependent variable is in turn the participation of women in the labor market or the attendance of children in daycare centers. The source used is a panel of 12 households, restricted to 000 for the analysis due to the east-west differences that lead to excluding the east from the analysis. The regressions are applied to children aged two to five and to those aged five or six. Whether on the working time of mothers, the fact that they are full-time, or on their participation in the labor market (extensive margin), the reform has no effect: no coefficient is significant — see table 9 of the publication (Busse & Gathmann, 000). These results are consistent with some studies finding little impact for Norway – difference-in-differences method comparing Norwegian municipalities and the effect of the 4 reform on the employment rate of mothers from a panel of administrative data representative of the entire population between 2018 and 1975 – (Havnes & Mogstad, 1967). To give an idea, the gain is not even 2006 additional mothers on the labor market (extensive margin) for 2009 additional places in daycare; by increasing daycare coverage by 765 points in the treatment group, the employment rate of mothers increases by 17 points.

Based on data on Swiss cantons from 2002 to 2012, the comparison of the treatment group and the control group shows that overall the employment rate of mothers does not increase, but that on the other hand there is an increase in the share of mothers working long part-time (from 20 to 36 hours of work per week), the average effect being two points on this type of employment, i.e. a modest increase in the employment rate of mothers (Ravazzini, 2018). The same kind of conclusions are also obtained in the United States, with a regression on discontinuity using the birth of children close to the eligibility threshold in Georgia and Oklahoma (Fitzpatrick, 2010), the establishment of free reception for preschool-age children in fact resulting in a 14% increase in registrations but without substantial impact on the employment rate of mothers, the coefficients which are otherwise of modest magnitude not being significant (see Table 5).

But these results are contrary to those obtained by other studies carried out in Spain and Germany.

In Switzerland, research has revealed the unmet demand for childcare services: three-quarters of parents report not needing childcare or being satisfied with the services they currently have ^[11]

It appears that the parents of 22,6% of preschool-age children and 29,6% of school-age children do not need childcare, either formal or informal. For 53,8% of preschool-age children and 49,2% of school-age children, their needs are met, i.e., the children receive care as much as the parents wish and in the form they wish. However, the parents of 3,7% of preschool-age children and 3,2% of school-age children indicated that they wished they did not have to provide them with as much childcare. The reasons for this are unknown. It may be that these parents do not work part-time or that the childcare facility sets a minimum care time. Finally, the parents of 19,9% ​​of preschool-age children and 18,0% of school-age children say they would need more childcare. When assessing unmet demand, however, it must be taken into account that needs can change rapidly in the event of changes in structural or individual factors, including, for example, the available supply or childcare prices, but also changes in the network of acquaintances to whom parents can entrust their children. (CHSS, 2018); (Bieri, Ramsden, & Felfe, 2017). Conversely, the parents of 19,9% ​​of preschool-age children and 18,0% of school-age children say they would need to have their children looked after more.

Figure 8: Childcare needs in daycare according to their parents in Switzerland

In Norway, similar results are obtained. The situation has changed significantly between 2002 and 2010. For children under one year old, 84% of mothers believe that the best childcare system is for the child to be looked after only by the parents, compared to 82% in 2010. For one-year-old children, 40% of mothers in 2002 and 34% in 2010 believe that the child should be looked after only by the parents. For two-year-old children, these figures are 2% and 16% respectively.

Mothers who only wanted daycare, nurseries or nursery schools as their ideal childcare option were a minority in 2002 for children under one year old (1%), one year old (9%), two years old (24%), and three years old (41%) but in 2010 these figures were 1%, 18%, 55%, 72% respectively. A significant proportion of mothers wanted to combine parental care or other solutions than collective daycare with collective care: very few wanted this combination for children under one year old (2% in 2002 and 3% in 2010), but more for one-year-olds (22% in 2002 compared to 31% in 2010), for two-year-olds (38% compared to 29%), for three-year-olds (44% compared to 23%). In 2002, even for children aged 4 or 5, half of mothers wanted only collective care as the ideal childcare method, the other half therefore did not want to use this solution alone. However, in 2010, only 18% and 14% of mothers wanted to combine for children aged 4 and 5 respectively. As for the number of hours of care desired, for children under three, less than half of mothers wanted seven hours of care per day. Even for children aged four or five, 48% to 67% of mothers wanted seven hours of care. Seven hours of care per day over five days corresponds to 35 hours of care per week. For two-year-olds, 63% of mothers in 2010 wanted less than seven hours of care per day, with this figure being even higher for younger children and in 2002. Taking into account travel time, these data suggest that a large proportion of mothers whose youngest child is aged 0-5 prefer not to work full-time.

Figure 9: Mothers with children aged 1 to 5 years, according to the type of childcare they think is best suited to children of different age groups in Norway (2002 and 2010)

Total workforce in 2002: 2879
Total workforce in 2010: 2151

After…

In Hungary, a study found that 70% of mothers approved of women's absence from the labour market for three years (Bencszik & Juhàsz, n.d.).

In Ireland, certainly 60% of parents believe that they do not have access to (quality) childcare ^[12] but mothers' return to work is motivated in 65% of cases by financial reasons ^[13] and in 18% of cases by career (Mcginnity, Murray, & McNally, 2013).

In Luxembourg, in 2015, 73% of parents who took parental leave were satisfied with it (TNS Ilres, 2015). The motivations for parental leave are, by far, the parent-child relationship (92%) and the child's well-being (92%). By comparison, the reason "no crèche found, no crèche" is only given in 6% of cases, "work schedule incompatible with crèche schedule" is given in 10% of cases and finally "cheaper than crèche" is only given in 21% of cases.

Figure 10: Satisfaction and reasons for parental leave in Luxembourg in 2015

In France, INSEE reveals that the main reasons for mothers taking full-time parental leave are above all to devote themselves to their child, with these two reasons being cited by around 90% of mothers, or almost all. Consequently, at most 10% of mothers on full-time parental leave are not on it for these two reasons combined. Other reasons given in the survey include the cost of childcare (for 33% of mothers) and the need to recover from the birth (30%), followed by the incompatibility between work schedules and childcare service schedules (17%) and the lack of childcare solutions nearby (17%) (Govillot, 2013).

Figure 11: Parental leave for mothers in France in 2010: above all for the good of the child

The INSEE results, which are authoritative due to their representativeness and are valid for France, are consistent with those of the OECD, although they focus on parental leave rather than part-time work. The common thread is that both highlight the significant role of choice in the total or partial withdrawal from the labor market, although it should not be ignored that constraints persist. For the first year following childbirth, women stop working not so much due to a lack of childcare but to spend time with their child, demonstrating that it is more a matter of family choice than family constraints, as sociologist Catherine Hakim points out: "The myth of childcare problems as ^[14] (Hakim, 1995). The INSEE data comes from the supplementary module associated with the 2010 Employment Survey, commissioned by Eurostat, on work-life balance. It concerns individuals aged 15 to 64 living in ordinary households. A total of 24,302 people responded to the survey. The scope of the study is limited to the 3,028 parents who met certain criteria at the time of the survey. According to INSEE: "It is unlikely that obtaining a place in a daycare center allowed more than 5% of the mothers who benefited from it to avoid interrupting their careers." (Pora, 2020).

These results obtained in France regarding chosen parental leave (90%) are consistent with other international studies (particularly in Anglo-Saxon countries) reporting tensions and low marital satisfaction among women who do not want to be in the labor market but are captive to it ^[15] (Moen, 1992), express the lowest levels of satisfaction with their marital life ^[16] (Hakim, 2004). It should be noted that the freedom to choose working hours is a good predictor of marital satisfaction in the United States, with greater satisfaction for women working part-time (Orden & Bradburn, 1969). Furthermore, “In most European countries, the provision of public childcare is very widespread, on average less expensive than private childcare (lower-income families receive larger subsidies), and is also widely recognized for its quality. Nevertheless, despite these attractive features, the use of public childcare is quite limited. Several studies have attempted to unravel this mystery.” ^[17] (Del Boca, Locatelli, & Vuri, 2005).

However, not all countries are in the same situation regarding the impact of daycare provision on women's participation in the labour market. In Italy, the employment rate for women is significantly lower than the European Union average, 52,5% in Italy (almost 20 percentage points lower than that of men in Italy) compared to 66,5% in the EU (Eurostat, 2018). In studies conducted on Italy, the impact of the supply of daycare taking into account subsidies of up to 100% (i.e. free) was conducted, distinguishing between territories with a limited supply (initial female employment rate of 40,8%) and those with an abundant supply (female employment rate of 61,5%): reducing the cost of care (100% subsidy) the employment rate increases slightly in the first case (+5,4 points) but jumps in the second case, increasing by 26,5 points (Del Boca & Vuri, 2007). These data suggest that a more abundant supply of affordable childcare is likely to increase the female employment rate very significantly. It should be noted, however, that even in countries with abundant and subsidized offers, the female employment rate remains slightly lower than the male employment rate: this is the case in Sweden (employment rate of 83,8% for men in 2017 compared to 79,8% for women), Norway (80,2% compared to 76,2%), Finland (75,9% compared to 72,4%) and Denmark (80,2% compared to 73,7%). Although narrower, the gaps persist.

Figure 12: Happiness experienced in marriage by spouses as a function of working time and whether it was chosen or imposed in 1963 (Orden & Bradburn, 1969)

In the same vein, it is also women who would prefer to work fewer hours but cannot who are the most dissatisfied in the United Kingdom in 2000 (McGovern, Hill, Mills, & White, 2008). It should be noted that McGovern et al. rely on a sample stratified by address, which ensures quality representativeness: approximately 3 people were interviewed in 500 (response rate of 1992%) compared to 71 in 2 (response rate of 100%).

Figure 13: Average dissatisfaction based on hours worked (7-point scale), according to men's and women's preference

In the United Kingdom as well, among mothers working part-time, only a minority would like to work more hours, according to local administrative data cross-referenced with 1999 survey data (Paull, Taylor, & Duncan, 2002). Among mothers in couples working part-time whose youngest child is three years old or younger, between 8,8% and 12,1% would like to work more hours if suitable childcare services were available but cannot because they have to look after their child(ren). These figures are higher for single mothers (22,4% to 25,4%), who are also a demographic minority (in 2004, among UK families with at least one child under 16 or aged 16 to 18 in school, single-parent families represented a quarter of the total ^[18] ). According to Paull, Taylor and Duncan, this means that the vast majority of mothers working part-time are satisfied with the number of hours worked and that the lack of daycare is not a constraint for the vast majority of them ^[19] . Furthermore, 20,8% to 25,4% of mothers in couples working part-time whose youngest child is between 4 and 11 years old report wanting to work more but being unable to do so due to childcare responsibilities. These figures are also higher for single mothers (35,8% to 39,9%), a demographically smaller group. This means that the vast majority of part-time work is genuinely a choice. These results are consistent with those of the OECD, whose external validity appears to be stronger.

Figure 14: Perception of mothers in part-time employment of the impact of maternal constraints on hours worked

Women in the UK have a preference for part-time work, as confirmed by other studies ^[20] (Booth & van Ours, 2008), particularly work of less than 30 hours per week ^[21] , with overtime being off-putting to them.

Based on Canadian data from 1989, namely the Labour Market Activity Survey (LMAS), 27% of women are part-time and among them 85% are voluntary part-time (Barrett & Doiron, 2001).

In Germany, Socioeconomic Panel data from 1992 to 1998 also shows that "life satisfaction of mothers in the years following the birth of a child is especially high when they return to part-time employment after the birth" ^{[22] (Holst & Trzcinski, 2003). Overall, for a given number of children, satisfaction is highest with part} work, then with non-employment, and finally with full-time work.

Figure 15: Life satisfaction of women and mothers in Germany in the interval from one year before the birth to five years after

In Denmark, in 2012, 23,7% of part-time women were part-time because they could not find a full-time job, compared to 28,8% in Finland, 27,3% in Iceland, 18,9% in Norway and 24,9% in Sweden (Lanninger & Sundström, 2014), based on the Eurostat source. Thus, among part-time women in these Nordic countries, three-quarters chose it.

This massively chosen nature of part-time work is reflected in job satisfaction (source European Social Survey (ESS)). In the Nordic countries in 2010 (Lanninger & Sundström, 2014) the share of women very satisfied in their jobs is similar for women working full-time and part-time (respectively 50,5% and 51,6% in Denmark, 28,2% and 32,9% in Finland, 38,4% and 33% in Norway, 30,5% and 25,7% in Sweden). In Denmark, 96,8% of women in full-time employment are satisfied or very satisfied compared to 96% of those working part-time. These figures are also high in Finland (94,5% for those working full-time compared to 89,1% for part-time) and in Norway (94,7% compared to 86,9%). In Sweden, 90,2% of women say they are satisfied or very satisfied among those working full-time, compared to 83,5% of those working part-time.

Figure 16: Job satisfaction in full-time and part-time jobs of women aged 20-64 in Denmark, Finland, Norway and Sweden in 2010

Dutch women are the happiest in Europe and so are their children. However, Dutch women are the ones who work part-time the most (75%) and report being satisfied with the way they share tasks with their spouse (78%) (Stellinga, 2009).

Figure 17: Share of men and women with above-average life satisfaction in OECD countries in 1999 and 2002

Figure 18: Female part-time work in OECD countries: record in the Netherlands

In the Netherlands, women in couples who work part-time have high levels of job satisfaction and are unlikely to change their working hours, with results suggesting that part-time work is what most Dutch women are looking for (Booth & van Ours, 2010). The researchers use internet-based panel data (by CentER); all individuals aged 16 and over are surveyed about their work, income, health and demographic characteristics. They are asked questions about their life satisfaction and health status. In the raw results, women appear happier than men regardless of the work volume bracket. Men are less satisfied when they work less.

Using an ordered logit model, by introducing a fixed effect on individuals, Booth and van Our arrive at certain conclusions that neutralize structural effects. The level of satisfaction of men is higher if they work more than 20 hours per week (long part-time or full-time). They prefer their wife to work part-time, although the introduction of household income in the specification reduces the impact of hours worked, in particular the life satisfaction of men is no longer influenced by the hourly volume of their partner (variable becoming non-significant). As for women, only their health counts; their life satisfaction depends neither on the fact that they work nor on the number of hours worked.

Booth and van Our point out that in the Netherlands, in 2007 only 3% of part-time women preferred to work full-time. Part-time work is therefore chosen in almost all cases.

Booth and van Our then use a representative sample of households in the Netherlands (OSA labour supply panel). This sample is cylindered to match the CentER sample previously used. Only women aged 22 to 49 in 1992 are included, couples where the man is over 60 in 2006 are excluded. This time, the data contain information on job satisfaction. Women's job satisfaction increases slightly with the volume of work. Men's job satisfaction is lowest when they work between 21 and 32 hours per week. Men working more than 40 hours per week have the highest level of satisfaction. Using a fixed-effects logit model, the researchers again found that men's satisfaction was lowest when they worked between 21 and 32 hours per week, but their highest level of satisfaction was this time when they worked less than 21 hours per week, although it should be noted that none of the parameters relating to working time slots were significantly different from zero. As for women, their highest level of satisfaction was reached when they worked between 33 and 40 hours per week.

About 9% of women want to increase their working hours while 12% want to reduce them. For men, these figures are 12% and 21% respectively. Using a fixed-effects logit model, in which the dependent variable is the desire to work more or vice versa, the researchers finally analyse preferences in terms of working hours. The more you work, the less you want to increase your working hours. Similarly, the less you work, the less you want to reduce your workload. Booth and van Our therefore estimate the equilibrium point where preferences are such that you neither want to increase nor decrease your working hours: in other words, this is the optimal weekly working time. Between 1985 and 2006, this optimal duration decreased for men from 36,9 hours to 32,5 hours while it remained stable for women at 21,7 hours. The researchers conclude that women in a relationship in the Netherlands have a high level of job satisfaction and that they do not really want to change their working hours. According to the authors, there are two opposing views on the efficiency of female part-time work. A negative view is that it is a waste of human resources and an under-utilization of human capital since many women working part-time are highly educated. But the authors point out that a positive view is that the female employment rate would be significantly lower in the absence of part-time work, given that if the alternative to full-time work was to leave the labor market, many women would choose to withdraw. Finally, the authors use a time-use survey from 2000 to 2005 to assess whether there is a gendered division of domestic labor. The data show that men's domestic work time (dishes, cleaning, vacuuming, watering, etc.) is not very sensitive to paid work time, which is not the case for women. For women, an increase of one hour of market work leads to a decrease of only half an hour of domestic work, which is consistent with the idea of ​​gendered division of labor. According to the researchers, the status quo seems to satisfy both men and women. Women's satisfaction at work, and in particular part-time work, is a result consistent with OECD data.

In Australia, Booth and van Our in 2009 show, using panel data, that women's life satisfaction decreases when they work full time, particularly when they work more than 40 hours per week. On the other hand, their life satisfaction increases when their partner works full time, particularly between 35 and 50 hours per week. Both men and women are satisfied when the man works between 35 and 50 hours. Men's satisfaction does not vary whether their wife works part time or not. On the other hand, women are happier when working part time. This result for Australia is consistent with what is found in many other socio-economically comparable countries. Other studies have found similar results for Australia, namely the level of satisfaction of people working part-time (8), regarding the balance between professional and family life, is higher than that of people working full-time (7,1) (Luppi & Arpino, 2016). The highest level of satisfaction is thus achieved when there is a combination of paid (nurseries) and free, i.e. informal (grandparents, family, friends) childcare which allows flexibility.

Figure 19: Effect of the share of free childcare provision within all childcare methods on satisfaction with the balance between professional and family life

In the United Kingdom, Booth and van Our in 2008, using panel data, show that women are happy to work part-time, even if it does not increase their life satisfaction. The life satisfaction of women without children does not depend on the volume of work, while women with children are happier with a job. It is known that in 2004 in the United Kingdom, 79% of women working part-time did not want to go full-time (Gregory & Connolly, 2008). In perspective, the results from the United Kingdom are comparable to the trends found in other Western countries.

These results – the relative appetite of some women for part-time work – appear consistent in many studies ^[23] (Hakim, Key Issues in Women's Work: Female Diversity and the Polarisation of Women's Employment, 2004), the literature review by sociologist Catherine Hakim provides an overview (pages 69 to 73), specifying that almost all part-time work is voluntary, in that it is preferred to full-time work, in the United Kingdom, Germany, Ireland, Denmark, and the Netherlands, with Ireland being the exception. Indeed, these results are consistent with other literature reviews: “Finally, increased usual hours of work are often associated with lower job satisfaction of women but not of men” (Bender, Donohue, & Heywood, 2005).

In 1992 and 2000, in the United Kingdom, the proportion of women who were totally or very satisfied with their working hours was higher than the proportion of men who were totally or very satisfied with their working hours (McGovern, Hill, Mills, & White, 2008): this proportion decreased for both men and women between the two years ^[24] , but the decline was more pronounced among women, meaning that the proportion of satisfied women was significantly higher in 1992, with the gap with men narrowing in 2000. A survey (one thousand women and one thousand men interviewed) ^[25] Regarding employees' preferred working hours, by gender and family structure in 2000 in the United Kingdom: 71% of women in couples stated they preferred to work the same hours and maintain their current salary, while 15% would like to work more and earn more. For men in couples, these figures were 65% and 23% respectively (McGovern, Hill, Mills, & White, 2008). For single women, these figures were 63% and 29% respectively (compared to 55% and 38% for single men). It thus appears that the vast majority of employees are satisfied with their working hours, and men are more likely to want to work more and earn more. Furthermore, 10% of women in couples would like to work less and earn less (9% of men in couples). McGovern, Hill, Mills, & White suggest that this is due to rising expectations as living conditions improve and education levels increase. The decline in satisfaction does not so much reflect a deterioration of living conditions as a rise in expectations. Nearly a third of women ask to reduce their working hours, which they obtain in 90% of cases, while one in nine men ask for it, with two thirds obtaining it ^[26] .

The role of institutional factors cannot be ignored, but only considerably nuanced. The reform of school hours for nursery and primary school pupils, implemented in France in 2013, was thus able to serve as a natural experiment to assess the reaction of mothers of young children when they now had classes on Wednesdays. On the impact of institutional constraints, econometric work shows that "the reform allowed a greater number of women to work on Wednesdays, leading, in less than two years, to a reduction of one third of their participation differential on this day of the week compared to women in the control group", however "mothers do not increase their total weekly working hours" and no impact on their salaries could be observed in the short term (Van Effenterre, 2017).

In Denmark, maternity and paternity leave rules provide that women are entitled to 18 weeks of maternity leave, while men are entitled to two weeks of paternity leave during the first 14 weeks after birth (Olsen, 2007). For these 20 weeks in total, a daily allowance is paid to the parents. After 14 weeks after birth, each parent is entitled to leave for up to 32 weeks, but together they are only entitled to 32 weeks of allowance. In addition, flexible options are provided (postponement until the child is 9 years old, extension of leave, part-time work). Women take the most leave – couples are happy with the distribution. The study shows that just over half of women take the leave that the parental couple is entitled to, a quarter of women take slightly less than what they are entitled to and another quarter of women share the leave with their male partner. But the gender distribution of leave does not mean that fathers and mothers are not satisfied with their part. Up to 100% of women and men respond that they are satisfied with the distribution. 68% of couples of parents take all their parental leave with the allowances to which they are entitled. However, there are significant differences in leave use between mothers and fathers. Mothers take 92% and fathers 8% of the leave granted to couples for children. All women take 14 weeks of maternity leave and 89% of men take two weeks of paternity leave. After the 14 weeks of maternity leave, 94% of women extend their leave in the form of parental leave, while a quarter of men take parental leave. Women have an average of 28 weeks of parental leave, which can be shared between parents, and men have 8 weeks of parental leave out of a total of 32 weeks. Only 6% of men use one or more of the three flexible options (postponement, extension or return to work part-time) in the rules relating to leave compared to 36% of women. In the report, parents were asked whether they would like to see a reform making part of the leave compulsory for the mother, part for the father and a period of freedom to distribute the leave between the parents. 60% of men and 72% of women do not advocate assignment but a large minority of men do (37%). Parents were also asked whether they were satisfied with their distribution of leave within the parental couple, and almost all men and women responded positively: 98% of men and 99% of women were satisfied. The importance of flexibility is evident in the study: 31,9% of mothers (compared to 3,8% of fathers) agree with the following statement: "It was important for me to be able to postpone the leave until later if unforeseen things happen to the child" [Table 2.13 in (Olsen, 2007)].

Part-time work is associated with a relative preference for flexibility over high pay or promotions ^[27] (Hakim, 1995).

Although part-time work among women is less prevalent in Finland, the share of women working part-time for less than 30 hours per week (16%) was double that of men (8%) in 2007 and only one in ten women cited childcare as a reason for working part-time (Marin & Zólyomi, 2017).

However, it is interesting to note some counterexamples regarding the preference for part-time work.

Part-time work has neither a positive nor a negative effect on women's life satisfaction in Chile (Montero & Rau, 2015), but the impact of part-time work on life satisfaction is negative.

In the case of Honduras (López Bóo, Madrigal, & Pagés, 2009) where men prefer full-time to part-time work, not surprisingly, but contrary to what is observed in developed countries, overall women do not have a preference for part-time work (however, there is heterogeneity) and in particular poor women in Honduras prefer full-time work. Without taking into account income, both men and women prefer full-time work, but when taking into account income, women's preference for full-time work appears to be much less marked than that of men. There is therefore heterogeneity. These results come from a representative survey of 2007 of approximately 8 individuals. Interestingly, married women in Honduras do not have a preference for part-time work, unlike single women, which could be explained by greater resource needs in the presence of children. This result is opposite to what we tend to find in developed countries where part-time work tends to increase with the presence of children, although we find counter-examples in Austria (Dearing, Hofer, Lietz, Winter-Ebmer, & Wrohlich, 300). Moreover, poor women have a preference for full-time work, which reflects the effect of the budget constraint. Unsurprisingly, married men have a strong preference for full-time work and especially their preference for long hours is clearly more marked compared to women. An interesting result is also that there is no difference in preferences for part-time work between employees and the self-employed.

These data suggest that what is true for developed countries is not necessarily true for developing countries, similar to the gender paradox in science, but we still find a more marked anchoring of men in the professional sphere, in particular by the employment rate and the volume of hours worked or preferred. More specifically, with regard to the volume of hours, even full-time a differential is observed. In France, "even when women work full-time, their working time is lower when they have dependent children while the working time of men is only reduced if they raise their children alone" (Sautory & Zilloniz, 2015) or even "even when we restrict the analysis to full-time, the working hours of male employees are higher: 1730 hours for men, 1600 hours for women, or 8% less" (Askenazy, Cohen, & Senik, 2017).

Looking at excessive working hours, it can be seen that women tend not to work excessive hours often in developed countries, while they are more exposed to these excessive hours (ILO, 2011) in developing or transition countries (Ethiopia, Jordan, Nigeria, Peru, Philippines, Saudi Arabia, Thailand and Turkey): the more economic progress there is, the more women tend to work part-time [see Figure 11 in (ILO, 2011)].

Conclusion

The majority of part-time work by women is voluntary, whether by direct measurement in a declarative manner (two thirds according to the employment survey) or indirectly due to the wishes expressed by job seekers (one quarter of part-time requests). Part-time work that is mainly chosen is largely found in OECD countries (OECD, 2010). This trend is therefore not limited to France. In particular, part-time work for civil servants provides insight because part-time work is not imposed there but entirely chosen by permanent employees. The similarity of the figures between part-time work chosen in the private sector and part-time work by definition chosen in the civil service confirms that part-time work chosen in the private sector probably does not mask part-time work for family reasons that would have been suffered due to a lack of full-time employment in the event of availability in the absence of family reasons. Doubt is all the more dispelled by satisfaction measures, particularly in the Netherlands, where women are both comparatively the most part-time workers and also the most satisfied in Europe. This is not a simple correlation but rather a causality highlighted by in-depth studies (Booth & van Ours, 2010). The possible inadequacy of childcare resources does not explain most of the hourly volume, as shown by the case of the United Kingdom: among part-time mothers, nine out of ten would not wish to increase their hourly volume despite the availability of childcare when they have a partner and three out of four are in this case in single-parent families (Paull, Taylor, & Duncan, 2002). Nevertheless, there remains a minority of part-time women who wish to increase their working time. They are thus more numerous than men to be in this situation.

Regarding parental leave, in 90% of cases it is the mother's choice (Govillot, 2013). Now, the offer of childcare for young children only changes the decision to retire from work for a tiny proportion of mothers (Pora, 2020), probably because mothers often prefer to stay with young children and also because the offer has increased over the years.