While previous studies have examined how an individual’s work and career-related activities may influence productivity, little research has examined the extent to which time spent in family-related activities are relevant. These issues are important given the consequences associated with family-work spillover and family-to-work conflict experienced by employed parents (Bakker and Geurts 2004; Comer and Stites-Doe 2006; Karimi and Nouri 2009; Maume and Houston 2001). This study proposes a theoretical model of productivity that pays particular attention to family-related determinants of productivity. It also examines the impact family obligations have on salary directly and indirectly via productivity.

To accurately capture the complex influences of family roles on worker productivity and salary, comparisons between men and women are pertinent. Human capital models examine differences in men’s and women’s productivity and work rewards and suggest that those who invest in their careers through education, experience, skills, and training will be rewarded accordingly in the workplace (Hagan and Kay 1995; Noonan et al. 2005). Implications from human capital theory suggest that since women tend to allocate more time to their current or future family roles, they invest less time and effort in their careers, which affects their work experience, career progress, and ultimately their productivity. Alternatively, men allocate more time to their careers than their familial responsibilities, resulting in greater workplace success and productivity (Becker 1993). In investigating how human capital and family responsibilities affect productivity in general, and whether men’s and women’s productivity is affected differently, this study examines lawyers in law firms.

Law firm lawyers were selected for this study because unlike workers in other types of law settings, or most other occupations, they report a standard measure of work productivity that spans across virtually all North American law firms. This indicator of productivity is referred to as billable hours (Epstein et al. 1995; Hagan and Kay 1995; Pierce 1995). Billable hours are a standard measure of calculating the fees charged to clients and viewed as an “objective” measure of the work done for the client (Epstein et al. 1999; Hagan and Kay 1995). The individual lawyer maintains a time diary that records accurate time accounts of correspondence and communication, legal research, as well as meetings and court appearances. Most firms set out annual billing targets or minimum quotas for lawyers to fulfill and monthly sheets are often circulated showing how each individual’s billings compare to their colleagues. In some firms, these figures are recorded in time increments as small as 6 min (Hagan and Kay 1995; McKenzie Leiper 2006). In others, computerized programs are used that enable lawyers and firms to monitor and track the time lawyers bill to clients (Fortney 2000). Both individual lawyers and firm managers are usually aware of a particular lawyer’s ongoing billings and ranking in the firm and the hours billed are viewed as an indicator of the lawyer’s productivity and value to the firm. It is important to note that billable hours do not reflect lawyers’ total amount of work time. Rather they represent the time billed directly to clients, which reflects approximately two-thirds of the total number of hours that lawyers actually work (Fortney 2000; Hagan and Kay 1995).

Lawyers are also productive during their non-billable work hours when participating in committee work, community work, continuing education, pro bono work or client development. The time devoted to these tasks, however, is not recorded in the same standardized and detailed manner and is therefore difficult to assess in terms of productive work hours. In addition, there has been considerable debate regarding the validity of billable hours as a measure of a lawyer’s valued labor to the firm. For example, billable hours only reflect the quantity of hours billed to clients without taking into account the quality of work provided to clients and the firm and therefore do not properly constitute the entire conceptual domain of productivity (Kordana 1995). Not only do billable hours fail to capture the quality of lawyers’ work, these figures do not indicate the degree to which clients are happy or satisfied with the legal services they receive. This measure of productivity is also limited because it refers to law firm lawyers only. Those in other types of law, such as government or in-house positions, do not record billable hours. Given these considerations, we acknowledge that the validity of our measure of productivity in the form billable hours may be open to debate, but argue it offers a unique, reliable and legitimate empirical test of the age-old assumption that women are “less productive” than men.

It is important to note that lawyers’ billable hours have been shown to differ significantly for men and women in law firms, which may explain salary disparities (Hagan and Kay 1995; Robson and Wallace 2001). Yet, whether and why differences exist between men’s and women’s billable hours has not been fully researched and the effect of workers’ family obligations on productivity has been neglected in previous studies. Therefore, this paper examines productivity amongst male and female lawyers in law firms. As mentioned previously, we also examine the relationship between productivity and salary. In doing so, this study addresses the following research questions:

  1. (1)

    To what extent do women and men differ in their work and family responsibilities?

  2. (2)

    Do male and female lawyers differ in their productivity and do different work and family responsibilities affect their productivity?

  3. (3)

    To what extent does productivity affect salary and does productivity and its determinants affect men’s and women’s salaries differently?

Related Literature

Women are generally viewed as working fewer hours and less successful in their careers than men because of women’s greater commitment to family responsibilities. Previous studies have reported that female lawyers work significantly fewer hours than male lawyers and are more likely to assume part-time positions and/or take a leave of absence to fulfill their childrearing responsibilities (Epstein et al. 1995; Hagan and Kay 1995; Wallace 2004b). Women make certain career decisions and sacrifices as they attempt to juggle work and family, or women face certain barriers because employers assume they will place priority on their family responsibilities that invariably interfere with their work (Damiano-Teixeira 2006; Reskin and Roos 1990). Men are not viewed in the same way. In fact, when men become fathers, they are considered more committed to their work as they often work longer hours to fulfill the breadwinner role and provide financially for their family (Hundley 2001; Jacobs and Gerson 2001; Seron and Kerry 1995). Given that women presumably work less because of family obligations, human capital theorists argue women are consequently rewarded less compared to their male counterparts (Christie-Mizell 2006; Epstein et al. 1995; Hagan and Kay 1995).

Hypothesis 1

Men will report greater productivity and salaries compared to women.

General and Firm Specific Human Capital Investments

According to human capital theory, general human capital comprises skills and training gained from experience that may be used in multiple work settings and is generally valued by many different employers (Becker 1994). Individuals with more work experience are expected to be more skilled and productive and as a result receive higher financial returns in the workplace. The literature reports that, on average, men accumulate more work experience than women (Blau et al. 2002; Nelson and Robinson 1999; Reskin and Roos 1990). Human capital theorists argue that these disparities result in part from women’s overriding family obligations, which restrict them from gaining experience in the labor force (Becker 1985).

Unlike general human capital investments, firm specific investments are first and foremost valuable to a particular firm and have little, if any, value outside that specific work context (Becker 1994). Firm specific experience refers to the length of time that an individual has worked at a particular firm (Kay and Hagan 1998; Long et al. 1993; Noonan et al. 2005). Since the more experience lawyers have in their current law firm is expected to lead to greater productivity, lawyers with longer organizational tenure should also receive a higher salary. Again, women usually have less firm specific experience because of their overriding family obligations, which interfere with their ability to get ahead in the workplace (Kay and Brockman 2001).

The second firm specific investment included in this model is firm position. In North American law firms, most lawyers occupy one of two positions: partners, who own the firm, or associates, who are employees of the firm. Partners may be more inclined to record more billable hours than associates because of the financial benefits they receive from the firm’s profits. Therefore, partners have more invested when it comes to their own and others’ productivity because the firm’s total profits from billable hours ultimately determines their own personal salary (Hagan and Kay 1995).

Men often have an advantage over women when it comes to promotion to partnership largely because men are seen as investing more in their careers, while women are considered primarily responsible for and committed to their family (Epstein et al. 1995; Kay and Brockman 2001; Pierce 1995). These expectations frequently lead women down the “mommy track” in the workplace, characterized by decreased advancement opportunities given women’s family-related commitments (Schwartz 1989). It is hypothesized that partners will be more productive than associates in law firms because they have more invested, and furthermore, it is expected that men will be more likely to hold partner positions than women.

Hypothesis 2

Human capital investments will have a positive effect on lawyers’ productivity.

Social Capital Investments

Researchers have argued that social capital matters to economic productivity and rewards (Schuller 2001; Simmons et al. 2007). Social capital refers to the resources that accumulate from socializing and networking with others in a community (Glaeser et al. 2000; Schuller 2001). Workers are more productive as a result of investing in social capital because of the support, opportunities, and resources it provides (Coleman 1990; Kay and Hagan 1999). This study examines both client investments and collegial investments in work-related social capital that are hypothesized to enhance productivity. Investments in one’s clients include the time spent in professional social activities and working for clients. External networking with clients through professional social activities is often referred to as “rainmaking.” According to Epstein et al. (1995) rainmaking is defined as the process whereby lawyers meet and recruit potential clients. Not all associates are designated “rain makers,” which may lead to variations in lawyers’ productivity. Alternatively, collegial investments refer to the time spent on professional social activities with colleagues and the access to resources offered by colleagues.

Epstein et al. (1995) explained that female lawyers report less time spent on these activities because of their overriding familial roles. For example, many women return home to perform a “second shift” of household responsibilities after their workday (Hochschild 1989), consequently missing socializing opportunities with clients. Based upon these explanations, female lawyers are expected to invest less time in professional social activities with clients and colleagues compared to male lawyers.

Hypothesis 3

Social capital investments will have a positive effect on lawyers’ productivity.

Family Responsibilities

In addition to work-related factors that contribute to workers’ productivity, there are a number of familial influences that appear relevant. Previous literature has typically accounted for these influences by examining demographic variables that reflect one’s family status, such as marital status and/or parental status (see Friedman and Greenhaus 2000; Monna and Gauthier 2008). In contrast, the current study includes more comprehensive measures of family responsibilities that capture the actual time spent on marital relationships, time spent caring for children, and time spent on household tasks.

Married people presumably dedicate time to their partner to successfully build and maintain their marital relationship. Much of the literature relies simply on marital status as an indicator of family responsibilities, where those who are married are hypothesized to spend more time with their spouse or partner than those who are not married. The actual time individuals spend on their marital relations from day-to-day may have important implications for workplace productivity or billable hours (Epstein et al. 1995; Hagan and Kay 1995; Noonan et al. 2005), where time spent on marital relations is expected to have a negative effect on productivity (Frone 2003; Frone et al. 1997). Research in this area indicates that women dedicate more time to all familial relations, including marital relations (Frone 2003; Waite and Bachrach 2000).

The second type of family time discussed in the literature reflects time dedicated to childcare and other parental responsibilities (Monna and Gauthier 2008). According to human capital theorists, parents spend time with their children providing education, care, and supervision to ensure their child’s healthy development (Becker 1993; Mincer and Polachek 1974). Time allocated to childcare, however, may intrude upon one’s role responsibilities in the workplace (Bakker and Geurts 2004; Clark 2000). Moreover, those with younger children are expected to experience greater family-to-work conflict because of the additional time pressures associated with younger children (Comer and Stites-Doe 2006; Moen and Yu 1999; Nomaguchi et al. 2005). Additionally, normative gender expectations prescribe that women should assume more childcare roles than men, especially when children are first born and in their preschool years (see Christie-Mizell 2006; Hays 1996; Monna and Gauthier 2008). Hays (1996) elaborated this further by explaining the ideology of “intensive mothering.” This is characterized by the expectation that first, mothers will provide the primary care for their children, and second, this care will be conducted in a specific manner. The manner of childcare is considered time consuming, given that it is child-focused, emotionally and labor intensive, and ultimately selfless (Hays 1996). The ideology of intensive mothering stands in stark contrast to the independent, self-centered, profit-driven mentality of the workplace, which encourages productivity and financial success. Men, however, are encouraged to maintain the workplace mentality of the drive for success that is congruent with their provisional obligations to the family and their primary role as the breadwinner (Christiansen and Palkovitz 2001). As a result, time pressures associated with childcare are expected to impinge upon women’s productivity and subsequent salary as women invest extensive time in these responsibilities (Bakker and Geurts 2004; Firestone et al. 1999; Frone 2003; Suitor et al. 2001).

The third family time variable examined in this study is time spent on household responsibilities (e.g., cleaning, cooking, and house and yard maintenance; Coltrane 2000), which is also expected to intrude upon one’s work-role responsibilities, thereby decreasing productivity (Frone 2003; Frone et al. 1997). It is important to note that time spent on these tasks is highly gendered. Men spend considerable time in the public sphere of work, whereas women stereotypically engage more in the private sphere, managing household responsibilities, which may subsequently affect workplace productivity and rewards (Christie-Mizell 2006; Friedman and Greenhaus 2000). Moreover, estimates of household labor state that, on average, men perform about 35% of the necessary tasks compared to women (Coltrane 2000; Presser 2003) and wives spend about twice as much time performing household tasks compared to their husbands (Bianchi et al. 2006). Women are also more likely to perform female-related household tasks that require attention daily, such as cleaning, cooking, and laundry, while men are more likely to more masculinized tasks, such as yard work and auto work (Estes et al. 2007). Recent research suggests some families are able to afford childcare or household assistance to relieve the time burden associated with parental obligations and household responsibilities (Bianchi et al. 2006). Despite these resources, it is still believed that for most individuals, time allocated to family and household responsibilities impedes work responsibilities, and therefore may influence workers’ productivity.

Hypothesis 4

Family responsibilities will have a negative effect on productivity.

Family-Friendly Firms

There are certain workplace characteristics that reflect whether the employing organization is a family-friendly firm that may also affect worker productivity. Family-friendly firms offer benefits and incentives to help workers balance their potentially conflicting work and family roles (Blair-Loy and Wharton 2002; Davis and Kalleberg 2006; Thompson et al. 1999). Some individuals pursue these employment settings to better suit their family situations, which has influenced family-friendly attitudes in a growing number of workplaces (Haddock et al. 2006). Previous studies of productivity do not comprehensively account for the influences of family-friendly firms. Rather, the size of the firm, the specialization of one’s work, and differences between core and periphery firms have been examined in relation to worker productivity (Kay and Hagan 1999; Leahey 2006; Noonan et al. 2005). In contrast, this study includes more comprehensive measures of family-friendly firms in terms of the availability of alternate work arrangements, the presence of a supportive work-family culture, and the percentage of women employed in the firm.

The availability of alternate work arrangements is considered a central component of family-friendly firms. Alternate work arrangements refer to any form of work that differs from the traditional full-time schedule, such as part-time work or reduced hours (Wallace 2006). Employees with children are often granted greater access to flexible or alternate work arrangements compared to those without children (Golden 2008). Greater flexibility and discretion in one’s work hours allow parents to structure their workdays to accommodate conflicting family obligations (Golden 2008; Haddock et al. 2006; Karimi and Nouri 2009). On one hand, alternate work arrangements may increase productivity, given that work may be completed anywhere, anytime (MacEachen et al. 2007). On the other hand, it is argued that such flexibility may reduce productivity if workers choose to use alternate work arrangements to dedicate more time to family-related responsibilities. For example, VandenHeuvel (1997) found that the availability of alternate work arrangements fosters workers’ family-related absenteeism, thereby decreasing individuals’ work hours. Others have found alternate work arrangements, or ‘flex-time’ may also have detrimental consequences for workers’ well-being (Campione 2008), which may subsequently affect productivity. The benefits of alternate work arrangements for productivity in family-friendly firms are therefore unclear.

For family-friendly firms to successfully promote employees’ work-family balance, a supportive culture is necessary so that employees are not penalized for using such family-friendly benefits. This is important because, regardless of whether family-friendly benefits are offered, their use depends upon the organizational culture of the employing firm (Blair-Loy and Wharton 2002; Thompson et al. 1999; Wallace 2006). For example, employees may not use family-friendly benefits, such as alternate work arrangements, if their work culture does not support doing so. Although some research suggests these patterns are changing (Hill et al. 2004), many argue that women still face consequences for using family-friendly benefits, including decreased job mobility, income, and security (Weeden 2005). The perceived negative consequences associated with using family-friendly benefits may affect productivity because unsupportive work settings emphasize work obligations over family obligations, which should promote long work hours. Alternatively, since those in more supportive work settings likely face fewer penalties when using family-friendly benefits, they may be more inclined to take advantage of such benefits, which will likely reduce lawyers’ billable hours and subsequent salary.

A final factor discussed in the literature emphasizes the concentration of women in family-friendly firms (Blair-Loy and Wharton 2002; Wallace 2006). Since women are more likely than men to seek balance between their work and family roles, family-friendly firms with benefits such as alternate work arrangements and supportive work-family cultures tend to be more attractive to women (Blair-Loy and Wharton 2002; Greenhaus and Beutell 1985). A greater percentage of women employed in an organization may signal lower productivity because these firms likely reflect supportive work-family cultures and family-friendly benefits that prioritize family obligations over work.

Literature on lawyers supports these arguments (Epstein et al. 1995; Hagan and Kay 1995). Women are overrepresented in law firms that are more family-friendly or that offer alternate work arrangements whereas men are less likely to work in such firms (Epstein et al. 1999; Hagan and Kay 1995; Pierce 1995). Since we predict lower productivity in these firms, we further hypothesize lawyers in family-friendly firms earn less.

Hypothesis 5

Family-friendly firm characteristics will have a negative effect on lawyers’ productivity.

Salary

The final component of this analysis examines the relationship between productivity and salary. According to human capital and social capital theory, workers who are more productive are rewarded with higher salaries (Becker 1994; Schuller 2001). Studies that focus on law firm lawyers’ salaries report that the number of hours worked, particularly in terms of billable hours, significantly increases their annual salary (Epstein et al. 1995; Hagan 1990; Hagan and Kay 1995).

Hypothesis 6

Productivity will have a positive effect on lawyers’ salaries.

Table 1 suggests that when considering the positive determinants of productivity and subsequent salary, men are expected to have an advantage over women. That is, men are expected to invest more in human capital and social capital, all of which are expected to enhance their productivity and salary. Similarly, when taking into account the determinants that may reduce productivity and salary, women are expected to be disadvantaged as they are expected to allocate more time to their family instead of their careers. These expectations logically support Hypothesis 1, which predicts women will report fewer billable hours than men and therefore earn less than men.

Table 1 Predicted gender differences in investments and firm characteristics and the predicted effects on productivity
Full size table

Although men and women are anticipated to vary in both their personal investments and workplace situations, the gendered effects of these determinants on productivity and salary are less obvious. For example, even if men and women have different amounts of firm specific experience, it is unclear whether firm specific experience has a greater effect on productivity for men compared to women or vice versa. Yet the literature does not suggest that these effects are necessarily the same for the two genders either (Hagan and Kay 1995; Noonan et al. 2005). Therefore, after testing the hypothesized relationships summarized in Table 1, this study also explores whether male and female lawyers’ productivity is affected differently by these determinants. In other words, does each of the determinants have the same effect on women’s and men’s productivity and salaries?

Methods

Data and Sample

The data analyzed in this paper are from the 2000 “Juggling It All Survey,” which collected information on lawyers’ work and family experiences and attitudes. The survey was distributed to all lawyers identified as currently practicing law within the province of Alberta, Canada. Of the 5,921 lawyers contacted, 1,829 completed the survey, yielding a 31% response rate. In comparison to the provincial figures from the Law Society of Alberta, similar proportions of lawyers are represented in the survey in regards to gender, workplace, and city (refer to Wallace 2004a for a detailed breakdown of these figures). In this paper, the sample is restricted to law firm lawyers who reported at least 25 billable hours in the previous fiscal year. Therefore, the sample was reduced to 670 respondents, providing information from 445 (66%) men and 225 (34%) women.

Measures

Productivity was measured by respondents’ reported billable hours from the year prior to the survey. Salary was measured by respondents’ total annual salary from the practice of law for the 1999 tax year, before taxes and other deductions were made. The variable salary was logged since its distribution was highly skewed (skewness index = 1.67; see Loether and McTavish 1980).

Gender was dummy coded 1 for males and 0 for females. Human capital investments were measured by respondents’ general law experience, firm specific experience, workplace status, and firm position. General law experience was calculated by subtracting the year that respondents were called to the Bar from the year they started working at their current firm. Firm specific experience was measured by subtracting the year that they started working at their present firm from the year that the survey was distributed (2000). Firm position was coded 1 for partners and 0 for associates.

Social capital investments were examined through the combination of client investments and collegial investments. It should be noted that professional social activities reflect information regarding respondents’ participation in both client and collegial social activities. This was measured by the number of times per month that respondents attend professional social activities, including times “Before 8 am,” “Lunches,” “Between 8 am and 6 pm (other than lunch),” “After 6 pm during the week,” and “Weekends (day or night).” Responses were summed and recoded 1 if respondents engaged in professional activities “less than 3 times per month,” 2 for “3 to 6 times per month,” and 3 if “more than 6 times per month.” Time spent working with corporate clients was measured as a proportion compared to the amount of time lawyers spent with all of their clients in total (Hagan and Kay 1995). Access to resources was measured by four items from Thomas and Ganster’s (1995) measure of coworker support. Respondents reported the extent to which lawyers they usually talk to about the stresses of their job offer suggestions or solutions, share ideas or advice, share relevant difficulties they experienced in their job, or help them figure out how to solve a work-related problem. Responses were coded 0 for “never,” 1 for “not very often,” 2 for “often,” and 3 for “most of the time.” The four items were summed and divided by the number of items to compute a mean score, which reflects the overall amount of access to resources (α = .84).

Family responsibilities include time spent in marital relations, parental obligations, and household responsibilities. Time in marital relations was measured by how often respondents and their partner spent time alone with each other talking or sharing an activity over the past month. Responses were coded 0 for “almost never,” 1 for “several times,” 2 for “once a week,” 3 for “several times a week,” and 4 for “almost daily.” Time in parental obligations was measured by asking respondents how much time they spend with their children on days that they work. Those having no children were assigned a value of 0. Similarly, information about respondents’ time in household responsibilities was obtained by asking how much time they spend on home chores, such as cooking, cleaning, repairs, shopping, yard work, or banking on days that they work.

Family-friendly firms were assessed by three characteristics including whether the firm offers alternate work arrangements, a supportive work-family culture, and the percentage of women working there. Alternate work arrangements was coded 1 if their firm allows part-time, reduced hours, or any other alternate arrangements for lawyers and 0 if not. Supportive work-family culture was measured using three Likert items from Thompson et al. (1999) that tap whether turning down work for family-related reasons seriously hurts one’s career in their firm, lawyers are resentful when people in their firm take extended leaves to care for new or adopted children, and lawyers in their firm who participate in work-family programs (e.g., part-time work) are viewed as less serious about their careers. Participants’ answers were summed and divided by the number of items to arrive at a mean score for supportive work-family culture (α = .74). The percentage of women working in a firm was computed by dividing the number of female lawyers into the total number of lawyers reported working in their firm and converting this figure to a percentage.

Control Variables

To appropriately specify the models of productivity and salary, family-related factors, including marital status, number of years in a relationship, and the presence of preschool children, were controlled for. In addition, work motivation, size of firm, and leaves of absence were also controlled. Marital status was coded 1 for those cohabitating/common law or married and 0 for all other responses. Number of years in a relationship was measured by the number of years they had been in their current relationship. The presence of preschool children was coded 1 for those currently living with any preschool children under six years of age and 0 for all others. Work motivation is defined by whether work is central to an individual’s life (Lait and Wallace 2002) and was measured by three Likert items tapping the extent to which respondents are absorbed in their work, their work is an important part of who they are, and they are deeply committed to their work. Responses were summed and divided by three to arrive at a mean score for work motivation (α = .72). Size of firm is the total number of associates and partners working in the respondent’s immediate office. Having taken leave was coded 1 for respondents who had taken any leaves of absence since they started practicing law and 0 for respondents who had not.

Statistical Analyses

To answer the questions posed in this paper, three types of statistical analyses were used. To answer Question 1, mean difference analyses were conducted using t-tests to determine whether men and women differ significantly in their work and family responsibilities and workplace characteristics (Table 2). To answer Questions 2 and 3, ordinary least squares (OLS) multiple regression techniques were used to estimate equations for productivity and salary. This technique assumes the errors are normally distributed and computes a solution that is the best fit for the distribution. In doing so, OLS attempts to minimize the errors of the estimates, also known as the residuals. In addition, OLS techniques yield standardized coefficients that allow us to compare the magnitude of effects across various determinants within a given model (Allison 1999). Given that OLS regression techniques are commonly conducted in path analysis when OLS assumptions are met (see Allison 1999), a similar approach is used here.

Table 2 Mean differences in productivity, salary, human capital, social capital, family time investments, and family-friendly firm characteristics for male (N = 445) and female (N = 225) lawyers
Full size table

To conduct the exploratory gender-interaction analysis raised in Questions 2 and 3, gender interaction terms for the determinants of productivity and salary were introduced into the regression analyses. This allows us to test whether the effect of each determinant on lawyers’ productivity or salary is the same for women and men. For example, the gender-by-general law experience interaction term allows us to test statistically whether the impact of each year of general law experience on productivity (or salary) is significantly different for women and men. For the productivity model 18 gender cross-product interaction terms were computed (e.g., gender-by-general law experience, gender-by-firm specific experience, gender-by-firm position, etc.). These interaction terms were added to the main-effects model for productivity in five blocks representing each set of theoretical variables in the model (e.g., human capital, social capital). After each block was added and the coefficients for that model examined, the set of interaction terms was removed and the next block was added to the model. Similarly, 18 gender interaction terms were tested in blocks for the salary model, with the addition of a gender-by-productivity term that was added separately to the model.

There was only one statistically significant interaction coefficient for productivity and there were three for salary. The only significant interaction for productivity was for alternate work arrangements (β = .20, t = 2.38, p = 0.01). The main-effect models for productivity was estimated for men and women separately (results not shown) to analyze the gender-specific effects of the significant interaction. Although the coefficient was positive for men (b = 31.65, t = .74, p = .24) and negative for women (b = −84.34, t = −1.41, p = .08), access to alternate arrangements failed to have a significant effect on productivity for either men or women.Footnote 1

The three significant gender interactions for salary are: firm specific experience (β = −.22, t = −2.70, p < .001); access to resources (β = −.32, t = −2.74, p < .001); and percentage of women in firm (β = .11, t = 2.29, p = .01). Results (not shown) indicate that firm specific experience has a similar, significant, positive effect on salary for men (b = .02, t = 5.42, p < .001) and women (b = .05, t = 5.83, p < .001). In contrast, access to resources was significant for men (b = −.10, t = −2.90, p = .002), but non-significant for women (b = .04, t = .75, p = .23). Finally, the specific gender effects of the number of women in a firm was not significant for either men (b = .001, t = 1.35, p = .09) or women (b = .001, t = −1.33, p = .09). These results provide little support for a salary model contingent upon gender. As a result, the main effects models for productivity and salary were used in all subsequent analyses.

Lastly, path analysis was used to answer Questions 2 and 3 and the hypotheses posed above. Recall these questions and hypotheses also refer to the direct and indirect effects of the determinants on lawyers’ salary through productivity. Path analysis techniques incorporate multiple regression equations to identify the indirect effects of a particular variable through another identified variable. The first OLS regression analysis identified productivity as the dependent variable and estimated the direct effects of work and family responsibilities and workplace characteristics on productivity (Table 3, Eq. 1). The next equations identified salary as the dependent variable. For example, in Eq. 2, the direct effects of work and family responsibilities and workplace characteristics on salary were estimated. The indirect (Eq. 3) and total effects (Eq. 4) for all of the explanatory variables on salary via productivity were then calculated. As Pedhazur (1997) explains, each indirect effect equals the product of the direct effect for that particular variable on salary and the direct effect of productivity on salary alone. The total effects for each variable on salary equal the sum of the indirect and direct effects. The results of the path analysis are also summarized in Fig. 1.

Table 3 Path analysis results for productivity and salary for male and female lawyers (N = 670)
Full size table
Fig. 1
figure 1

Path analysis results (standardized regression coefficients) for productivity and salary for male and female lawyers that are significant at the .05 level (one-tailed test). Note: Excludes control variables

Full size image

Results

Mean Differences Results

The mean difference results are presented in Table 2 and show that men (mean = 1520 h) and women (mean = 1513 h) report similar levels of productivity, contrary to what we expected. As predicted, men (mean = $157,652 CDN unlogged; mean = 11.73 logged) report significantly higher salaries than women (mean = $101,424 CDN unlogged; mean = 11.32 logged). Men (mean = 11 years) have more firm specific experience than women (mean = 6.11 years). The results show that 64% of men are partners compared to 34% of women, which is also consistent with our predictions.

As shown in Table 2, all three social capital indicators differ significantly for women and men. Men (mean = 2.44) attend more professional social activities than women (mean = 2.14) and men spend a greater proportion of their time with corporate clients (i.e., 63% of their time) compared to women (52% of their time). Women, however, report more access to resources (mean = 2.92) than men (mean = 2.68), which is the opposite of what was hypothesized.

As predicted, women (mean = 1.28 h) spend significantly more time parenting on workdays than men (mean = 1.03 h). Women (mean = 1.77 h) also spend significantly more time on household responsibilities compared to men (mean = 1.27 h). Statistically significant differences are also found for men and women in regards to the characteristics of their workplaces. For example, 74% of women report access to alternate work arrangements compared to 66% of men, and women report significantly more women in their firm (33%) compared to men (23%). These results support the predictions made in Table 1. Contrary to what was predicted, men (mean = 3.15) work in more supportive firms compared to their female counterparts (mean = 2.77).

In regards to the control variables, 89% of men were married compared to 72% of women. Men (mean = 13.1 years) had been married longer than women (mean = 8.0 years). Further, results suggest that men and women are equally motivated in their work, but significantly more women (38%) have taken a leave from their legal careers than men (18%). Lastly, women (mean = 47 lawyers) work in significantly larger firms than men (mean = 39 lawyers).

Regression Results: Productivity

Figure 1 provides a visual representation of the statistically significant paths from the multiple regression analyses. Table 3 presents the regression results for lawyers’ productivity (Eq. 1) and lawyers’ salary (Eqs. 2, 3, and 4). After taking into account lawyers’ work and family responsibilities and firm characteristics, gender does not have a statistically significant effect on productivity. These results fail to support Hypothesis 1. According to the results in Table 3, firm specific experience is the only human capital variable that significantly increases productivity (β = .30), which offers partial support for Hypothesis 2. Also note that firm specific experience is the most important predictor of productivity included in the model. The results provide mixed support for Hypothesis 3 as lawyers’ access to resources in the firm does not impact productivity, but attending professional social activities does (β = .09). Also, working with corporate clients (β = .21) significantly increases lawyers’ billable hours.

According to the results, the time spent alone with one’s partner does not affect productivity. However, time spent with children (β = −.15) and on household chores (β = −.10) significantly reduces productivity in terms of one’s billable hours. In particular, annual productivity decreases by roughly a week’s worth of work for every additional hour spent on parental obligations (approximately 48 h) and household responsibilities (approximately 42 h). These results offer partial support for Hypothesis 4. Turning next to firm characteristics, although supportive work-family cultures (β = −.10) significantly reduce billable hours, alternate work arrangements do not. In addition, the percentage of women working in firms (β = .08) significantly increases lawyers’ billable hours positively, which contradicts Hypothesis 5.

According to the results presented in Eq. 1, most of the control variables have significant effects on productivity. The number of years an individual is married negatively affects productivity (β = −.11); however, whether they are married does not. Also, worker motivation (β = .13) and firm size (β = .22) both increase productivity. Lawyers with preschool children (β = .12) also appear to be more productive; those who have preschool children surprisingly bill 115 h more per year compared to those who do not have preschool children. Finally, those who have taken a leave from law (β = −.11) bill less hours compared to those who have not. Overall, this model accounts for 30% of the variation in productivity.

Regression Results: Salary

The direct effects of the human capital, social capital, family responsibilities, firm characteristics, and control variables on lawyers’ salaries are presented in Eq. 2 of Table 3 and visually represented in Fig. 1. After accounting for productivity, as well as the other variables in the model, gender is not significantly related to lawyers’ salary. These results fail to support Hypothesis 1, which predicts differences in men’s and women’s salaries. The results in Eq. 2 also indicate that productivity (β = .30) has a significant effect on salary; more productive lawyers earn more.

According to Eq. 2, human capital and social capital investments have the greatest direct effects on salary, whereas family responsibilities have no direct effect on salary. General law experience (β = .20), firm specific experience (β = .31), and firm position (β = .40) all significantly increase lawyers’ salaries. In fact, firm specific experience has the greatest direct and total effects on salary. Professional social activities (β = .09) and time spent with corporate clients (β = .12) also significantly increase salary. Of the variables representing family-friendly firms, only alternate work arrangements affects lawyers’ salaries (β = .06). Results indicate that the size of firm also significantly increases salary, while the other control variables have no effect. Overall, this model accounts for 68% of the variation in lawyers’ salaries.

The total effects in Eq. 4 of Table 3 represent the combined direct and indirect effects of gender and work and family situations on salary. These results show that firm specific experience (β = .40), firm position (β = .39), and productivity (β = .30) have the greatest total effect on lawyers’ salaries, while the family-related variables have relatively small effects overall.

Discussion and Conclusions

This study set out to examine work and family-related determinants of productivity, whether these determinants differed by gender, and whether the effects of these determinants on productivity differed by gender. We also assessed the effect of productivity on salary and whether it differs for men and women. Perhaps the most surprising finding of this study is that there is no significant difference between men’s and women’s productivity. This finding is particularly striking because it holds even though women spend more time on family obligations and men allocate more time to their careers (refer to Table 2). Moreover, when examining how time allocations affect workplace productivity, the findings support theoretical arguments where career investments increase productivity and family responsibilities reduce productivity. These results, combined with the findings that men spend more time on career investments and women spend more time on family investments, led us to expect that career investments would be more important in increasing men’s productivity and alternatively, family responsibilities would be more important in decreasing women’s productivity. Accordingly, we predicted men would be more productive than women; however, this was not the case. Instead, the different variables affect men’s and women’s productivity in the same ways as indicated by the lack of significant gender interactions. And, as mentioned above, we found no gender differences in productivity. As a result, the similarity in men’s and women’s overall workplace productivity becomes a subject for further discussion.

In addition, the salary model reveals several important findings. Human capital and social capital increase salary, whereas family-related investments have no effect. These results differ from the productivity model where family-related investments significantly decrease lawyers’ productivity. These findings somewhat contradict human and social capital arguments that suggest women’s greater time spent in family obligations account for diminished rewards in the workplace.

The earnings model also suggests that the effects of productivity on salary are the same for men and women. This finding indicates that equally productive men and women are paid the same in law firms, which challenges comparable worth arguments that women are valued less and therefore paid less than men in the workplace (Acker 1989). Based on these findings, not only are men and women equally productive, they are also paid comparable amounts for their productivity. Additional issues raised by this study question why family responsibilities and family-friendly firm characteristics negatively affect productivity but have no effect on lawyers’ salary. That is, even though working in a family-friendly firm results in lower billable hours, working in a family-friendly firm does not translate into lower earnings. Perhaps the culture of these firms signify that lawyers are not financially penalized for taking advantage of working fewer hours in order to achieve a better balance between their work and family responsibilities.

Overall, perhaps the most important issue raised by this study is how women can invest less in their careers and more time in family and household responsibilities and still be as productive as their male counterparts. Given these contradictory findings, the following section discusses theoretical causes and consequences of comparable productivity among men and women in law firms. In doing so, efficiency, role overload, and enrichment arguments are considered.

If in fact women are dedicating more time to family obligations than men, but recording comparable billable hours, it is possible that women are performing their jobs more efficiently. If so, women may successfully juggle family demands without neglecting their work obligations. Efficiency arguments support these conclusions and suggest that women are more efficient in performing their work tasks compared to men, which results in women being able to spend less time at work and more time with their family (Bielby 1992). Similar arguments have been used to explain female lawyers’ dedication and effort to the profession (Wallace 2004b). Kay and Hagan (1998) suggested that the standards and expectations set out for female lawyers are higher than those placed on men. These double standards require women to exert extraordinary commitment to gain comparable rewards and recognition in the firm (Kay and Hagan 1998), which many women do (Wallace 2008). Based on this perspective, future research may explore whether women are more efficient at work than men by accomplishing more tasks in the same amount of time in order to meet these exceptional expectations required by the firm.

We recognize that time spent on family obligations does not necessary equal time lost at work. Rather, women may take time away from themselves in order to successfully juggle work and family obligations. Women’s comparable productivity in law firms may be at the expense of their personal well-being. As a result of their “second shift” of parental and household responsibilities (Hochschild 1989), women may feel overworked, over-stressed, and ultimately overloaded. This explanation is based on the notion of role overload, or having insufficient time to complete all the necessary demands associated with a set of given roles. A number of consequences may result from role overload. Since multiple roles require more time, energy, and effort, they may compromise one’s physical health and psychological well-being (Voydanoff 2005, 2007). Research highlights that family-related obligations are a primary determinant of role overload (Elliott 2003). Since the results of this study show that, in comparison to men, women allocate equal effort to their work roles, but more time to their family roles, it may be that when comparing men and women with equal productivity, women experience more role overload.

Yet alternative arguments advocate that multiple roles may be enriching and energizing, rather than draining (Thompson and Bunderson 2001). Emotional investments, relationship management at work, and family role quality may foster greater job satisfaction and work-family facilitation, rather than increased conflict between different roles (Pedersen et al. 2009; Seery et al. 2008). Mothers employed outside of the home experience better health and well-being as a result of the benefits associated with work, including income and collegial support (Schnittker 2007). Furthermore, research on lawyers has shown that mothers have similar commitment and job satisfaction compared to non-mothers (Wallace 2004b). Accordingly, women with complex work and family roles may have additional energy, good health, and high work commitment, which may facilitate their ability to successfully accomplish work and family demands. Given these competing explanations, long-term effects of women’s work and family obligations on physical and mental well-being are complex and should be further explored.

Although this study has implications for the broader context of workplace inequality, the results are applicable first and foremost to lawyers in law firms, which may be considered a limitation of the study. Law firm lawyers are known for the excessively long hours they work, since they are expected to reach annual billable targets, some as high as 2400 h (Hagan and Kay 1995; Wallace 2006). For these reasons, law firm lawyers are distinct from lawyers in other settings, such as in-house counsel and government lawyers, as well as unique in comparison to other professionals.

Second, the current study uses billable hours as a measure of men’s and women’s productivity. While this measure is quite reliable in its standardized and universal use across lawyers and law firms, there are important issues of validity that should be acknowledged. For example, billable hours do not capture the quality of work lawyers provide to their clients or their firm, nor the degree to which clients or the firm are satisfied with the legal services they provide. Moreover, as noted above, lawyers are also productive during their work time that is not billed directly to clients. Nonetheless, we argue that by using billable hours as a proxy for law firm lawyers’ productivity, we are able to provide an initial empirical assessment of the widely held assumption that women are “less productive” than men. Future research should attempt to incorporate other indicators of productivity that tap not only the quantity but quality of work performed.

Third, this study was based on cross-sectional data. Longitudinal data may be important in studying productivity as it may change over the course of life. Longitudinal analyses would also help establish the causal relationship between productivity and human capital, social capital, family responsibilities, and family-friendly firms, which is important for the purposes of model specification. For example, we are unable to study the long term effects of lawyers’ social capital and therefore cannot measure whether social activities ultimately generate more clients, subsequently contributing to greater productivity.

In conclusion, there are a number of noteworthy findings that emerge from this study. As expected, men invest more in their careers, which enhances productivity, and women allocate more to their family, which diminishes productivity. Yet despite these differences, men and women are equally productive in the workplace. Furthermore, equally productive men and women receive comparable salaries. However, it remains unclear how women manage additional family obligations, while maintaining levels of productivity similar to their male counterparts. Perhaps this research signals further exploration into men’s and women’s work efficiency, as well as the long term effects of work-family obligations on women’s well being.