Gender-Specific Duration of Parental Leave and Current Earnings

Abstract

Although male employees are increasingly making use of parental leave, gender differences in both usage and duration of parental leave are still prevalent. Based on signalling theory and the masculinities concept, the article explores the role of gender in the relationship between the incidence/duration of parental leave and wages/compensation after returning to a job. It is shown that pay gaps associated with parental leave are much more severe for male than they are for female middle managers in the German chemical industry.

Keywords

compensation earnings gender parental leave signalling wages

Introduction

For quite some years now, policymakers have been trying to increase fathers’ share of parental leave in many countries in order to align the labour market outcomes for females and males. Although legislative reforms have shown some positive effects (see e.g. Ekberg et al., 2013 for Sweden), gender differences in both usage and duration of parental leave are still prevalent (Destatis, 2019b; Statistics Sweden, 2018; World Economic Forum, 2018).

The research comprehensively addresses the role of parental leave for household issues such as childcare, housework and maternal health (Bünning, 2015; Persson and Rossin-Slater, 2019; Schober, 2014). Scholars have also investigated the situation in which employees return to work, pointing out earnings effects of previous career interruptions in general (Beblo and Wolf, 2002; Spivey, 2005) and of parental leave in particular. Most of these studies focus on women, as they have been more likely to take parental leave than men (Anderson et al., 2003; Budig and Hodges, 2010; Buligescu et al., 2009; Gangl and Ziefle, 2009; Lott and Eulgem, 2019; Schmelzer et al., 2015). Only a few studies concentrate on parental leave of males, i.e. paternal leave (Coltrane et al., 2013; Evertsson, 2016; Rege and Solli, 2013). Usually, lower wages subsequent to such career interruptions are explained by human capital arguments, including that of skill depreciation (Mincer and Polachek, 1974), or by negative signalling effects of parenthood (Coltrane et al., 2013). Evidence for gender comparisons of parental leave for future careers and compensation is rare and has predominantly examined the Swedish labour market; the results of Stafford and Sundström (1996) and Albrecht et al. (1999) indicate that parental leave is related to lower subsequent wages for male employees in particular. Moreover, results from Evertsson (2016) hint at effects on wages for males even in cases of relatively short parental leave spells compared to interruptions for females.

Our contribution extends the literature by explicitly exploring possible gender differences in the relation between the incidence or duration of parental leave and later earnings after returning to the job. Possible explanations of gender differences may include (i) gender-specific depreciation rates of human capital based on job segregation (Görlich and de Grip, 2009; Kunze, 2002) or (ii) gender-specific signalling effects that might be relevant in the sense that males may be violating social norms and thus not meeting the expectations of employers by making use of parental leave (Albiston, 2007; Wharton et al., 2008). We make use of a large longitudinal sample of middle managers in the German chemical sector. The dataset comprises a very homogeneous group of employees regarding education and jobs compared to much broader samples used in previous studies. The observed managers were all working in middle management positions within the German chemical sector and held a university degree in the STEM area so job segregation did not play a major role. This article also contributes to the literature on gender pay gaps by highlighting differences for groups of employees with and without periods of parental leave. Since expected or actual gender pay gaps interact with inequalities of labour supply decisions within households (Albanesi and Olivetti, 2009), insights from this article are relevant from a policy point of view, too.

The remainder of the article is organised as follows: Section 2 presents the relevant theoretical considerations and provides an overview of previous empirical research. On this basis, hypotheses on how parental leave may be associated with earnings consequences are derived. The data and variables are described in Section 3. Section 4 presents the results. Section 5 gives further estimations, robustness checks and remaining limitations. Section 6 provides a discussion and Section 7 concludes.

Theoretical considerations, previous empirical studies and hypotheses

Observed negative effects of periods of parental leave on future earnings after returning to work are often explained by human capital theory in conjunction with the skill depreciation hypothesis. According to this theory, workers’ remuneration depends positively on their stock of human capital. Workers increase their value to the employer through the accumulation of work skills and knowledge, resulting in greater productivity (Becker, 1964; Mincer, 1958; Mincer and Polachek, 1974). During periods of parental leave, employees forgo work experience and skill development and defer investments in on-the-job training until they return to the labour market. The subsequent payoff period of investments in on-the-job training is consequently shortened. In expectation of lower total returns over the further course of the career, rational employees hence provide fewer investments in training activities than workers with continuous career paths (Corcoran, 1979; Mincer and Ofek, 1982). It is also conceivable that the stock of already acquired human capital may even depreciate during the period of absence (Mincer and Polachek, 1974). Employees may forget accustomed operating processes because they are no longer using them regularly. Besides, they may fail to adopt new techniques or to keep up with technical progress during their absence from work and may, therefore, be confronted with technical as well as economic obsolescence of their human capital (Beblo and Wolf, 2002; Edin and Gustavsson, 2008; De Grip and van Loo, 2002). Even without depreciation of human capital, a negative relation between the incidence and duration of parental leave and earnings after returning to the labour market may still be observed, because the human capital has still decreased relatively compared to that of workers without career interruptions, who have continued to increase human capital during that period (Neuman and Weiss, 1995). In addition to human capital depreciation, parental leave spells may evoke negative expectations on the part of the employers regarding the employee’s commitment and work dedication (Cohen and Single, 2001; Coltrane et al., 2013). In addition, employers may make statistically discriminatory wage offers to employees, when sufficient information is lacking for establishing the latter’s productivity levels. Accordingly, employees with parental leave spells would receive lower wages because they are either perceived to be less productive on average than those without career interruptions or because their productivity levels are more difficult to assess due to these levels being subject to a higher variance. This leads to:

Hypothesis 1: Employees who have taken parental leave currently receive lower pay compared to employees with a continuous career path.

The arguments presented above become even more relevant with the length of parental leave. This directly leads to:

Hypothesis 2: The interruption pay gap increases with the duration of parental leave.

The question then arises of whether male and female employees are affected to the same extent. Exploring employment gaps of managers of a large US financial service organisation, Judiesch and Lyness (1999) do not find any significant gender-specific effects with regard to promotions and wage increases after general leaves of absence. Stafford and Sundström (1996), Albrecht et al. (1999) and Gerst and Grund (2019) find a more pronounced negative payment effect after parental leave for male compared to female workers in Sweden and Germany. Evertsson (2016) focusses on wages during the first years subsequent to returning to the labour market. Her results indicate that the wages of males are already affected after relatively short parental leave spells, whereas females only suffer when making use of much longer maternal leave spells in the case of Sweden.

From a theoretical perspective, the combination of a number of approaches is likely to be relevant here.¹ It can be presumed that gender-specific social norms, the construct of masculinities and signalling effects reinforce one another. Signalling theory dates back to Spence (1973). He introduced the argument that firms can use observables, such as educational degrees, as credible positive signals for otherwise unobservable individual productivity or ability of employees (e.g. for determining wages). There are theoretical considerations and empirical evidence that employees’ long working hours act as another positive signal (Golden and Altman, 2008; Landers et al., 1996; Rebitzer and Taylor, 1995). In contrast, career interruptions, such as unemployment spells, are argued to be interpreted by firms as negative signals, which can lead to some stigma effects regarding affected employees (Gibbons and Katz, 1991; Vishwanath, 1989). Coltrane et al. (2013) have convincingly applied the theoretical argument to possible stigma effects due to reductions of working hours for family reasons, although the acceptance among employers and colleagues may have increased in recent years (Lott and Klenner, 2018).

A first argument for differences between men and women may stem from stereotypes that lead to gender-specific social norms. Research in social psychology indicates that traditional gender roles and the gender division of labour act as main reasons here (Eagly, 1987; Eagly and Steffen, 1984; Rudman and Phelan, 2008). Acker (1990) and Williams (2001) put forward the concept of “ideal workers”, who are completely and fully dedicated to their work and unencumbered by external family obligations. The ideal worker concept is attributed more to male workers, who traditionally take on the role of the breadwinner or good provider, whereas women are expected to serve as the primary caregivers (Rudman and Mescher, 2013). Hence, gender-specific social norms are closely connected to the construct of masculinities in the workplace. An important facet of the masculine model is to show endurance (Gould, 1974: 97) revealed by continuous employment (Collinson and Hearn, 2005: 294).

The model is primarily used to explain gender inequalities with advantages for men, who reinforce their position of power at work. These norms can then lead to gender-specific expectations from employers towards the use of parental leave. Indeed, evidence shows that fathers on average work even more hours than childless men (e.g. Hodges and Budig, 2010). However, these advantages may only hold as long as males fulfil the attributes or expectations towards masculinities. Already earlier work stresses the relevance of heterogeneity among the group of men and that some subgroups, e.g. with regard to skin colour or sexual orientation, do not have the same privileges as others (Messner, 1997: 6). Accordingly, males with parental leave spells still contravene against social norms and may lose usual male privileges.

In contrast, women may face only small future salary cuts subsequent to actual parental leave spells, if firms already anticipate the likelihood of maternal leave after recruiting females (Schneer and Reitman, 1990). While women tend to fulfil the norm when making use of maternal leave, men, in contrast, are violating the male norm of the ideal worker (Acker, 1990; Williams, 2001). This can lead to the relevance of gender specific signals of identical behaviour: paternal leave may act as a negative signal for males with regard to their lacking career commitment (Albiston, 2007; Wharton et al., 2008). In contrast, this may not be or may be less the case for females. Consequently, males who take periods of parental leave will be stigmatised more than female workers, resulting in the former’s lower earnings and limited future career opportunities (Cohen and Single, 2001; Coltrane et al., 2013). These considerations lead to:

Hypothesis 3: The negative relation between parental leave and subsequent compensation is more pronounced for males than it is for females.

Data, variables and descriptive statistics

This article utilises data from a yearly salary survey among middle managers in the German chemical sector during 2013 to 2018. The respondents were members of the German Association of Managers in the Chemical Industry (official abbreviation: VAA), which is a German professional association for non-tariff and executive employees. Contrary to most regular employees in the sector with collectively agreed wages and wage increases, there is much more discretion in negotiations of the wage level and in determination of wage increases between middle managers and employers.

The survey collected individual information on workers’ demographics, such as work experience, field of study and job characteristics (e.g. level of hierarchy, functional area and firm size). For the main analysis, the sample was limited to university graduates with a STEM degree, working full-time in the chemical industry. In addition, top managers were excluded from the analysis, as the wage-setting process for them might also differ considerably from that for middle managers. Companies in the chemical industry are predominantly located in Western Germany (Heller, 2014: 6). Since there are still some path-dependent differences in wage settings, observations of the federal states in Eastern Germany (formerly the German Democratic Republic: GDR) were excluded. The analysis was further restricted to employees who have had either no career interruption at all or no interruption other than periods of parental leave. These restrictions resulted in a rather homogenous sample of managers who had a similar educational background and were working in the same industry (n=17,141). The sector is somewhat male-dominated, which was also reflected in the sample. About 13% of observations did come from females, though (n=2,187).

Throughout the study, the yearly total compensation of managers was used as the main dependent variable. All observations were deflated by the consumer price index to the base year of 2013. On average, managers in the sample received a total compensation of €134,000. Compensation differed considerably between males (€138,000) and females (€112,000, see Table 1).

Table 1.

Descriptive statistics of total compensation by gender (in €).

		Whole sample (n=17,141)	Females (n=2,187)	Males (n=14,954)
Total compensation	Mean	134,397	112,098	137,659
	SD	57,006	45,719	57,758

Survey participants were asked to report the total number of months that they had taken for parental leave during their professional careers, i.e. after graduating from university. One out of ten individuals reported having taken parental leave during their career. This share was considerably higher among females (22%) than it was for males (8%). Male and female employees also differed with respect to the total duration of parental leave during their careers. More than half of the male workers with interruptions had made use of up to two months of parental leave only, whereas this was true for only a small minority of females (5% of employees with interruptions due to parental leave). In contrast, rather long-term interruptions of more than 12 months were much more relevant for female employees than they were for males. Both the median and the mode of parental leave duration by women pointed to 12 months, while they were two months for male managers (see Figure 1).

Figure 1.

Distribution of parental leave duration by gender in sample.

These gender-specific distributions are likely to be affected by formal parental leave regulations in Germany. In addition to a mandatory “maternity protection” period of 14 weeks for mothers surrounding childbirth, parents are eligible to take job-protected parental leave of up to three years. A parental allowance as a percentage of the previous salary is provided by public entities for up to 14 months in total for the two parents to compensate for the unpaid leave from work. Each parent is entitled to the parental allowance for a maximum of 12 months, though. More representative data from Germany confirm that most fathers with parental leave spells (79%) receive the parental allowance for two months only. Mothers, on the other hand, receive the parental allowance for 12 months on average (BMFSFJ, 2016). Consequently, the duration of parental leave has been divided into three categories according to the aforementioned thresholds. In the following, these are (i) up to two months of parental leave, (ii) three to twelve months and (iii) more than twelve months of parental leave. Using these categories allowed the identification of possible non-linear relations between parental leave duration and current remuneration. Table 2 shows the corresponding distribution of observations. These differ considerably by gender.

Table 2.

Distribution of parental leave duration categories by gender.

	Whole sample (n=17,141)	Females (n=2,187)	Males (n=14,954)
No parental leave	0.899	0.781	0.916
1-2 months	0.041	0.012	0.045
3-12 months	0.047	0.126	0.036
>12 months	0.013	0.081	0.003

Next to parental leave, several covariates were used in the subsequent regression to adjust for potential confounding effects on compensation. Along with gender, other individual characteristics, such as work experience (in years), field of study (seven dummies) and a dummy for holding a doctoral degree, were taken into account. It is evident that middle managers in the sample were very experienced, with an average of 22 years of professional experience. This also reflects the aging workforce in the chemical industry (Gehrke et al., 2018). Confounding effects caused by job-specific characteristics were also controlled for in additional estimations, which allowed a disentangling of wage effects between and within comparable jobs. The German chemical sector is characterised by rather long-term employment relations, which was expressed by an average amount of firm tenure of 17 years. Almost half of the middle managers were employed in large firms with more than 10,000 employees. In addition, participants were asked to assign themselves to one of three hierarchical levels of middle and senior management. The top management (Level 1) was excluded because of essential different compensation principles. Being either employed in middle or senior management, two-thirds of the managers had already moved up in the hierarchy, leaving the lower middle management (Level 4) with some management responsibilities behind them. Next, the functional areas that managers worked in (11 dummies) and managers’ actual weekly working hours were also queried. Women in the sample had both less professional experience and shorter firm tenure (t-test, p<0.001). They also held a position at the lowest level of the hierarchy more often than male employees do (50% > 31%, chi-squared<0.001). Table 3 provides sample descriptive statistics of these variables.

Table 3.

Descriptive statistics.

	Whole Sample(n=17,141)		Females(n=2,187)		Males(n=14,954)
	Mean	SD	Mean	SD	Mean	SD
Individual characteristics
Female	0.128
Experience (in years)	22.06	8.726	18.60	9.295	22.57	8.523
Field of study
Chemistry	0.602		0.560		0.609
Engineering	0.200		0.065		0.220
Biology	0.054		0.116		0.045
Physics	0.027		0.012		0.029
Medical science	0.019		0.055		0.014
Pharmaceutics	0.060		0.129		0.050
Other natural sciences	0.038		0.063		0.034
Doctoral degree	0.807		0.782		0.810
Job characteristics
Tenure (in years)	16.53	9.630	12.49	9.119	17.12	9.560
Level of hierarchy
Level 2	0.115		0.064		0.122
Level 3	0.552		0.437		0.568
Level 4	0.334		0.499		0.310
Functional area
Production	0.192		0.110		0.204
Research & Development	0.344		0.451		0.328
Technical & Process Engineering	0.087		0.025		0.096
Applications Engineering	0.059		0.049		0.061
Sales, Marketing, Logistics, Sourcing	0.085		0.074		0.087
Finance, Controlling, Human Resources	0.039		0.028		0.041
Technical supervision	0.057		0.048		0.058
IT	0.020		0.013		0.021
Other	0.117		0.202		0.105
Firm size
Fewer than 1,000	0.182		0.226		0.175
1,000–5,000	0.204		0.204		0.205
5,001–10,000	0.121		0.113		0.123
>10,000	0.493		0.457		0.498
Weekly working hours	45.83	5.392	44.75	5.372	45.99	5.376
Year
2013	0.185		0.190		0.184
2014	0.177		0.161		0.179
2015	0.168		0.167		0.168
2016	0.163		0.164		0.163
2017	0.158		0.167		0.156
2018	0.150		0.151		0.150

Main results

In order to explore the relation between the incidence or duration of parental leave and earnings after returning to the job, Mincer-type wage regressions were run. The results were obtained using pooled ordinary least square (OLS) estimations with robust standard errors clustered at the firm level. Individual fixed effects estimations would have been inappropriate for the case at hand, since parental leave spells mainly occurred prior to the observation period, which is why there is hardly any variation within a participant over the observation period.²

Hypotheses 1 and 2 were tested by conducting a two-stage hierarchical regression analysis (see Table 4). Both presented models regressed the logarithm of the employees’ annual total compensation on previous parental leave spells. Leaves of different duration were taken into account: “No leave”, “1-2 months”, “3-12 months” and “>12 months”. Model (1) considered only individual characteristics, showing significantly lower earnings for managers who had taken parental leave spells compared to those employees without career interruptions. These increased with the duration of parental leave. The results are thus in line with the baseline hypotheses 1 and 2 and confirm previous empirical results.

Table 4.

Pooled OLS regressions on total earnings.

Log total compensation	(1)		(2)
Parental leave (base: No leave)
1–2 Months	-0.0170*	(0.010)	-0.00419	(0.006)
3–12 Months	-0.0455***	(0.015)	-0.0156	(0.010)
>12 Months	-0.0965***	(0.022)	-0.0902***	(0.016)
Female (1=yes)	-0.0672***	(0.010)	-0.0280***	(0.008)
Experience	0.0650***	(0.003)	0.0431***	(0.003)
Experience squared	-0.000937***	(0.000)	-0.000652***	(0.000)
Field of study (base: Chemistry)
Engineering	0.123***	(0.029)	0.0705***	(0.018)
Biology	-0.0326	(0.023)	-0.000739	(0.015)
Physics	0.0315	(0.032)	0.0258	(0.020)
Medical Science	0.0936**	(0.042)	0.102***	(0.024)
Pharmaceutics	0.0900***	(0.023)	0.0900***	(0.016)
Other natural sciences	0.0313*	(0.018)	0.0262*	(0.014)
Doctoral degree	0.0810***	(0.011)	0.0450***	(0.010)
Tenure			0.00506***	(0.001)
Level of hierarchy (base: Level 3)
Level 2			0.253***	(0.038)
Level 4			-0.165***	(0.017)
Functional area (base: R&D)
Production			0.0203**	(0.010)
Technical & Process Engineering			-0.00164	(0.013)
Applications Engineering			-0.00869	(0.018)
Sales, Marketing, Logistics, Sourcing			0.0691***	(0.014)
Finance, Controlling, HR			0.0480***	(0.012)
Technical Supervision			-0.0183	(0.012)
IT			-0.0297*	(0.018)
Other			-0.0124	(0.013)
Firm size (base: 1,000–5,000 employees
Fewer than 1,000			-0.133***	(0.014)
5,001–10,000			0.0651**	(0.027)
>10,000			0.137***	(0.014)
Actual weekly working hours			0.00944***	(0.001)
Year dummies	Yes		Yes
Intercept	10.72***	(0.049)	10.52***	(0.096)
Observations	17,141		17,141
Adj. R-squared	0.463		0.662

Notes: Cluster-robust standard errors in parentheses; significance level: *p < 0.10. **p < 0.05. ***p < 0.01.

Next, the job-specific characteristics were entered in model (2). After accounting for tenure, hierarchy level within the firm, functional areas, working hours and firm size, a negative correlation remained but only those spells of leave that lasted longer than 12 months were significant. Step-by-step inclusion of the variables (not reported) showed that the level of hierarchy and weekly working hours were major reasons for the decrease in the coefficients representing parental leave durations. Those employees who had taken parental leave had lower working hours than those with continuous career paths, even among those with full-time contracts in the sample. This hints at an altered time allocation towards taking on more family responsibilities after returning from parental leave on average. Similarly, Bünning (2015) found that fathers who had taken parental leave subsequently reduced their working hours and increased their involvement in childcare even after short spells of leave. In addition, employees with previous parental leave spells were assigned to jobs at lower levels of the hierarchy, indicating that taking parental leave conflicts with climbing the internal hierarchical ladder or achieving inter-firm career progress. These findings are in line with the results of Judiesch and Lyness (1999), who found evidence that leaves of absence are associated with fewer subsequent promotions. Besides, individuals who had taken parental leave for more than a year faced pay gaps amounting to almost 10% even within similar jobs.

Table 4 also reveals the relevance of gender pay gaps in the chemical sector.³ Despite the consideration of parental leave spells and the adjustment for confounding effects, a gender pay gap of around 3% was still observed. In a next step it was investigated whether the reported pay gaps due to parental leave were more severe for males, as postulated in hypothesis 3. Expecting a moderating effect, parental leave dummies were interacted with gender in Table 5. Again, model (1) provided the baseline model, in which individual characteristics were controlled for, next to the categorical variable of parental leave, indicating various durations. The female dummy now showed a significant gender pay gap amounting to 8% among employees without parental leave. Male workers who had taken parental leave received significantly lower earnings than men without career interruptions. On average, men with up to two months of parental leave received 2% lower earnings than male workers with continuous career paths. This difference increased to almost 25% for men who had taken more than twelve months away from work due to caring for their children. Most importantly, the interaction between parental leave and gender was also significant for periods of parental leave longer than three months. In addition, applying a Wald test on the interaction term, which provides a test that all coefficients of the term are simultaneously equal to zero, yielded a negative result (Wald test, p=0.002). Therefore, there was a statistically significant interaction between the two variables, which provides support for hypothesis 3. These results were reinforced by additionally considering job characteristics in model (2), although differences were considerably smaller and as they were captured by further control variables.

Table 5.

Pooled OLS regressions on total compensation evaluating the interaction of gender and duration of parental leave.

Log total compensation	(1)		(2)
Parental leave (base: no leave)
1–2 Months	-0.0215**	(0.010)	-0.00979	(0.007)
3–12 Months	-0.0739***	(0.023)	-0.0295**	(0.014)
>12 Months	-0.245***	(0.041)	-0.128***	(0.023)
Female (1=yes)	-0.0815***	(0.010)	-0.0355***	(0.008)
1-2 Months x Female	0.0408	(0.106)	0.104	(0.073)
3-12 Months x Female	0.0901***	(0.029)	0.0439**	(0.018)
>12 Months x Female	0.194***	(0.046)	0.0532**	(0.025)
Individual characteristics	Yes		Yes
Job characteristics	No		Yes
Year dummies	Yes		Yes
Intercept	10.73***	(0.050)	10.52***	(0.096)
Observations	17,141		17,141
Adj. R-squared	0.464		0.662

Notes: Cluster-robust standard errors in parentheses; significance level: **p < 0.05. ***p < 0.01.

Individual characteristics: experience, experience squared, field of study, doctoral degree.

Job characteristics: firm tenure, level of hierarchy, functional area, firm size, actual weekly working hours.

The empirical pattern becomes even more illustrative through plotting the relationship between total compensation and parental leave categories in Figure 2.⁴ First, the plot shows that parental leave was much more noticeably linked to lower earnings for males. In contrast, females with minor parental leave duration of up to two months even received a somewhat higher wage than those without leave spells (not significantly, though). In addition, average payments for female employees with 3-12 months of parental leave during their careers did not differ significantly compared to those for women with continuous career paths. A significant pay gap was only observed between female managers who have had career interruptions of more than 12 months and those without any interruption. Men, on the other hand, faced significant earning disadvantages already after being on parental leave for more than three months during their careers. A possible explanation for the moderating effect of gender is that employers still have gender-specific expectations about the usage of parental leave, as discussed above. Taking parental leave for a certain period of time may lead to gender-specific signals with regard to career commitment. Whereas employers may be positively surprised by short maternal leave decisions of two months, firms may assess longer paternal leave decisions as a bad signal. This issue will be discussed in the conclusion below.

Figure 2.

Interaction plot for parental leave and gender (controlled for individual and job characteristics).

Figure 2 can also be read as gender pay gaps for separate parental leave categories. The typical gender pay gaps in favour of males for employees without parental leave were observed, which might result from statistical discrimination based on gender-specific ex-ante expectations. In this sense, though, expectations are not met if males make use of parental leave and if females take only a few months off from work. As a result, employers may adapt their ex-ante expectations of individuals’ career commitment or behaviour, which may have led to the observed diminishing or even reverse gender pay gaps among employees with parental leave spells. Within the group of managers with parental leave periods, the pay gap between men and women was not significant for any of the parental leave categories.

Further estimations, robustness checks and remaining limitations

The estimated associations were robust to a variety of checks. First, the patterns reported for fixed salaries and bonus payments hardly differed from those reported for total compensation (see Figure B and Table A in the Appendix for corresponding estimation results). Second, the group of employees without career interruptions consisted of employees without children and of parents who choose not to take parental leave. It is conceivable that employees with children but without parental leave spells differ in certain characteristics, e.g. career commitment, both from those employees without children and from those who had taken parental leave. Hence, the model specifications used for Table 5 were re-estimated on two sub-samples to detect potential distortions. Considering only parents among the employees, the results were found to be robust, albeit the pay gaps compared to employees without parental leave spells and the corresponding interaction terms with gender were even somewhat larger than previously described (see Table 6).⁵ This hints at a positive signal of career commitment from those parents without interruptions, in particular from women.

Table 6.

Re-estimations of Table 5 for a sub-sample of employees with children.

	(1)		(2)
Parental leave (base: no leave)
1–2 Months	-0.0596***	(0.012)	-0.0307***	(0.009)
3–12 Months	-0.109***	(0.020)	-0.0459***	(0.011)
>12 Months	-0.263***	(0.048)	-0.144***	(0.022)
Female (1=yes)	-0.0594**	(0.028)	-0.0417**	(0.021)
1-2 Months x Female	0.0486	(0.084)	0.132**	(0.055)
3-12 Months x Female	0.0836**	(0.041)	0.0587**	(0.029)
>12 Months x Female	0.163***	(0.060)	0.0618*	(0.037)
Individual characteristics	Yes		Yes
Job characteristics	No		Yes
Year dummies	Yes		Yes
Intercept	10.78***	(0.036)	10.59***	(0.102)
Observations	10,578		10,578
Adj. R-squared	0.358		0.608

Notes: Cluster-robust standard errors in parentheses; significance level: *p < 0.10. **p < 0.05. ***p < 0.01.

Individual characteristics: experience, experience squared, field of study, doctoral degree.

Job characteristics: firm tenure, level of hierarchy, functional area, firm size, actual weekly working hours.

Third, it was also controlled for the number of previous employers of the employees. Previous research generally hints at a positive relation between external job mobility and compensation (Lam et al., 2012). However, due to the predominantly long-term employment relationships in this particular sector, only few external job changes occurred without significant gender differences in the sample. Accordingly, the results were robust with regard to this additional control. In addition, the results were also robust with regard to an additional consideration of past part-time employment spells, which are more relevant for women, especially when they want to/have to take care of their children (Destatis, 2019a).

Fourth, earnings penalties for parental leave may differ across the earnings distribution (Budig and Hodges, 2010). Estimating the association between parental leave and compensation at the mean may therefore produce an incomprehensive image of the true relationship. Gender-specific quantile estimations (available upon request) show, however, that parental leave pay gaps were quite stable across the distributions. If at all, they were slightly more pronounced at the bottom of the distribution for female workers, while being more pronounced at the top for male counterparts.

Some limitations remain. Parental leave spells have only been observed in retrospect, and survey participants did not have to provide information on the exact timing of their parental leave spell(s). The majority of parental leave spells occurred prior to the observation period. Thus, short-term effects could not be precisely disentangled from long-term ones. Neither was an ex-ante ex-post analysis feasible. The issue was addressed by running separate estimations on total compensation for younger and for more mature employees, performing a median-split at 25 years of experience. The results indicate that differences in pay gaps between male and female workers were more pronounced among the group of employees with less than 25 years of work experience. The lack of statistical significance of the interaction effect for the older group may be explained by the fact that male workers hardly ever took parental leave during their early careers in previous decades; thus, the general link between parental leave and compensation is almost completely represented by women. Among the group of employees with more than 25 years of work experience, male and female employees both faced considerable compensation disadvantages already after parental leave spells that lasted longer than three months in total. In contrast, women in the younger group even had a compensation advantage by taking parental leave of up to 12 months, similar to what was already described for the entire group in Table 5 (now significant, though). At the same time, male employees in the more experienced cohort faced significant earnings disadvantages already after three months of parental leave (see Figure C and Table B in the Appendix).

Lastly, we examined whether individuals who have taken spells of leave were able to catch up or whether pay gaps actually increased over time. Therefore, the relative wage increases were compared between employees with and without parental leave spells. Rates of change were calculated based on information provided by the survey participants in two consecutive years (n=8,244 observations). Employees without interruptions faced an average increase in total compensation amounting to +0.041 (n=7,391). The average increase of those employees who had taken parental leave spells in the past was slightly higher at +0.056 (n=853). Interestingly, individuals who had taken leave spells of up to 12 months in total did have even a slightly higher salary adjustment (+0.059), while employees whose leave spells altogether lasted longer than 12 months only received an increase of +0.024 on average. However, when controlling for individual (and job-related) characteristics, by applying OLS estimations on the wage increases, no considerable differences in wage increases were found between employees who had taken parental leave spells and those who had not (see Table 7). Coefficients were rather negative for males, but not for females, though (see Table C in the Appendix).

Table 7.

Pooled OLS regressions on relative increases in total compensation.

	(1)		(2)
Parental leave (base: no leave)
1–2 Months	-0.00526	(0.008)	-0.00501	(0.008)
3–12 Months	0.00422	(0.007)	0.00689	(0.007)
>12 Months	-0.0161	(0.011)	-0.0157	(0.012)
Female (1=yes)	0.00170	(0.005)	0.00379	(0.005)
Individual characteristics	Yes		Yes
Job characteristics	No		Yes
Year dummies	Yes		Yes
Intercept	0.132***	(0.017)	0.122***	(0.018)
Observations	8,244		8,244
Adj. R-squared	0.025		0.030

Notes: Cluster-robust standard errors in parentheses; significance level: ***p < 0.01.

Individual characteristics: experience, experience squared, field of study, doctoral degree.

Job characteristics: firm tenure, level of hierarchy, functional area, firm size, actual weekly working hours.

Discussion

The results presented above challenge current theory and may provide a basis for its further development. As argued in Section 2, human capital theory can certainly be used as a possible explanation for the interruption pay gaps, which increase with the duration of gaps in general (parental leave in this case). Since a rather homogeneous sample of employees, all with similar educational backgrounds and all holding similar jobs in the chemical industry, is used for this study, the possible relevance of gender-specific human capital for gender-specific impact of parental leave on subsequent remuneration cannot be tested. Previous evidence has cast doubt, however, on whether human capital theory may be considered to be a major explanatory approach in this context because of only minor gender differences in job-specific human capital depreciation (England, 1982).

Signalling theory also serves as an explanatory approach. Accordingly, employees may reveal information about their lower career commitment through the use of parental leave. The results of this contribution can be used for a possible extension of signalling theory by considering gender-specific signals, since employers’ expectations can differ by gender. While males who have taken parental leave deviate from the expectation or norm of the ideal worker and send a negative signal, females can actually send a positive signal about their career commitment by opting for considerably less than the full year of maternal leave that is taken by most mothers in our sample and that is more in line with the prevailing social norm. Certainly, these implications are highly dependent on gender-specific expectations in the given context of this study. Future empirical and theoretical work might take precise gender-specific expectations or social norms more explicitly into account. This call to action is not limited to parental leave; it applies to every topic regarding work and employment, where gender-specific expectations are likely to be relevant (e.g. full- vs part-time employment, assignments abroad/longer business trips, or participation in further training).

The possible negative signal of males who make use of paternal leave is also likely to be relevant for the masculinities concept by highlighting an additional dimension of heterogeneity within the group of males. Male privileges, such as wage premiums, only hold when expectations are met, and they are not only lost, but can even reverse completely when men violate expectations. Two different comparisons are relevant: first, certain groups of males can lose privileges and fall short of other males who conform to the masculinity norms. In this context, Coltrane et al. (2013) speak of a possible flexibility stigma of males who make use of flexible work arrangement. Second, males deviating from masculinity norms can even be disadvantaged (e.g. regarding compensation) relative to females with corresponding characteristics and employment histories. In consequence, a kind of privilege reversal is relevant, if the expected traits or behavioural patterns are not met. These considerations do not question the relevance of male privileges in general, however, they raise necessary awareness for within gender heterogeneity and gender-based expectations by firms that lead to distinct gender specific signals elicited by the same behaviours. This contribution can therefore be seen as a first step towards a linkage between signalling theory and the masculinities concept.

The results presented above are also relevant for the much broader debate on inequality in organisations and the labour market as a whole. First, they reveal that subgroups, e.g. with regard to the use of parental leave spells, have to be taken into account when focusing on gender inequality. In a more general approach, England et al. (2020) show a slowdown (even a stalling of some indicators) in the movement towards gender equality regarding work and employment over five decades in the United States. These findings lead to their plausible conclusion that further progress towards gender equality will only be likely with substantial institutional change. The previously presented considerations also indicate that gender-specific expectations, differences in individual decisions, and employment inequality will most probably reinforce each other in a seemingly infinite loop without such change.

Conclusion

This study finds evidence of the important role of gender in the relation between previous parental leaves and current remuneration. Men face much more severe interruption pay gaps than females do. This even turns around gender pay gaps among employees with parental leave in favour of females. Although the proportion of fathers taking parental leave has increased in recent years, the incidence/duration of parental leave is still dominated by mothers. The results of this article cast doubt on a full alignment in the near future if interruption pay gaps continue to be relevant for fathers in particular. Assuming that these differences are not completely caused by differences in preferences between males and females, this leads to a dilemma in the sense that female-dominated parental leave and more pronounced interruption pay gaps for males reinforce one another.

The results are important for theory and policy at the firm as well as on the societal level: there may already be approaches at the firm and the societal level to overcome this dilemma by implementing some institutional changes. Firms may have their own incentive for promoting workplace practices that support parental leave for males as well as females. Although empirical results hint at a positive relationship between family-friendly practices and productivity (Konrad and Mangel, 2000), Bloom et al. (2011) show that this relation is mainly driven by the quality of other implemented management practices.

One possibility at the societal level, in addition to subsidised early child care (Baker et al., 2008), is to provide parental leave benefits for up to a maximum of n months of allowances, where each parent has the opportunity to make use of a maximum of n/2 months without the possibility of transferring unused leave entitlements to the other parent. Consequently, a gender alignment in parental leave would be incentivised more intensely, which may also diminish gender-specific expectations and social norms of firms, gender pay gaps and gender-specific interruption pay gaps in the future. The implementation of a comparable parental leave allowance system in Iceland has shown expected effects on the participation of fathers in child care (Gíslason and Símonardóttir, 2018). Various measures have been introduced in order to encourage fathers to use parental leave in Sweden. While the introduction of one month reserved specifically for each parent under a “use it or lose it” system in 1995 and its stepwise extension to three months nowadays have increased the proportion of fathers taking parental leave, while an additional introduction of tax credits for gender equality has not affected behaviour that much (Duvander and Johansson, 2012).

Unterhofer and Wrohlich (2017) argue that next to the direct effects of such incentives in terms of individual decisions, also indirect, more long-term effects can be prevalent, including slowly changing social norms and attitudes as well as a culture towards gender roles within society. Future work may also try to integrate the different facets of the topic at the individual and the household level with those at the corporate and the societal level and validate the results for different institutional contexts and for other facets of the employment relationship.

Supplemental Material

sj-docx-1-wes-10.1177_09500170221090163 – Supplemental material for Gender-Specific Duration of Parental Leave and Current Earnings

Supplemental material, sj-docx-1-wes-10.1177_09500170221090163 for Gender-Specific Duration of Parental Leave and Current Earnings by Benedikt Gerst and Christian Grund in Work, Employment and Society

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/ or publication of this article.

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Christian Grund

Supplemental material

Supplemental material including the appendix for this article is available online.

Notes

Benedikt Gerst received his doctoral degree from RWTH Aachen University in 2021. He studied Business Administration and Economics at Stellenbosch University, the University of Passau (B.Sc., 2010) and the University of Göttingen (M.Sc., 2014).

Christian Grund is Professor of Human Resource Management and Personnel Economics at RWTH Aachen University. He studied economics and business administration at the Universities of Hannover (Diplom-Ökonom, 1997) and Bonn (Dr. rer. pol., 2001). He was a visiting scholar at Stanford University (Graduate School of Business), Aarhus Business School and McGill University (Montreal). Before he moved to Aachen in 2012, Christian Grund held professorial positions at the University of Würzburg and the University of Duisburg-Essen.

References

Acker

(1990) Hierarchies, jobs, bodies: a theory of gendered organizations. Gender & Society 4(2): 139–158.

Albanesi

Olivetti

(2009) Home production, market production and the gender wage gap: incentives and expectations. Review of Economic Dynamics 12(1): 80–107.

Albiston

(2007) Institutional perspectives on law, work, and family. Annual Review of Law and Social Science 3(1): 397–426.

Albrecht

Edin

P-A

Sundström

(1999) Career interruptions and subsequent earnings: a reexamination using Swedish data. Journal of Human Resources 34(2): 294–311.

Anderson

Binder

Krause

(2003) The motherhood wage penalty revisited: experience, heterogeneity, work effort, and work-schedule flexibility. Industrial and Labor Relations Review 56(2): 273–294.

Baker

Gruber

Milligan

(2008) Universal child care, maternal labor supply, and family well-being. Journal of Political Economy 116(4): 709–745.

Beblo

Wolf

(2002) Die Folgekosten von Erwerbsunterbrechungen. Vierteljahrshefte zur Wirtschaftsforschung 71(1): 83–94.

Becker

(1964) Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. New York: NBER Books.

Bloom

Kretschmer

van Reenen

(2011) Are family-friendly workplace practices a valuable firm resource? Strategic Management Journal 32(4): 343–367.

10.

BMFSFJ (2016) Bilanz 10 Jahre Elterngeld. Bundesministerium für Familie, Senioren, Frauen und Jugend. Berlin: BMFSFJ.

11.

Budig

Hodges

(2010) Differences in disadvantage: variation in the motherhood penalty across White women’s earnings distribution. American Sociological Review 75(5): 705–728.

12.

Buligescu

de Crombrugghe

Menteşoğlu

, et al. (2009) Panel estimates of the wage penalty for maternal leave. Oxford Economic Papers 61(1): i35–i55.

13.

Bünning

(2015) What happens after the “daddy months”? Fathers’ involvement in paid work, childcare, and housework after taking parental leave in Germany. European Sociological Review 31(6): 738–748.

14.

Cohen

Single

(2001) An examination of the perceived impact of flexible work arrangements on professional opportunities in public accounting. Journal of Business Ethics 32(4): 317–328.

15.

Collinson

Hearn

(2005) Men and masculinities in work, organizations, and management. In: Kimmel

Hearn

Connell

(eds) Handbook of Studies on Men & Masculinities. Thousand Oaks, CA: SAGE, 289–310.

16.

Coltrane

Miller

DeHaan

(2013) Fathers and the flexibility stigma. Journal of Social Issues 69(2): 279–302.

17.

Corcoran

(1979) Work experience, labor force withdrawals and women’s wages: empirical results using the 1976 panel study of income dynamics. In: Lloyd

Andrews

Gilroy

(eds) Women in the Labor Market. New York: Columbia University Press, 216–245.

18.

De Grip

van Loo

(2002) The economics of skills obsolescence: a review. In: de Grip

van Loo

Mayhew

(eds) The Economics of Skills Obsolescence: Theoretical Innovations and Empirical Applications. Amsterdam; Boston, MA: JAI, 1–26.

19.

Destatis (2019a) Lebenssituation von Männern: Ergebnisse des Mikrozensus 2017. Wiesbaden: Federal Statistical Office of Germany.

20.

Destatis (2019b) Statistik zum Elterngeld: Leistungsbezüge – 2018. Wiesbaden: Federal Statistical Office of Germany.

21.

Duvander

A-Z

Johansson

(2012) What are the effects of reforms promoting fathers’ parental leave use? Journal of European Social Policy 22(3): 319–330.

22.

Eagly

(1987) Sex Differences in Social Behavior: A Social-Role Interpretation. Hillsdale, MI: Lawrence Erlbaum Associates.

23.

Eagly

Steffen

(1984) Gender stereotypes stem from the distribution of women and men in social roles. Journal of Personality and Social Psychology 46(4): 735–754.

24.

Edin

P-A

Gustavsson

(2008) Time out of work and skill depreciation. Industrial & Labor Relations Review 61(2): 163–180.

25.

Ekberg

Eriksson

Friebel

(2013) Parental leave: a policy evaluation of the Swedish “daddy-month” reform. Journal of Public Economics 97: 131–143.

26.

England

(1982) The failure of human capital theory to explain occupational sex segregation. The Journal of Human Resources 17: 358–370.

27.

England

Privalko

Levine

(2020) Has the gender revolution stalled? The Economic and Social Review 51: 463–488.

28.

Evertsson

(2016) Parental leave and careers: women’s and men’s wages after parental leave in Sweden. Advances in Life Course Research 29: 26–40.

29.

Gangl

Ziefle

(2009) Motherhood, labor force behavior, and women’s careers: an empirical assessment of the wage penalty for motherhood in Britain, Germany, and the United States. Demography 46(2): 341–369.

30.

Gehrke

Gulden

V-S

Ingwersen

(2018) Branchenanalyse Chemieindustrie: Der Chemiestandort Deutschland im Spannungsfeld globaler Verschiebungen von Nachfragestrukturen und Wertschöpfungsketten. Study der Hans-Böckler-Stiftung 395, Dusseldorf: Hans-Böckler-Stiftung.

31.

Gerst

Grund

(2019) Career interruptions and current remuneration. International Journal of Manpower 40(5): 850–878.

32.

Gibbons

Katz

(1991) Layoffs and lemons. Journal of Labor Economics 9(4): 351–380.

33.

Gíslason

Símonardóttir

(2018) Mothering and gender equality in Iceland: irreconcilable opposites? Social Policy and Society 17: 1–10.

34.

Golden

Altman

(2008) Why do people overwork? Over-supply of hours of labor, labor market forces and adaptive preferences. In: Burke

Cooper

(eds) Culture Cause Consequences and Choices. Bingley: Emerald, 61–83.

35.

Görlich

de Grip

(2009) Human capital depreciation during hometime. Oxford Economic Papers 61: i98–i121.

36.

Gould

(1974) Measuring masculinity by the size of a paycheck. In: Pleck

Sawyer

(eds) Men and Masculinity. Englewood Cliffs, NJ: Prentice-Hall, 96–100.

37.

Grund

(2015) Gender pay gaps among highly educated professionals: compensation components do matter. Labour Economics 34(C): 118–126.

38.

Heller

(2014) Strukturanalyse der ostdeutschen Chemieindustrie: Gutachten im Auftrag des Verbandes der Chemischen Industrie e.V. (VCI), Landesverband Nordost und des Arbeitgeberverbandes Nordostchemie e.V. Dresden: ifo Inst.

39.

Hodges

Budig

(2010) Who gets the daddy bonus? Organizational hegemonic masculinity and the impact of fatherhood earnings. Gender and Society 24: 717–745.

40.

Judiesch

Lyness

(1999) Left behind? The impact of leaves of absence on managers’ career success. Academy of Management Journal 42(6): 641–651.

41.

Konrad

Mangel

(2000) The impact of work-life programs on firm productivity. Strategic Management Journal 21(12): 1225–1237.

42.

Kunze

(2002) The timing of careers and human capital depreciation. IZA discussion Institute of Labor Economics paper 509. Bonn: IZA.

43.

Lam

Feldman

(2012) The relationship between external job mobility and salary attainment across career stages. Journal of Vocational Behavior 80(1): 129–136.

44.

Landers

Rebitzer

Taylor

(1996) Rat race redux: adverse selection in the deter-mination of work hours in law firms. The American Economic Review 86(3): 329–348.

45.

Lott

Eulgem

(2019) Lohnnachteile durch Mutterschaft: Helfen flexible Arbeitszeiten? WSI report 49. Dusseldorf: Wirtschafts- und Sozialwissenschaftliches Institut (WSI) der Hans-Böckler-Stiftung.

46.

Lott

Klenner

(2018) Are the ideal worker and ideal parent norms about to change? The acceptance of part-time and parental leave at German workplaces. Community, Work & Family 21(5): 564–580.

47.

Messner

(1997) Politics of Masculinities: Men in Movements. Thousand Oaks, CA: SAGE.

48.

Mincer

(1958) Investment in human capital and personal income distribution. Journal of Political Economy 66(4): 281–302.

49.

Mincer

Ofek

(1982) Interrupted work careers: depreciation and restoration of human capital. Journal of Human Resources 17(1): 3–24.

50.

Mincer

Polachek

(1974) Family investments in human capital: earnings of women. Journal of Political Economy 82(2): 76–108.

51.

Neuman

Weiss

(1995) On the effects of schooling vintage on experience-earnings profiles: theory and evidence. European Economic Review 39(5): 943–955.

52.

Persson

Rossin-Slater

(2019) When dad can stay home: fathers’ workplace flexibility and maternal health. IZA discussion paper series 12386. Bonn: IZA Institute of Labor Economics.

53.

Rebitzer

Taylor

(1995) Do labor markets provide enough short-hour jobs? An analysis of work hours and work incentives. Economic Inquiry 33(2): 257–273.

54.

Rege

Solli

(2013) The impact of paternity leave on fathers’ future earnings. Demography 50(6): 2255–2277.

55.

Rudman

Mescher

(2013) Penalizing men who request a family leave: is flexibility stigma a femininity stigma? Journal of Social Issues 69(2): 322–340.

56.

Rudman

Phelan

(2008) Backlash effects for disconfirming gender stereotypes in organizations. Research in Organizational Behavior 28: 61–79.

57.

Schmelzer

Kurz

Schulze

(2015) Einkommensnachteile von Müttern im Vergleich zu kinderlosen Frauen in Deutschland. Kzfss Kölner Zeitschrift für Soziologie und Sozialpsychologie 67(4): 737–762.

58.

Schneer

Reitman

(1990) Effects of employment gaps on careers of M.B.A.’s: more damaging for men than for women? Academy of Management Journal 33(2): 391–406.

59.

Schober

(2014) Parental leave and domestic work of mothers and fathers: a longitudinal study of two reforms in West Germany. Journal of Social Policy 43(2): 351–372.

60.

Spence

(1973) Job market signaling. The Quarterly Journal of Economics 87(3): 355–374.

61.

Spivey

(2005) Time off at what price? The effects of career interruptions on earnings. Industrial and Labor Relations Review 59(1): 119–140.

62.

Stafford

Sundström

(1996) Time out for childcare: signalling and earnings rebound effects for men and women. Labour 10(3): 609–629.

63.

Statistics Sweden (2018) Women and Men in Sweden 2018. Örebro: Population and Economic Welfare, Statistics Sweden.

64.

Unterhofer

Wrohlich

(2017) Fathers, parental leave and gender norms. IZA discussion paper 10712. Bonn: IZA – Institute of Labor Economics.

65.

Vishwanath

(1989) Job search, stigma effect, and escape rate from unemployment. Journal of Labor Economics 7(4): 487–502.

66.

Wharton

Chivers

Blair-Loy

(2008) Use of formal and informal work-family policies on the digital assembly line. Work and Occupations 35(3): 327–350.

67.

Williams

(2001) Unbending Gender: Why Family and Work Conflict and What to Do about It. New York: Oxford University Press.

68.

World Economic Forum (ed.) (2018) The Global Gender Gap Report 2018. Geneva: World Economic Forum.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB