Do Family-Friendly Firms Perform Better? - CiteSeerX

7 downloads 479 Views 190KB Size Report
Jan 14, 2005 - Florida State University, Tallahassee, FL 32306-2180,. David A. Macpherson ...... Relations, Monash University, Melbourne. Vanderkolk, B., and ...
Do Family-Friendly Firms Perform Better? Frank Heiland∗ Department of Economics Florida State University, Tallahassee, FL 32306-2180,

David A. Macpherson Department of Economics Florida State University, Tallahassee, FL 32306-2180.

January 14, 2005

Abstract This study investigates whether offering family-friendly benefits is beneficial to the firm. Using representative data on matched Australian workplaces and employees, we test whether workplaces that offer family leave, provide or assist in child care, offer flexible work schedules, and other programs to help employees balance work and family life earn greater profits or achieve higher labor productivity than workplaces that do not provide such benefits. We find some evidence that workplaces that offer the option to switch to permanent-part-time employment and offer greater control over the working schedule are more likely to be leaders in labor productivity but are no more likely to be profitable. Overall, there is no strong evidence to support the hypothesis that family-friendly practices increase the performance of the average workplace. Our findings, based on representative workplace data, qualify the existing evidence drawn primarily from case studies and non-random samples. While firms that pioneer certain family-friendly practices may outperform their peers—as suggested by previous studies—for the average workplace this advantage has ceased to exist or never existed. ∗ Tel.:

+1-850-644-7083; fax: +1-850-644-4535. E-mail addresses: [email protected] (F. Heiland); [email protected] (D. Macpherson). We received valuable comments and suggestions from Dhaval Dave, Hugo Ben´ıtez-Silva, Tim Salmon, Mary Burke, Kislaya Prasad and session participants at the 2004 Annual Meetings of the Southern Economics Association. Yunwei Gai provided excellent research assistance.

1

1 Introduction As women pursue professional careers in greater numbers, populations age rapidly in many industrialized countries, service economies boom, and technological advances make new types of workplace arrangements possible, family-friendly programs such as flexible time, work-based child care, or parental leave are becoming more widely available at workplaces around the world.1 Firms are said to have an interest in providing such non-monetary benefits to employees as part of their compensation strategy to increase the productivity of their workforce and to reduce overall labor costs by lowering turnover and monetary compensation. For example, Lisa Benenson, Editor-in-Chief of Working Mother Magazine, who publishes the popular list of the 100 Best Companies for Working Mothers in the US, was quoted in 2000 as follows: “The number of companies on the list that have initiated flextime and other work/life initiatives has increased dramatically. Companies are realizing that the work/life needs of their employees are very real and, by instituting these programs, companies increase production and lessen attrition” (cf. SocialFunds.com, 9/7/2000).2 Estimates suggest that firms incur substantial costs from labor turnover and absenteeism, and that conflict between family and work contributes significantly to these costs.3 If productivity in a traditional work environment suffers from the family demands of their employees or if the compensation is largely monetary, then workplaces that provide (more) family-friendly benefits may outperform their peers with respect to overall workplace productivity and profitability. However, to date there is little research on the hypothesis that workplace-provided family-friendly programs benefit company performance. The existing empirical evidence on the implications for productivity and profitability is limited. While there are numerous studies that consider the effect of the introduction of a new program on performance in a single workplace (with flexible time arrangements 1 For

a discussion of the availability of family-friendly work practices in the US, see Schwartz (1994) and Glass and Estes (1997). For a similar survey on Britain, see Dex (1994) or Budd and Mumford (2003, 2004). Spearritt and Edgar (1994) provide a collection of company case studies from Australia, Western Europe, Britain and the US. 2 Other prominent sources of support for the productivity hypothesis include Vanderkolk and Young (1991). 3 Campbell (1993) estimates the average turnover costs at $10,000 in 1980 US-dollars for salaried workers in the US. Similar figures have been reported for the UK (e.g., Institute of Personnel and Development 2000) and Australia (e.g., Abbott et al. 1998).

2

the most widely-studied innovation), much less research on workplace success exists that directly considers the implications of benefit provision in a typical workplace, i.e. using representative samples of firms. This is due, at least in part, to a lack of representative firm-level data (for countries like the US) that contain measures of family-friendly benefits at the workplace. In this paper we investigate the popular hypothesis that family-friendly firms perform better than their peers. We discuss the theoretical basis of the hypothesis and examine it empirically using previously unexamined employee-employer-matched benefit and performance data from representative surveys of more than 2,000 Australian workplaces (with 20 or more employees) and 19,155 employees in 1995. To estimate the effect of family-friendly practices on relative labor productivity and profitability, we utilize the variation in the availability of family-friendly benefits and workplace performance across firms. Since 1990, enterprise collective bargaining is the dominant form of determining pay and working conditions of employees in Australia (e.g., Berg et al. 2004). Statutory regulation regarding employer provided family-friendly practices are very limited. In particular, there is no general provision of paid maternity or paternity leave.4 Since Australian workplaces enjoy relative bargaining autonomy and low regulatory influence, their decision whether or not (and to what extent) to provide family-friendly policies is more likely to be consistent with their economic goals than in an environment of strong government-sponsored or regulated work and family policies (as it exists in many Western European countries).5 Thus, if there are workplace performance gains to offering (certain) family-friendly benefits, the case of the Australian workplace is well-suited to an investigation of these effects. There are several difficulties in detecting the possible effects of family-friendly benefits on workplace performance in this data. It may be that more successful companies provide more generous 4 Following

a test case before the Conciliation and Arbitration Commission in 1979, unpaid maternal leave for one year after birth is provided to the mother if she has been employed continuously with the same employer for at least one year. This statute was amended in the early 1990s to include paternity and adoption leave (Industrial Relations Act 1988/Workplace Relations Act 1996). For certain public service employees of the Commonwealth three months of paid maternity leave is available (Maternity Leave (Commonwealth Employees) Act 1973). 5 In the US, the provision of family-related benefits is even less regulated than in Australia. The most significant program, the Family and Medical Leave Act 1993, provides certain eligible employees in companies with 50 or more workers with 12 weeks of annual unpaid job-protected leave for child birth, child care, or family and personal health needs.

3

benefits. We address reverse causality with three robustness checks: an estimation of the benefit effects using large sets of additional workplace characteristics; a test for increased rent sharing in unionized and public sector workplaces; and an estimation in subsamples of workplaces expected to be more homogeneous. Furthermore, the analysis might be hindered by multicollinearity due to correlation across types of benefits. To address this issue we conduct additional analysis on the basis of a composite benefit measure. Overall, our analysis reveals little evidence to support the performance hypothesis. In this sample of typical Australian workplaces with 20 or more employees, few family-friendly practices are associated with superior labor productivity and the majority of benefits have no effect on whether the company makes a profit. These findings stand out because they question the prevailing beliefs as well as the findings of most previous studies. The only two studies using broadly representative workplace data, ours and Gray (2002), both find only weak support for the performance hypothesis. In the next section, we review the possible motives for firms to provide such non-monetary benefits. Despite popular perception a more rigorous analysis reveals that while some labor productivity advantage may arise in more family-friendly workplaces, profit advantages for the average workplace do not necessarily follow. Section 3 discusses the data and measures used in the present study to investigate the role of family-friendly benefits on workplace productivity and profitability. The results presented in Section 4 show small productivity advantages in workplaces that offer certain benefits but no effect on profits. The final section concludes.

2 Background 2.1 Theory Compensation is the central mechanism of firms to attract and retain a productive workforce. Typically compensation is in monetary terms through wages, bonuses, and health and retirement benefits. In a broader sense compensation also includes non-monetary benefits provided to employees individually or

4

as a group. Family friendly policies in the workplace (beyond what is required by law) such as return to work programs, flexible time schedules, child care assistance, etc. are part of this type of firm-specific non-monetary compensation. Several arguments that support a positive effect of generous family-friendly benefits on workplace productivity have been put forth. (I) Efficiency wage: a more generous compensation may increase worker productivity. With better compensation the job becomes more valuable to the employee. As the cost of losing the job rise the amount of shirking declines.6 (II) Psychic costs: a work environment that addresses the needs of the many employees who are or are going to be parents with small children may reduce family related stress situations that can negatively affect the productivity of the employee (via lower performance and higher absenteeism). (III) Reciprocal gift exchange: employees who experience a work environment that addresses their (future) family needs beyond what would be expected may display more effort on the job, reduce shirking and absenteeism as a result of a greater sense of loyalty. Overall productivity at the workplace may rise as a result of individual productivity gains and lower employee turnover7 . (IV) Job matching: more productive workers may be recruited if the firm is family-friendly since the pool of applicants may be larger and more diverse which increases the chance of realizing a good workplace-employee match. Offering family-friendly benefits may also increase profits. By definition profits rise if the cost of producing one dollar of revenue falls. If providing additional benefits increases productivity (and revenue) faster than costs, profits can increase. Even if there are no productivity gains from providing family-friendly benefits profits could be higher in companies that offer more benefits. This would be the case if firms can provide a benefit that is valued by the employee in a way that the monetary component of the compensation package can be lowered by more than the cost incurred by the firm to 6 This

motive for why providing a family-friendly workplace may be beneficial for firms’ performance is familiar from the standard efficiency wage theory (e.g., Akerlof and Yellen 1986). 7 An employee may not be immediately eligible for certain family-friendly benefits which reduced the amount of job switching. In the case of maternity or paternity leave the time-to-eligibility is defined explicitly by law, for other benefits the exchange follows an implicit contract.

5

provide the benefit. It requires that wage and non-wage compensation are sufficiently substitutable for the employee (e.g., Woodbury 1983 and Brown 1980) and a cost advantage in the production of the benefit by the firm over the market (if one exists) in which the employees could otherwise purchase this benefit. For example, this could be the case for firm provision of day care. If the industry structure of the market for child care is monopolistic (which is likely to be the case in more rural regions), a large firm may choose to provide child care on site and reap the benefits by ’selling’ the service to its employees through lower wages.8 Whether a workplace can realize performance gains from providing benefits may crucially depend on several factors. First, not all family programs may result in higher labor productivity at the individual or workplace level. It may be that employees who switch to more flexible work schedules, work from home, or take family leave shirk more on the job (e.g., due to less supervision) or cause workplace productivity declines due to workflow disruptions. Secondly, and most importantly, the potential gains in profit may be only temporary if other firms that compete for the same workers offer similar benefits or adjust their monetary compensation accordingly. To see that, consider the case where households have preferences over both types of benefits and where firms maximize profits and act in a competitive labor market environment. In the hedonic equilibrium an optimal compensation bundle is found for each firm such that workers are matched optimally and no firm can obtain higher profits by trading an additional unit of non-monetary benefits for wage-benefits and households cannot improve by switching to a different employer.9 In equilibrium, no firm would choose to provide more benefits since the costs would exceed the gains (e.g., from lower turnover). The compensation bundle could differ between firms even if they compete for the same skill type of labor. Some firms would mostly employ workers that have a high valuation of the family-friendly workplace practices (those likely to utilize the costly benefits) using contracts that offer more benefits and less monetary compensation (see Deardorff and Stafford 1976, Duncan and Stafford 8 By

integrating the provision of family-friendly services like child care into the workplace the employer may avoid the double marginalization problem, i.e. a situation with lower profits and consumer welfare that arises in the absence of coordination between the employer (’retailer’) and the monopoly service provider (’manufacturer’). 9 For an early model of labor supply and demand for non-pecuniary benefits see Atrostic (1982).

6

1980).10 The key insight from this illustration is there is little reason to expect workplaces that offer more generous benefits to enjoy higher profits. The fact that firms are just beginning to provide familyfriendly benefits, however, may indicate an out-of-equilibrium scenario. If firms are just realizing that some households have a preference for non-monetary benefits or that there are productivity gains to being family-friendly, some firms may enjoy a productivity advantage and higher profits as a result of being the first to offer a benefit. Subsequently, other firms that compete for the same type of labor will either follow suit in providing the benefit or adjust their monetary compensation. Thus, profitability advantages will disappear over time. If firms in the same industry ultimately provide the same compensation package the average productivity of a worker and the overall productivity in the workplace will be the same across firms. While the productivity advantage vanishes, the level of labor productivity in each workplace may be higher than in the state without family-friendly benefits due to reduced psychic costs and increased reciprocal gift exchange. Alternatively, if competing firms target workers with different benefit valuation (but the same skill set), workplaces that provide more generous familyfriendly programs may enjoy consistently higher labor productivity than workplaces who do not due to a greater reduction in psychic costs and a larger increase in reciprocal gift exchange. In either scenario, the productivity gains come at the expense of lower overall labor costs. In the remainder of this article, we investigate the effect of benefit provision on the following two dimensions of workplace performance: whether the firm makes a profit in a given year and how the firm’s labor productivity compares to its competitors. The (relative) labor productivity measures the overall (relative) labor productivity of a workplace—including the productivity of individual workers on the job and turnover etc.—and profitability captures its overall economic performance. The preceding theoretical discussion suggests family-friendly workplaces have at best a first-mover gain that they temporarily reap from providing family-friendly benefits. Such a gain is more likely if providing benefits is associated with significant worker or firm level productivity gains, employees value benefits 10 Employers

are likely able to use demographic characteristics such as gender to infer the benefit valuation type of an applicant. Equal pay legislation—as in place in Australia—makes discrimination costly and should increase the sorting by types across firms.

7

highly and the firm has a cost advantage in providing benefits (e.g., due to economies of scales) over its competitors or there are non-competitive structures in the markets where workers can purchase the benefit (a substitute). In general we should not expect to find strong evidence that family-friendly companies are more likely to be profitable in a representative sample of workplaces since firm competing with early movers will have adjusted their compensation. However, the theoretical discussion does not rule out productivity differences—even for firms that compete for workers with the same skill. Family friendly workplaces will be more likely to experience higher overall labor productivity than their competitors if they provide the kinds of benefits that are associated with significant worker or firm level productivity gains and if employees that value family benefits highly sort themselves correctly to the firms that offer these benefits.

2.2 Existing Evidence The broader earnings literature supports the idea that better compensation of employees increases individual labor productivity (Hamermesh 1990, Kahn and Sherer 1990, Campbell 1993) and lowers turnover and hiring costs (Holzer 1990). In the existing literature on non-monetary compensation, the evidence on the labor productivity implications of providing certain characteristics of a family-friendly workplace is inconclusive. Among the best documented benefits is control of the employee regarding the work schedule (’flexitime’).11 Most studies on the effect of flexible work schedules suggest that employee attendance and performance increase under such arrangements (Shepard et al. 1996, Kim et al. 1981, Orpen 1981, Schein et al. 1977). However, Haines et al. (1999) investigate the effect of flexible work schedules on turnover and absenteeism in Canada using employee pairs with differing family responsibilities. They find no evidence that turnover rates or the use of sick leave decrease for employees with family responsibilities under flextime. Lanoie (2001) follows workers over time and finds that workers’ productivity decreased after a job sharing program was implemented. Recently, studies have begun to test whether family-friendly work practices and benefits improve 11 Flexible

scheduling may be demanded by the workplace as well (e.g., Houseman 2001 and Gramm and Schnell 2001).

8

firm’s performance using samples of firms from different industries.12 Using data from 120 employers in upstate New York, Baughman et al. (2003) find that companies who offer flexible sick leave and child care assistance experience lower turnover. Baughman et al. also find that firms were able to offer lower entry-level wages confirming results from broader studies on the substitutability of wage and non-wage benefits (Macpherson and Hirsch 1995, Waldfogel 1998, Glass 2001).13 Meyer et al. (2001) investigate companies that were named among the 100 most family-friendly companies in the US by Working Mother Magazine in 1991-95. They find a positive effect of the provision and utilization of some work-family programs on profitability. The availability of a sick leave policy and greater utilization of working from home were found to increase profits while job sharing arrangements were detrimental. No effect on profits was found for flexible time and part time arrangements. Using Fortune 500 companies, Arthur and Cook (2004) find that companies that are pioneering a family-friendly workplace practice experience increases in their equity value. Using a representative sample of British workplaces with 10 or more employees from the 1998 Workplace Employee Relations Survey (WERS98), Gray (2002) finds that the availability of financial help with child care and paternity leave increases the likelihood that a workplace outperforms the average workplace financially while eight other policies have no effect and the option of switching from full-time to part-time employment reduces that chance.14 She also finds that the availability of several family-friendly benefits is associated with a lower quit rate which supports the turnover hypothesis but only one practice (financial help with child care) is found to increase the chance that a workplace experiences above-average overall labor productivity (cf. Table 6, p. 30). There is evidence that suggests a positive effect of some family-friendly workplace initiatives on company performance (labor productivity, profitability or equity value) but—with the exception of 12 Most

of the evidence had been based on case studies or non-random samples. For a detailed survey of that literature see Dex and Scheibl (1999). 13 Using an improved identification strategy and linked employer-employee data from the 1998 Workplace Employee Relations Survey, Heywood et al. (2003) provide strong evidence for the substitutability of wages and family-friendly benefits in British workplaces. Similar evidence is provided for Canada by Gagne (2003). 14 The eight other practices are parental leave, working from home, term-time-only working, job-sharing, workplace nursery, paid time-off for child care at short-notice, flexitime, and 4 21 -day week (cf. Table 6, p. 30).

9

Gray—it is based on workplace case studies and non-random samples of firms.15 The present study attempts to test more generally for a positive relationship between the availability of various familyfriendly policies and performance. Specifically, we investigate the effects of a range of programs offered by a company—including flexible time schedules, provision of parental leave and child care—on profitability and labor productivity using a large representative sample of Australian workplaces in 1995.

3 Data and Econometric Model Our empirical analysis employs data from the 1995 Australian Workplace Industrial Relations Survey (AWIRS95).16 AWIRS95 is the second comprehensive workplace survey conducted by the Australian Commonwealth Department of Industrial Relations. The survey covers all major industries except agriculture, forestry, fishing, and defense across all states and territories (main survey). We use the matched workplace-employee sample that consists of 2001 workplaces with 20 or more employees. The instruments for family-friendly workplaces are obtained from an interview with the most senior manager and from an employee survey conducted in most workplaces. The latter returned data on 19,155 employees (out of a random sample of more than 30,000 employees) from the 1,896 survey workplaces (95%) where management gave permission to conduct the employee survey. The sample represents the work situation of more than half of all Australian employees and the employee sample is representative of the labor force in Australian workplaces of 20 or more employees (Morehead et al. 1997). 15 For

example, the Fortune 500 companies studied in Arthur and Cook and the firms on the 100 most family-friendly firms list used in Meyer et al. are far larger than the average workplace in the US. 16 The analysis employs data from the General Management Questionnaire, the Employee Relations Management Questionnaire, the Workplace Characteristics Questionnaire and the Employee Survey Questionnaire. The data are available through the Australian Social Science Data Archive: http://assda.anu.edu.au/studies/D0977.html.

10

3.1 Data Dependent Variables We consider the effect of family-friendly practices at the workplace on two measures of firm performance: profitability and relative labor productivity. While the relative labor productivity captures the performance of a workplace compared to its main competitors with regard to how quality and skill of its workforce is transformed into goods and services produced, profitability reflects the economic performance of the workplace at large. It is clear that the two aspects of performance are related since the profitability of a workplace is a function of the (labor) productivity and the costs of employing the production factors. Measures on these key performance aspects are gathered as ordered categorial variables in the manager questionnaire. The instrument for profit is based on the following three-category question: In the last financial year, did this workplace make a pre-tax profit, break even or make a loss? The instrument for labor productivity uses the question: How does the level of labor productivity here compare with your major competitors? The five-point answer scale is: ”A lot higher”, ”A little higher”, ”About the same”, ”A little lower”, and ”A lot lower.” There are 1,281 firms with non-missing profit information (64% non-missing) and 1,819 firms with non-missing relative labor productivity information (91% non-missing). The number of firms without profit information is high since the sample also contains non-commercial workplaces (see Table 1). Of the commercial workplaces, 75% reported a profit compared to 9% who broke even and 16% who reported a loss.17 Almost 50% of firms consider their labor productivity a little bit or much higher than their main competitors while the other half ranks itself as same or lower.18 17 At the time the Australian Economy was in an expansion that began in 1992. In 1995, the Australian GDP grew at a rate close to 4% and unemployment rate was 8% (down from 11% in early 1993). 18 Firms appear to be overly optimistic about their relative labor productivity but these data are still useful. Gray (2002) uses similar management-reported instruments to measure relative labor productivity and relative financial performance of workplaces in the WERS98. She cites evidence that shows that managers assess the performance of their workplaces correctly. The measure of the relative overall workplace productivity in our data (AWIRS95), while subjective (as in the WERS98), has the advantage that it is easy to interpret and readily comparable across industries. If family-friendly firms systematically overstate their performance then this kind of measurement error should lead to larger estimated benefit effects (i.e. upward bias).

11

Independent Variables Family Friendly Workplace Practices In the analysis of the effect of family-friendly practices on firm’s performance, we investigate workplace practices reported from two sources: (1) benefits made available to the majority of employees according to the manager; (2) benefits available to employees according to the employee survey. Since there is some overlap between the types of benefits measured in both surveys the two sets of measures help gauge the robustness of the estimated effects. In particular, the information provided by the employees may reflect better the actual utilization of services while the information from the manager survey may be primarily indicative of the availability of certain support programs.19 This difference is important as higher utilization implies higher costs and may induce higher productivity as argued above. Based on the manager data, we consider the following aspects of family-friendly workplace practices: availability of family leave, maternity leave, paternity leave, paid leave, and child care. From the employee data we construct the fraction of employees who can switch to permanent part-time employment, have access to maternity or paternity leave, can get child care support, have access to a phone they can use for family purposes, could work from home. To avoid reductions in sample size, we create binary indicators that flag missing information on covariates. Table 1 lists the mean and the fraction missing for the workplace practice variables and Table 2 illustrates the availability of these policies by selected workplace characteristics. We find that less than one percent of the managers report that no family leave programs are offered to most employees to care for family members. More than 80% of the employers report having an explicit annual family leave policy in place but other arrangements such as seniority leave or sick leave are also available to care for family members at around 60% of workplaces. Given the coding of the leave question it is unclear to what extend an employee can use paid leave to care for a family member but 75% of the workplaces offer unpaid leave to do so. 19 Responses

by employees who do not know whether a benefit is offered do not enter the averages reported in Table 1.

12

The manager survey does not specifically ask for the availability of unpaid maternity or paternity leave—a policy required by law as discussed above—but it does collect information on the availability of paid maternity and paternity leave. The former is offered in about one quarter of the commercial workplaces and in about one third of all workplaces. The greater availability of paid maternity leave in public workplaces reflects a special provision for female employees of the Commonwealth. Paid paternity leave programs are only about half as likely to be available as paid maternity leave programs. The employee survey does not specifically ask for access to paid maternity/paternity leave but for maternity/paternity leave in general. It suggests that on average about 70% of employees in a workplace have access to maternity or paternity leave.20 Work-based provision of child care or financial assistance with child care is uncommon in Australian workplaces; only about 10% of workplaces have some child care support practice in place. Only 4% of all workplaces have work-based child care (2% in commercial workplaces) and about 3% of employers offer child care subsidies. According to the employee survey about 4.5% of employees have access to some form of child care subsidy. The limited corporate provision of child care reflects the high level of public provision of work-related child care in Australia (cf. Spearritt and Edgar 1994 and Wooden 2002). In the multivariate analysis below we use an indicator for the presence of a child care center and a binary child/elder care assistance indicator capturing any availability of child/elder care assistance programs. The employee survey also collects information on the access of employees to permanent part time employment, a phone they can use to make family-related calls and to working from home, and whether an employee has control over the start/finish time. As shown in Table 1, about 75% of employees have access to a family phone and more than half could change to permanent part time at this workplace. Having the option to work from home is much less common (18-20%). More than 65% of employees have at least some control over their work schedule.21 20 Federal

law requires that a workplace offers a 52 week maternity/paternity leave to all regular employees. The much lower access figure reported by the employees may be the result of some employees interpreting the question as to refer to paid leave or employees being unaware of the maternity/paternity leave policies available. 21 A detailed analysis of the characteristics of employees who report having access to family-friendly practices in the AWIRS95 can be found in Gray and Tudball (2003). Heywood et al. (2003) provide estimates of these determinants in the

13

In sum, the descriptives in Table 1 confirm that practices that are more costly to the employer—such as paid leave and child care-related benefits—are less frequently available in Australian workplaces than other benefits. Table 1 and Table 2 furthermore show that non-commercial, larger, more unionized workplaces and workplaces with a greater share of female full-time employees are on average more likely to offer family-friendly benefits.22 These effects are consistent with the expectation that non-commercial and more unionized workplaces are more likely to result in additional rent-sharing between employer and employees. The theoretical discussion earlier suggests that larger workplaces should be more likely to provide benefits since they may enjoy economies of scales in the benefit provision. Finally, we also find support for the idea that if there is sorting by benefit-valuation then workplaces with a larger fraction of female employees should offer more benefits. Interestingly, Table 2 also shows that workplaces that face no or few competitors in the market of their main product or service (predominantly workplaces of commercial organizations) are typically no more likely to offer a benefit than the average (commercial) workplace. Workplaces that experience less competition in the product market may feel less pressure to adopt innovative workplace practices compared to firms in more competitive industries since their workforce has fewer options of alternative employment (all else constant) provided that workers have industry-specific human capital.

Other Determinants In addition to the workplace practices that are designed to appeal directly to family-oriented employees, we also account for broader workplace conditions in the analysis. We constructed controls for the existence of a written occupational health and safety policy (more than 80%), whether employee counseling/assistance is provided (37-46%), and whether the workplace is an equal opportunity/affirmative action employer (67-75%). To investigate the robustness of the estimated benefit effect we also estimate models accounting for WERS98 using an improved identification strategy. 22 Working from home and flexible work schedule is less likely to be available to employees in all-union-workplaces compared to no-union-workplaces. This ’exception’ has been observed in the WERS98 as well (see Budd and Mumford 2004). It results from the fact that the all-union workplaces contain a greater share of workplaces that have a production flow that imposes strict work schedules (e.g., manufacturing).

14

a variety of factors that may independently affect firm’s performance and that are potentially correlated with the availability of family-friendly benefits. To that end we constructed controls for company’s workforce and its demographics (including fraction of full-time and part-time female employees), the fraction of employees that are unionized, industry type of workplace based on main activity, a set of human resource management practices (including performance pay for non-managerial employees, prevalence of individual contracts, quality of management-employee relations, management focus on quality improvement and human resources), time in business, ownership and organizational status, share of labor costs in total costs, and characteristics of its main product market. We also employ a measure of seniority of the manager who was interviewed. Summary descriptives for these covariates are presented in Table 1.

Identification Strategy In the multivariate analysis below, we estimate the effect of a set of family-friendly practices on the probability of having higher labor productivity than their main competitors and on the probability of being profitable using four specifications: (1) specification with family-friendly workplace practices and welfare, equal opportunity and safety measures only; (2) in addition to (1) also controlling for industry dummies based on main activity of the workplace; (3) in addition to (1) also including basic controls regarding operation, human resource management practices, and workforce and (4) in addition to (1) and (3) also including detailed controls regarding operation, workforce, ownership, and demand. In the first specification, the estimated benefit coefficients capture the net effects of benefits on (relative) productivity and profit; the other specifications estimate the effects of work practices on performance holding other determinants of labor productivity and profits that may be correlated with the benefits constant. A concern in the analysis of the effect of family-friendly benefits on workplace performance is the possibility of reverse causality.23,24 It may be that more successful companies provide more generous 23 This

is of greater concern for the profit measure since it reflects profits from the last financial year. alternative identification strategy using exogeneous variation in benefit provision is not pursued here due to a lack of data on variation in benefit (regulation) over time and across regions. 24 An

15

compensation packages. We attempt to address this endogeneity concern in two ways.25 First, we assess the robustness of the estimated effects of benefits on firm’s performance with regard to a large set of firm and industry characteristics. As more characteristics are controlled for the role of omitted factors may be diminished. Also, reverse causality should be a greater concern in workplaces that are part of public organizations and in companies where unions are very influential since more rent sharing can be expected there. Hence we investigate the estimated effects in models with interaction terms between benefits and union coverage (indicator of ’no-union-workplace’) and private versus public organization (private enterprise indicator). Finally, we summarize estimates of the benefit effects obtained from using more homogeneous samples to reduce the potential reverse-causality bias in the results section below. Another important issue is multicollinearity in the estimated benefit effects as a result of employing correlated practices. This problem concerns primarily the set of leave policies that can be used for family leave. We report results from an additional analysis that uses a measure of the number of such leave programs in the results section below. We also discuss results from interacting the benefit measures with firm size (number of employees), fraction of female full-time employees, and lack of competition in the firm’s product market (no or few competitors indicator). A greater share of female employees may be associated with higher labor productivity as discussed above. Providing benefits with large fixed cost components (such as child care on site) should be less costly and hence less detrimental for profits in larger workplaces. Lack of product competition may increase the likelihood that workplaces that offer a benefit are innovators in their industry and should hence increase performance. 25 We

note that if there is positive reverse causality—firms that inherently perform better are more likely to offer familyfriendly benefits—then the estimated benefit effects should be biased upwards, i.e. represent upper bounds to the true effects of benefits on performance.

16

3.2 Econometric Model As shown in Tables 3 and 4, an effect of providing family-friendly benefits on profitability or relative productivity is not visible. Given the potential correlation between the various firms’ characteristics, multivariate analysis is necessary to sort out the separate effects of the provision of family-friendly work practices on firm performance. To that end we estimate the effect of the family-friendly benefits on firms’ profitability and relative productivity using Probit models.26 The estimates capture the net effect of the practices on the likelihood of having higher labor productivity than the main competitors and on the likelihood of making a profit, using the variation in benefits offered and performance between workplaces and controlling for a large number of other firm-specific determinants of success. Denoting the firm’s relative performance measured on a continuous positive scale by y∗ , we seek to estimate the coefficient vector β in y∗ = β0 X + ε,

(1)

where X contains the matrix of independent factors and ε captures unmeasured firm-specific characteristics that affect the probability of successful performance. It is assumed to be independent of the covariates, X. We observe, (2) y = 0 if y∗ ≤ 0, = 1 if 0 < y∗ . Assuming that ε is normally distributed with mean zero and unity variance, we obtain the familiar Probit model:27 (3) 26 Given

the ordered categorial nature of our dependent variables, we also considered the Ordered Probit framework. Since the results were qualitatively similar to the standard binary Probit estimates and the representation of the marginal effects is more easily accessible in the latter case we only present the Probit results here. The additional results are available from the authors upon request. 27 The variance is not identified and hence normalized to unity.

17

Pr(y = 0) = 1 − Φ(β0 X), Pr(y = 1) = Φ(β0 X), where Φ is the cumulative normal density function. In the next section we employ the Probit framework to test for the effects of available benefits—as reported by the manager and the employees—on the chances of having higher labor productivity than the competitors and on the chances of reporting a profit. The estimated effects capture the net effect of such family-friendly workplace practices. The effects are identified from the variation in benefits offered and performance between firms controlling for a large number of other determinants.

4 Results We estimate profit and relative labor productivity models. For each performance outcome we present the effects of family-friendly practices using different multivariate specifications. Tables 5-6 present the effects for practices as reported by the manager while Tables 7-8 show the effects when the familyfriendly benefit measures from the employee survey are used. For easy of interpretability of the magnitude of the effects, we report the effects of the workplace characteristics as marginal probability effects holding other covariates at their sample means. We also discuss results from interacting key familyfriendly workplace practices with firm size, an indicator for a workplace with no unionized employees, whether the workplace is commercial, the fraction of female full-time employees, and an indicator for workplaces that face no or few competitors in their product/service market.28 The empirical fit of the models is modest but the covariates are jointly significant across specification. The specifications with detailed industry identifiers and with basic and product market and ownership covariates reach (pseudo) R2 s of up to 11% (8%) for the profit (relative labor productivity) model.29 We observe that the coefficients of most benefits tend to be stable in magnitude across 28 The

interaction terms enter the equation in addition to the individual covariates that are included in the set of basic controls. The detailed results of the models with interactions are available from the authors upon request. 29 The pseudo coefficient of determination is based on the ratio of the log-likelihoods of the probit model shown and a constant-only probit model.

18

specifications in the profit and labor productivity model. In the presence of bias from positive reverse causality, in contrast, we would expect larger (positive) effects in the specification without control variables.

4.1 Relative Labor Productivity The results using the measures from the manager interview do not provide evidence that more familyfriendly work practices lead to higher labor productivity. As shown in Table 5, workplaces that offer career leave to employees to care for family members (8% of all workplaces) have higher labor productivity. The likelihood that a workplace outperforms the competition in terms of labor productivity increases by 9 percentage points if career leave is offered. For all other types of leave arrangements to care for family members there is no statistically significant positive effect. Moreover there is no evidence that paid maternity or paid paternity leave lead to higher labor productivity. In fact the effect of paid maternity leave on productivity is negative (paid maternity leave appears to reduce the chance of being a productivity leader by up to 9 percentage points) but it is only statistically significant in specifications with additional controls. Analysis with interaction terms shows that paid leave is particularly detrimental for relative labor productivity in public-sector workplaces and somewhat less detrimental in workplaces that have a larger fraction of female employees. There is no evidence that workplaces with a child care center or other child/elder care programs have higher labor productivity than their main competitors. As shown in Table 7, the measures constructed from the employee survey do not provide strong support for the hypothesis that workplaces that provide more generous family-friendly benefits than others are relatively more productive. Access to two out of the six family-friendly practices has a positive effect on productivity in models with no or few controls, while the other four have no statistically significant effect. Having the option of changing permanently to part time employment (available to 55% of employees on average) is associated with a greater chance that the own labor productivity exceeds the one of competitors. Analysis with interaction shows that this effect is particularly large in

19

private-sector workplaces. An increase in employee access to this benefit by 10 percentage points increases the probability that the workplace outperforms its main competitors by about 1% according to the model without controls and with industry dummies (columns (1) and (2) in Table 7). Control over the work schedule (available to 65% of employees on average) is associated with higher relative labor productivity: the marginal effects range from 0.6 to 1.3 for a ten percentage point increases in the share of employees with access (cf. Table 7). Controlling for interactions we find that flexibility of the work schedule is particularly beneficial for labor productivity in workplaces with a higher share of female employees and in public-sector workplaces. However, the positive effects of the part time option and the flexibility in the work schedule are not robust to the inclusion of additional control variables.

4.2 Profit The results in Table 6 show that career leave programs are associated with significantly higher profitability. Workplaces that offer career leave to employees to care for family members are about 10% more likely to report a profit. Analysis of this effect with interactions shows that the magnitude of the effect of providing career leave on profits is smaller in private sector firms than in government for profit and other commercial companies. Other leave programs do not have significant effects on profits. In particular, annual leave, by law the standard leave policy, has no effect. On the other hand, paid maternity leave (offered at 24% of the commercial workplaces) has a negative effect on the chance of being profitable. The size of this effect is comparable to the effect of career leave programs. In workplaces of private sector organizations paid leave provisions are less detrimental than in public organization which is consistent with the idea that public organizations are more likely to share rents with employees.30 Finally, while there is no evidence that having a child care center or providing other forms of child/elder care assistance is beneficial for profitability. To the contrary, workplaces that offer workbased child care have a more than 20 percentage point smaller chance of reporting a profit. Analysis with interaction terms shows that offering on-site child care is less detrimental in workplaces with a 30 Interestingly

we do not find the same pattern by degree of unionization of the workplace.

20

larger workforce or less competition in the product market. Access to most benefits measured in the employee survey does not raise profitability as shown in Table 8. A higher fraction of employees who report access to maternity/paternity leave or workingfrom-home is associated with a greater chance of having a profit but the effects are not statistically significant. Access to flexible work schedule, to subsidized child care, and to a phone that can be used for family purposes reduce the chance of being profitable while this effect is only statistically significant for child care subsidies. A five percentage point increase in access to subsidized child care (available to 4.4% of employees on average) can lower the chance of being profitable by up to 1%. Analysis of models with interaction terms shows that provision of child care assistance is less detrimental for profits in larger companies. Also, flexible work schedules are less detrimental in larger firms. This is consistent with the idea that larger workplaces can provide certain benefits at a lower cost due to economies of scales. On the other hand, there is some evidence that access to a family phone reduces profits more in workplaces that are larger.

4.3 Summary of Additional Analysis We conducted additional analysis of the role of family-friendly benefits in workplace performance.31 We investigated the robustness of the benefit effects in the manager survey data. Instead of using the set of leave policies available for family care we defined a single variable that measures the total number of leave programs offered to care for a family member. Using this broader category that helps to reduce potential misclassification error and collinearity, we did not find any evidence that a more generous leave policy increases relative labor productivity or profits. We also estimated the workplace performance models using more homogeneous samples. Specifically, we used separate samples of the private sector workplaces, government sector workplaces, service industry workplaces, non-service industry workplaces, workplaces with above-average number of older employees, above-average number of female full-time employees, and an above-average share of labor 31 These

additional results are available from the authors upon request.

21

costs in total costs. Estimates from these more homogeneous samples are less likely to be subject to the reverse causality problem discussed above and allow us to detect if the benefit effects differ across types of workplaces. For example, in workplaces of the service industry labor turnover may be particularly important. Workplaces that employ more female employees may be more responsive to the provision of family benefits while that may be less the case in workplaces that employ mostly older worker. The results in the analysis of the subsamples were mostly consistent with the earlier findings of a positive effect of career leave on performance, negative effects of paid maternity leave, on-site child care, and subsidized child care (employee survey) on profitability, and some indications of a positive effect of the permanent-part-time option and some work-schedule flexibility on relative labor productivity. Consistent with our expectations, the labor productivity effects tended to be somewhat larger in the sample of service workplaces and smaller in the sample of workplaces with above-average number of older employees. Finally, in addition to the analysis of the benefit effects on relative labor productivity reported above, we studied the effects of benefits on two important dimension of absolute level of workplace labor productivity: labor turnover and employee absenteeism. We found that flexibility with respect to the work schedule and on-site child care reduce absenteeism but there was little evidence that family-friendly benefits lower worker turnover. While there are other important channels through which family-friendly workplace practices are expected to affect labor productivity that are not captured in this analysis (e.g., on-the-job productivity of a worker), this evidence suggests that the productivity gains from implementing family-friendly practices may be limited even in absolute terms.

5 Discussion and Conclusion This paper provides one of the first comprehensive empirical investigations of the effect of familyfriendly practices on relative labor productivity and profitability of average firms using representative matched workplace-employee data from Australia. The analysis provides some evidence that having the option to change to permanent part time em22

ployment and having (some) control over the work schedule—both policies that may be particularly important to female employees who have or plan to have children—are associated with higher labor productivity but have no affect or even a negative effect on profits.32 This suggests that while the average workplace may be able to realize some productivity gains from offering more flexible forms of employment there are also significant costs associated with these programs. We also find that the availability of a special leave programs to care for family such as career leave is associated with a higher chance of being a leader in labor productivity and a greater probability of reporting a profit while paid maternity leave is associated with lower productivity and profits. This suggest that programs that help to address a specific family leave need by an employee can be adopted in a way that is not detrimental to the firm. On the other hand, the results reject the idea that paid parental leave policies may “pay for themselves.” Instead, programs that are costly in terms of working time and salary—such as paid leave—are likely to have no effect or perhaps even a negative effect on productivity and profits. Overall, the findings do not provide strong support for the hypothesis that (more) family-friendly practices increase the performance of the workplace. Few family-friendly practices in Australian workplaces are associated with superior labor productivity, most have no effect on the chance of making a profit, and some even appear to depress profits (paid maternity leave, child care center). The finding of some labor productivity advantage but essentially no profit gain of providing family-friendly benefits in the average workplace is consistent with the results from the study of British workplaces by Gray (2002). These findings from representative firm-level data-sets qualify the earlier conclusions based on evidence from case studies and non-random samples (e.g., Arthur and Cook 2004, Meyer et al. 2001): While there may be a first-mover performance advantage in terms of productivity and profits for workplaces that pioneer certain family-friendly practices—as suggested in that literature, for the average family-friendly workplace that profit advantage has ceased to exist or never existed. The rationale for the absence of any profit advantage and the lack of a strong labor productivity advantages is that 32 The

detrimental effect of generous benefits on profitability may reflect the firm’s inability to make such benefits available while offering significantly lower monetary compensation to demographic groups that value benefits highly. Statutory equal opportunity employment regulations and existing collective bargaining agreements contribute to this effect (e.g., Gruber 1994).

23

workplaces that compete for the same skill-type of labor respond by either providing similar benefits or by attracting employees that value family-friendly benefits less. The average workplace that offers compensation packages that attract employees who value family-friendly benefits highly may enjoy some lasting overall labor productivity gains as a result of lower turnover and/or higher on-the-job productivity but it also incurs higher costs as these programs are implemented, maintained and utilized.

24

References Abbott, J., De Cieri, H., and R. Iverson (1998), ”Costing Turnover: Implications of work/Family Conflict at the Management Level,” Asia Pacific Journal of Human Resources, 36(1), 25-43. Akerlof, G.A., and J.L. Yellen, eds. (1986), Efficiency Wage Models of the Labor Market. New York: Cambridge University Press. Arthur, M.M., and A. Cook (2004), ”Taking Stock of Work-Family Initiatives: How Announcements of ”Family-Friendly” Human Resource Decisions Affect Shareholder Value,” Industrial and Labor Relations Review, 57(4), 599-613. Atrostic, B.K. (1982), ”The Demand for Leisure and Nonpecuniary Job Characteristics,” American Economic Review, 72(3), 428-440. Baughman, R., DiNardi, D., and D. Holtz-Eakin (2003), ”Productivity and Wage Effects of ”FamilyFriendly” Fringe Benefits,” International Journal of Manpower, 24(3), 247-259. Berg, P., Appelbaum, E., Bailey, T., and A.L. Kalleberg (2004), ”Contesting Time: International Comparisons of Employee Control of Working Time,” Industrial and Labor Relations Review, 57(3), 531-549. Brown, C. (1980), ”Equalizing Differences in the Labor Market,” Quarterly Journal of Economics, 94(1), 113-134. Budd, J.W., and K. Mumford (2003), ”Family-Friendly Work Practices in Britain: Availability and Effective Coverage,” Department of Economics and Related Studies, University of York, Discussion Paper 02/01. Budd, J.W., and K. Mumford (2004), ”Trade Unions and family-friendly Policies in Britain,” Industrial and Labor Relations Review, 57(2), 204-222. Campbell, Carl M. III. (1993), ”Do Firms pay Efficiency Wages? Evidence with Data at the Firm Level,” Journal of Labor Economics, 11, 442-470. Deardorff, A.V., and F.P. Stafford (1976), ”Compensation of Cooperating Factors,” Econometrica, 44(4), 671-684. Dex, S., and F. Scheibl (2001), ”Flexible and family-friendly Working Arrangements in UK-Based SMEs: Business Cases,” British Journal of Industrial Relations, 39(3), 411-431. Dex, S., and F. Scheibl (1999), ”Business Performance and family-friendly policies,” Journal of General Management, 24(4), 22-37. Duncan, G., and F. Stafford (1980), ”Do Union Members Receive Compensation Wage Differentials?” American Economic Review, 70(3), 355-371. Even, W.E., and D.A. Macpherson (1996), ”Employer Size and Labor Turnover: The Role of Pensions,” Industrial and Labor Relations Review, 49(4), 707-728. 25

Glass, J., and S.B. Estes (1997), ”The family responsive workplace,” Annual Review of Sociology, 23, 289-311. Glass, J. (2001), ”Blessing or Curse: Family-Responsive Policies and Mother’s Wage Growth Over Time,” Paper presented in the Seminar Series of the Center for Policy Research, Syracuse University, October 11, 2001. Gramm, C., and J. Schnell (2001), ”The Use of Flexible Staffing Arrangements in Core Production Jobs,” Industrial and Labor Relations Review, 55(2), 245-258. Gray, H. (2002), ”Family-Friendly Working: What a Performance! An Analysis of the Relationship Between the Availability of family-friendly Policies and Establishment Performance,” Centre for Economic Performance, London School of Economics, CEP Discussion Paper 0529. Gray, M., and J. Tudball (2003), ”Family-Friendly Work Practices: Differences within and between Workplaces,” The Journal of Industrial Relations, 45(3), 269-291. Gruber, J. (1994), ”The Incidence of Mandated Maternity Benefits,” American Economic Review, 84(3), 622-641. Haines, V.Y., III, Guerin, G., and S. St-Onge (1999), ”Les effets de l’horaire flexible sur le taux de roulement et l’absent´eisme d’employ´es ayant des responsabilit´es parentales (Effects of Flexible Hours on Turnover Rates and Absenteeism of Workers with Children),” Canadian Journal of Administrative Sciences, 16(4), 232-233. Hamermesh, D.S. (1990), ”Shirking or Productive Schmoozing: Wages and the Allocation of Time at Work,” Industrial and Labor Relations Review, 3, Special Issue: Do Compensation Policies Matter?, 121S-133S. Heywood, J.S., Siebert, W.S., and X. Wei (2003), ”The Implicit Market for Family Friendly Work Practice,” Hong Kong Institute of Economics and Business Strategy, University of Hong Kong, Working Paper 1053. Holzer, H.J. (1990), ”Wages, Employer Costs, and Employee Performance in the Firm”, Industrial and Labor Relations Review, 3, Special Issue: Do Compensation Policies Matter?, 147S-164S. Houseman, S.N. (2001), ”Why Employers use Flexible Staffing Arrangements: Evidence from an Establishment Survey”, Industrial and Labor Relations Review, 55(1), 149-170. Institute of Personnel and Development (2000), ”Staff Turnover Costs Rise by 10.5 per cent,” People Management, 26 October 2000, 8. Kahn, L., and P. Sherer (1990), ”Contingent Pay and Managerial Performance,” Industrial and Labor Relations Review, 3, Special Issue: Do Compensation Policies Matter?, 107S-120S. Kim, J.S., and A.F. Campagna (1981), ”Effects of Flexitime on Employee Attendance and Performance: A Field Experiment,” The Academy of Management Journal, 24(4), 729-741.

26

Lanoie, P., Raymond, F., and B. Shearer (2001), ”Work Sharing and Productivity: Evidence from Firm Level Data,” Applied Economics, 33(9), 1213-20. Macpherson, D.A., and B.T. Hirsch (1995), ”Wage and Gender Composition: Why do women’s jobs pay less?” Journal of Labor Economics, 13(3), 426-471. Meyer, C.S., Mukerjee, S., and A. Sestero (2001), ”Work-family benefits: Which ones maximize profits?,” Journal of Managerial Issues, 13(1), 28-44. Morehead, A., Steele, M., Alexander, M., Stephen, K., and L. Duffin (1997), Changes at Work: The 1995 Australian Workplace Industrial Relations Survey. South Melbourne: Longman. Orpen, C. (1981), ”Effects of flexible working hours on employee satisfaction and performance: a field experiment,” Journal of Applied Psychology, 66(1), 113-115. Schein, V.E., Mauer, E.H., and J.F. Novak (1977), ”Impact of flexible working hours on productivity,” Journal of Applied Psychology, 62(4), 463-465. Shepard, E.M., Clifton, T.J., and D. Kruse (1996), ”Flexible Work Hours and Productivity: Some Evidence from the Pharmaceutical Industry,” Industrial Relations, 35(1), 123-139. SocialFunds.com (2000), ”Family-Friendly Companies Reap Economic Rewards,” Article at SocialFunds.com from 9/7/2000, http://www.socialfunds.com/news/article.cgi/360.html. Spearritt, K., and D. Edgar (1994), ”The Family-Friendly Front,” National Key Centre in Industrial Relations, Monash University, Melbourne. Vanderkolk, B., and A. Young (1991), The Work and Family Revolution: How Companies Can Keep Employees Happy and Business Profitable. New York: Facts on File. Waldfogel, J. (1998), ”The Family Gap for Young Women in the United States and Britain: Can Maternity Leave Make A Difference?,” Journal of Labor Economics, 16(3), 505-545. Wooden, M. (2002), ”Childcare Policy: An Introduction and Overview,” Australian Economic Review, 35(2), 173-179. Woodbury, S. A. (1983), ”Substitution between wage and non-wage benefits,” American Economic Review, 73(1), 166-182.

27

Table 1: Summary of Workplace Samples Variable

Definition

For Profit (N=1,281) Mean (Valid)

Profit Profit Break Even Loss

Workplace made a profit in the last year Workplace broke even in the last year Workplace had a loss in the last year

0.746 0.092 0.162

(1.00) (1.00) (1.00)

Labor Productivity Much Higher A Little Higher Same A Little Lower Much Lower

Labor product. Labor product. Labor product. Labor product. Labor product.

0.145 0.352 0.368 0.114 0.021

(1.00) (1.00) (1.00) (1.00) (1.00)

0.145 0.345 0.389 0.101 0.021

(1.00) (1.00) (1.00) (1.00) (1.00)

Workplace Practices Manager Survey Family Leave Career Leave Special Leave Unpaid Leave Flex. Leave Sick Leave Annual Leave Seniority Leave Other Leave No Leave Paid Maternity Leave Paid Paternity Leave Child Care Center Care Assistance Care Subsidy Holiday Care Other Child Care Elder Care No Child/Elder Care Welfare Equal Opportunity Safety

Family leave avail. to look after family Career leave avail. to look after family Special leave avail. to look after family Unpaid leave avail. to look after family Flex leave avail. to look after family Sick leave avail. to look after family Annual leave avail. to look after family Long serving leave avail. to look after family Other leave avail. to look after family No leave avail. to look after family Paid maternity leave offered Paid maternity leave offered Work-based child care center Financial assistance for child care Subsidy/reserved place in child care center Holiday care program Other child care assistance provided Elder care assistance provided No child/elder care provided welfare, employee assistance/counselling equal opportunity/affirmative action employer Written occupational health & safety policy

0.236 0.061 0.437 0.744 0.391 0.616 0.864 0.606 0.055 0.008 0.241 0.118 0.020 0.013 0.026 0.017 0.022 0.003 0.919 0.365 0.667 0.821

(1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (0.98) (0.98) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (0.98)

0.285 0.083 0.488 0.746 0.426 0.585 0.840 0.656 0.063 0.007 0.342 0.174 0.040 0.013 0.033 0.027 0.023 0.003 0.894 0.456 0.752 0.861

(1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (0.98) (0.97) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (0.98)

Employee Survey Not Surveyed Frac. Part Time Frac. M/P Leave Frac. Child Care Frac. Family Phone Frac. Home-Work Frac. Flex Schedule

No employees surveyed in this workplace Fraction with access to permanent part time Fraction with access to maternity/paternity leave Fraction with access to subsidized child care Fraction with access to phone for family purposes Fraction that could work from home Fraction with control over schedule (start/finish)

0.097 0.535 0.692 0.044 0.755 0.183 0.654

(1.00) (0.88) (0.86) (0.87) (0.90) (0.91) (0.90)

0.086 0.554 0.742 0.046 0.740 0.199 0.655

(1.00) (0.91) (0.88) (0.88) (0.91) (0.91) (0.91)

Other Characteristics PSE GBE NCSA FPSD SPSD ONC OC Commercial

Private Sector Organization Govt. Business Enterprise/Comm. Stat. Auth. Non-commercial Statutory Authority Federal Public Service Department State Public Service Department Other Non-commercial Other Commercial Commercial Workplace (=PSE+GBE+OC)

0.856 0.123 0.000 0.000 0.000 0.000 0.021 1.000

(1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00)

0.605 0.098 0.061 0.019 0.132 0.071 0.014 0.716

(1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00)

much higher than main competitors a little higher than main competitors same as main competitors a little lower than main competitors much lower than main competitors

All Firms (N=1,819) Mean (Valid)

Table 1 (continued). Summary of Workplace Samples Variable

Definition

For Profit (N=1,281) Mean (Valid)

All Firms (N=1,819) Mean (Valid)

Other Characteristics Employees Fixed Term Workers 0% under 21 1-10% under 21 11-25% under 21 26-50% under 21 50+% under 21 0% over 50 1-10% over 50 11-25% over 50 26-50% over 50 50+% over 50 Frac. Female PT Frac. Female FT Union Members All Union No Union Employer Assoc. Annual Evaluation Performance Pay Frac. Perf. Paya Individ. Contracta Quality Focusb Human Resourcesb Relationsb Operates < 2 Yrs. Operates 2-5 Yrs. Operates 5-10 Yrs. Operates 10-20 Yrs. Operates 20-50 Yrs. Operates 50+ Yrs. Domestic 100% Domestic 51+% Domestic 50% Foreign 51+% Foreign 100% Admin. only Head Office Product domestic Prod. mostly dom. Prod. mostly export Prod. not appl. Many Competitors Few Competitors No Competitors Demand expanding Demand stable Demand contracting Labor Costs < 20% Labor Costs 21-40% Labor Costs 41-60% Labor Costs 60-80% Labor Costs 80+%

Number of employees on payroll Number of employees with fixed term contracts 0% of employees under 21 years of age 1-10% of employees under 21 years of age 11-25% of employees under 21 years of age 26-50% of employees under 21 years of age 50+% of employees under 21 years of age 0% of employees over 50 years of age 1-10% of employees over 50 years of age 11-25% of employees over 50 years of age 26-50% of employees over 50 years of age 50+% of employees over 50 years of age Share part time female employees of total part time Share full time female employees of total full time Number of unionized employees All employees unionized No unionized employees Workplace is member of employer association Any non-managerial employees evaluated annually Any non-managerial empl. receive performance pay Share of non-managerial empl. w/ performance pay Share of non-managerial empl. w/ individual contracts Quality improvement dominates labor cost reduction Considerable resources towards human resources Relationship between employees and management Main activity undertaken less than 2 years Main activity undertaken 2-5 years Main activity undertaken 5-10 years Main activity undertaken 10-20 years Main activity undertaken 20-50 years Main activity undertaken 50+ years Wholly Australian owned Predom. Australian owned (51% or more) Equally 50% Australian and 50% foreign owned Predom. foreign owned (51% or more) Wholly foreign owned Administrative Function only Workplace is head office in Australia Product Market is domestic only Product Market is domestic, some export Product Market is primarily export Not applicable, admin. office Many competitors in product market Few competitors in product market No competitors in product market Demand for product is expanding Demand for product is stable Demand for product is contracting Labor cost as % of total costs is 80%

163.7 4.7 0.151 0.619 0.115 0.078 0.038 0.076 0.593 0.242 0.080 0.009 0.826 0.330 87.1 0.043 0.280 0.745 0.688 0.462 1.858 1.853 2.012 2.086 1.738 0.009 0.038 0.114 0.217 0.387 0.236 0.669 0.127 0.016 0.054 0.135 0.053 0.452 0.635 0.294 0.062 0.010 0.669 0.299 0.032 0.510 0.355 0.134 0.167 0.365 0.265 0.152 0.051

201.1 10.1 0.213 0.610 0.089 0.060 0.029 0.066 0.581 0.255 0.090 0.008 0.841 0.370 101.3 0.076 0.219 0.636 0.707 0.359 1.410 1.713 2.000 2.068 1.760 0.008 0.037 0.097 0.184 0.370 0.304 0.757 0.092 0.012 0.041 0.098 0.087 0.400 0.647 0.283 0.061 0.009 0.667 0.300 0.033 0.535 0.349 0.116 0.124 0.289 0.226 0.224 0.137

(1.00) (1.00) (0.99) (0.99) (0.99) (0.99) (0.99) (1.00) (1.00) (1.00) (1.00) (1.00) (0.53) (0.92) (0.90) (0.90) (0.90) (0.99) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (0.81) (0.81) (0.81) (0.81) (0.80) (0.80) (0.80) (1.00) (1.00) (1.00) (0.99) (0.99) (0.99) (0.99) (0.99)

(1.00) (1.00) (0.99) (0.99) (0.99) (0.99) (0.99) (0.99) (0.99) (0.99) (0.99) (0.99) (0.59) (0.91) (0.84) (0.84) (0.84) (0.99) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (0.56) (0.56) (0.56) (0.56) (0.56) (0.56) (0.56) (1.00) (1.00) (1.00) (0.98) (0.98) (0.98) (0.98) (0.98)

Notes: a Scale of 1-7: 1=0%,7=100%. b Scale of 1-5: 1=’Strongly Agree’/’Very Good’, 5=’Strongly Disagree’/’Very Poor’.

Table 1 (continued). Summary of Workplace Samples Variable

Definition

For Profit (N=1,281) Mean (Valid)

All Firms (N=1,819) Mean (Valid)

Other Characteristics Mining-1 Mining-2 Mining-3 Mining-4 Mining-5 Manufacturing-1 Manufacturing-2 Manufacturing-3 Manufacturing-4 Manufacturing-5 Manufacturing-6 Manufacturing-7 Manufacturing-8 Manufacturing-9 Electric & Gas Water & Sewerage Construction-1 Construction-2 Wholesale-1 Wholesale-2 Wholesale-3 Retail-1 Retail-2 Retail-3 Accommodation/Restaurant Transport-1 Transport-2 Transport-3 Transport-4 Transport-5 Transport-6 Storage Communal Finance Insurance F&I Services Property Services Business Services Govt. Admin. Education Health Services Community Services Film, Radio, TV Library, Museum, Art Sport, Recreational Personal Services Other Services

Coal Oil & Gas Metal Ore Other Mining Mining-Services Food, Beverages & Tobacco TCF Wood & Paper Print & Publishing Petroleum etc. Non-Metal Metal Machinery & Equipment Other Manufacturing Electric & Gas Water & Sewerage General Construction Construction-Trade Basic Materials Machines & Auto Personal & Household Food Personal & Household Motor Accommodation, Cafes and Restaurants Road Rail Water Air Other Transportation Transport-Services Storage Communal Services Finance Insurance Finance and Insurance Services Property Services Business Services Government Administration Education Health Services Community Services Film, Radio, TV Library, Museum, Art Sport, Recreational Personal Services Other Services

0.014 0.004 0.013 0.012 0.008 0.034 0.057 0.020 0.023 0.025 0.013 0.042 0.063 0.009 0.027 0.007 0.022 0.028 0.011 0.023 0.025 0.062 0.041 0.018 0.069 0.021 0.003 0.002 0.006 0.002 0.014 0.005 0.043 0.042 0.016 0.012 0.004 0.059 0.004 0.006 0.025 0.003 0.015 0.007 0.021 0.008 0.010

0.010 0.003 0.010 0.009 0.007 0.024 0.038 0.012 0.016 0.016 0.008 0.029 0.039 0.005 0.025 0.010 0.026 0.020 0.009 0.015 0.016 0.046 0.026 0.012 0.053 0.016 0.006 0.002 0.006 0.002 0.010 0.007 0.031 0.033 0.011 0.009 0.004 0.057 0.069 0.076 0.067 0.014 0.115 0.015 0.018 0.008 0.043

(1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00)

(1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00) (1.00)

Table 2: Availability of/Access to Family Friendly Practices by Firm Characteristics No Union

All Union

Employeesa

Female FTa

Female PTa

PSE

No/Few Competitors

Manager Survey Family Leave Career Leave Special Leave Unpaid Leave Flex. Leave Sick Leave Annual Leave Seniority Leave Other Leave No Leave Paid Maternity Leave Paid Paternity Leave Child Care Center Care Assistance Care Subsidy Holiday Care Other Child Care Elder Care No Child/Elder Care Welfare Equal Opportunity

0.150 0.033 0.438 0.703 0.390 0.601 0.826 0.471 0.060 0.018 0.178 0.080 0.024 0.009 0.027 0.012 0.021 0.000 0.925 0.231 0.467

0.364 0.100 0.533 0.743 0.425 0.548 0.814 0.707 0.076 0.002 0.473 0.252 0.054 0.012 0.032 0.029 0.020 0.000 0.890 0.576 0.885

0.469 0.121 0.545 0.801 0.519 0.557 0.882 0.720 0.057 0.005 0.346 0.153 0.093 0.017 0.069 0.052 0.050 0.005 0.779 0.682 0.953

0.313 0.095 0.508 0.769 0.426 0.571 0.796 0.645 0.052 0.006 0.372 0.194 0.061 0.013 0.039 0.033 0.028 0.002 0.862 0.443 0.782

0.286 0.080 0.492 0.741 0.429 0.579 0.843 0.654 0.066 0.005 0.340 0.173 0.036 0.014 0.029 0.025 0.024 0.004 0.901 0.456 0.737

0.222 0.056 0.410 0.738 0.369 0.635 0.865 0.585 0.058 0.009 0.200 0.111 0.014 0.010 0.020 0.016 0.019 0.003 0.936 0.320 0.655

0.240 0.062 0.445 0.733 0.407 0.588 0.881 0.599 0.053 0.009 0.301 0.160 0.012 0.015 0.018 0.018 0.009 0.003 0.943 0.429 0.689

Employee Survey Frac. Part Time Frac. M/P Leave Frac. Child Care Frac. Family Phone Frac. Home-Work Frac. Flex Schedule

0.546 0.642 0.044 0.757 0.288 0.740

0.565 0.785 0.043 0.709 0.193 0.638

0.578 0.819 0.061 0.747 0.201 0.662

0.679 0.816 0.046 0.740 0.199 0.676

0.530 0.726 0.045 0.736 0.196 0.650

0.559 0.685 0.043 0.751 0.188 0.661

0.504 0.690 0.050 0.762 0.177 0.635

Variable

Notes: a Workplaces with above average realizations of the characteristic based on the averages in column 3 in Table 1.

Table 3: Labor Productivitya by Availability of/Access to Family Friendly Practices Benefit / Characteristic Applicable

Variable

Lower

Labor Productivity is... Same Higher

Benefit/Characteristic Not Applicable

Lower

Labor Productivity is... Same Higher

Manager Survey Family Leave Career Leave Special Leave Unpaid Leave Flex. Leave Sick Leave Annual Leave Seniority Leave Other Leave No Leave Paid Maternity Leave Paid Paternity Leave Child Care Center Care Assistance Care Subsidy Holiday Care Other Child Care Elder Care No Child/Elder Care Welfare Equal Opportunity Safety

0.13 0.13 0.13 0.12 0.11 0.12 0.13 0.13 0.10 0.08 0.14 0.12 0.14 0.26 0.14 0.20 0.24 0.33 0.11 0.14 0.13 0.13

0.36 0.31 0.38 0.39 0.39 0.38 0.38 0.39 0.41 0.33 0.40 0.42 0.40 0.17 0.37 0.27 0.32 0.50 0.39 0.37 0.39 0.39

0.51 0.56 0.49 0.49 0.50 0.50 0.49 0.48 0.50 0.58 0.46 0.47 0.47 0.57 0.49 0.53 0.44 0.17 0.50 0.48 0.48 0.48

0.12 0.12 0.11 0.13 0.13 0.12 0.10 0.10 0.12 0.12 0.12 0.13 0.12 0.12 0.12 0.12 0.12 0.12 0.18 0.10 0.10 0.07

0.40 0.40 0.39 0.40 0.39 0.40 0.44 0.39 0.39 0.39 0.37 0.38 0.39 0.39 0.39 0.39 0.39 0.39 0.36 0.40 0.39 0.39

0.48 0.48 0.49 0.47 0.48 0.48 0.46 0.50 0.49 0.49 0.51 0.50 0.49 0.49 0.49 0.49 0.49 0.49 0.46 0.50 0.51 0.54

Employee Surveyb Part Time M/P Leave Child Care Family Phone Home-Work Flex Schedule

0.10 0.12 0.16 0.13 0.11 0.10

0.38 0.39 0.36 0.37 0.39 0.37

0.51 0.48 0.48 0.50 0.50 0.53

0.15 0.13 0.12 0.12 0.14 0.15

0.39 0.39 0.39 0.40 0.38 0.40

0.46 0.48 0.49 0.47 0.48 0.45

Other Characteristic Female PTc Female FTc All Union No Union PSE Employeesc

0.12 0.10 0.09 0.09 0.12 0.16

0.37 0.37 0.43 0.36 0.38 0.35

0.51 0.53 0.48 0.55 0.50 0.49

0.13 0.14 0.13 0.14 0.13 0.11

0.36 0.40 0.38 0.39 0.40 0.40

0.51 0.46 0.49 0.47 0.48 0.49

Notes: a Labor productivity compared to main competitors. b Access/Availability is above (columns 1-3) or below (columns 4-6) average availability in all workplaces (cf. Table 1). c Characteristic available above (columns 1-3) or below (columns 4-6) average compared to all workplaces (cf. Table 1).

Table 4: Profitability by Availability of/Access to Family Friendly Practices Benefit / Characteristic Applicable Workplace reported... Break-Even Profit

Benefit/Characteristic Not Applicable

Loss

Workplace reported... Break-Even Profit

Variable

Loss

Manager Survey Family Leave Career Leave Special Leave Unpaid Leave Flex. Leave Sick Leave Annual Leave Seniority Leave Other Leave No Leave Paid Maternity Leave Paid Paternity Leave Child Care Center Care Assistance Care Subsidy Holiday Care Other Child Care Elder Care No Child/Elder Care Welfare Equal Opportunity Safety

0.11 0.10 0.18 0.16 0.16 0.16 0.16 0.14 0.11 0.30 0.21 0.20 0.23 0.29 0.12 0.18 0.07 0.00 0.16 0.16 0.15 0.16

0.09 0.05 0.08 0.08 0.09 0.09 0.09 0.09 0.14 0.30 0.10 0.07 0.23 0.12 0.09 0.05 0.07 0.00 0.09 0.07 0.08 0.09

0.80 0.85 0.74 0.76 0.75 0.75 0.75 0.77 0.74 0.40 0.69 0.73 0.54 0.59 0.79 0.77 0.86 1.00 0.75 0.77 0.77 0.75

0.18 0.17 0.15 0.17 0.16 0.17 0.17 0.19 0.17 0.16 0.14 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.17 0.17 0.16

0.09 0.09 0.10 0.12 0.09 0.09 0.11 0.10 0.09 0.09 0.08 0.09 0.09 0.09 0.09 0.09 0.09 0.09 0.09 0.10 0.12 0.10

0.73 0.74 0.75 0.72 0.74 0.74 0.71 0.71 0.75 0.75 0.77 0.75 0.75 0.75 0.75 0.75 0.74 0.75 0.76 0.73 0.70 0.74

Employee Surveya : Part Time M/P Leave Child Care Family Phone Home-Work Flex Schedule

0.14 0.16 0.16 0.16 0.17 0.16

0.11 0.08 0.11 0.09 0.10 0.10

0.75 0.75 0.72 0.74 0.74 0.74

0.18 0.16 0.16 0.15 0.15 0.16

0.08 0.11 0.09 0.10 0.10 0.09

0.74 0.73 0.75 0.75 0.75 0.75

Other Characteristic Female PTb Female FTb All Union No Union PSE Employeesb

0.11 0.16 0.19 0.17 0.15 0.11

0.09 0.09 0.09 0.11 0.09 0.06

0.80 0.75 0.72 0.73 0.75 0.83

0.21 0.16 0.16 0.16 0.21 0.18

0.06 0.09 0.09 0.08 0.08 0.10

0.73 0.75 0.75 0.76 0.71 0.72

Notes: a Access/Availability is above (columns 1-3) or below (columns 4-6) average availability in all workplaces (cf. Table 1). b Characteristic available above (columns 1-3) or below (columns 4-6) average compared to all workplaces (cf. Table 1).

Table 5: Relative Labor Productivity: Marginal Effects of Workplace Practices (Manager Survey) Variable

(1)

(2)

(3)

(4)

0.03 (0.03) 0.09∗∗ (0.04) 0.01 (0.02) 0.02 (0.03) 0.00 (0.03) 0.01 (0.02) 0.04 (0.04) -0.04 (0.03) 0.01 (0.05) -0.05 (0.03) 0.01 (0.04) -0.01 (0.06) -0.02 (0.05) 0.00 (0.03) 0.01 (0.03) -0.06 (0.04)

0.03 (0.03) 0.09∗∗ (0.05) 0.00 (0.03) 0.02 (0.03) -0.02 (0.03) 0.01 (0.03) 0.03 (0.04) -0.02 (0.03) 0.01 (0.05) -0.05 (0.03) 0.00 (0.04) -0.01 (0.06) -0.03 (0.05) 0.00 (0.03) -0.01 (0.04) -0.05 (0.04)

0.03 (0.03) 0.09∗ (0.05) 0.00 (0.03) 0.01 (0.03) -0.01 (0.03) 0.01 (0.03) 0.06 (0.04) -0.02 (0.03) 0.04 (0.05) -0.09∗∗ (0.03) 0.02 (0.04) -0.02 (0.07) -0.05 (0.05) 0.01 (0.03) 0.01 (0.04) -0.04 (0.04) 0.02 (0.06) 0.06 (0.06)

0.02 (0.03) 0.09∗ (0.05) 0.00 (0.03) 0.00 (0.03) 0.00 (0.03) 0.01 (0.03) 0.04 (0.04) 0.00 (0.03) 0.05 (0.05) -0.07∗∗ (0.03) 0.00 (0.04) -0.03 (0.07) -0.04 (0.05) 0.00 (0.03) 0.00 (0.04) -0.04 (0.04) 0.00 (0.07) 0.05 (0.06)

Controls

none

2-digit industry dummies

basic controls

basic controls+ product market/ownership

LL N Pseudo R2

-1,247.000 1,816 0.010

-1,210.530 1,816 0.038

-1,182.636 1,803 0.053

-1,154.567 1,799 0.075

Family Leave Career Leave Special Leave Unpaid Leave Flex. Leave Sick Leave Annual Leave Seniority Leave Other Leave Paid Maternity Leave Paid Paternity Leave Child Care Center Child/Elder Care Support Welfare Equal Opportunity Safety Fraction Female PT Fraction Female FT

Notes: The binary dependent variable measures higher labor productivity in the workplace relative to main competitors based on the scale: ’a lot higher’ or ’a little higher’ compared to ’about the same’, ’a little lower’, or ’a lot lower’. HuberWhite corrected standard errors in parentheses. ∗ Statistically significant at the .10 level; ∗∗ at the .05 level (two-tailed test); ∗∗∗ at the .01 level (two-tailed test).

Table 6: Profitability: Marginal Effects of Workplace Practices (Manager Survey) Variable

(1)

(2)

(3)

(4)

0.03 (0.03) 0.11∗∗ (0.05) -0.02 (0.03) 0.02 (0.03) 0.00 (0.03) -0.01 (0.03) 0.00 (0.04) 0.03 (0.03) 0.02 (0.05) -0.10∗∗∗ (0.04) 0.03 (0.04) -0.22∗∗ (0.11) 0.04 (0.05) 0.03 (0.03) 0.07∗∗ (0.03) -0.04 (0.04)

0.01 (0.03) 0.09∗∗ (0.05) -0.03 (0.03) 0.02 (0.03) 0.02 (0.03) -0.01 (0.03) -0.03 (0.04) 0.02 (0.03) 0.05 (0.05) -0.09∗∗ (0.04) 0.00 (0.05) -0.20∗ (0.12) 0.04 (0.05) 0.05∗ (0.03) 0.06∗ (0.04) -0.03 (0.04)

0.02 (0.03) 0.12∗∗∗ (0.04) -0.02 (0.03) 0.02 (0.03) 0.00 (0.03) -0.02 (0.03) 0.01 (0.04) 0.02 (0.03) 0.00 (0.06) -0.09∗∗ (0.04) 0.03 (0.04) -0.28∗∗ (0.12) 0.00 (0.06) 0.01 (0.03) 0.06 (0.04) -0.04 (0.04) 0.11∗ (0.06) -0.07 (0.06)

0.03 (0.03) 0.12∗∗∗ (0.04) -0.03 (0.03) 0.02 (0.03) 0.00 (0.03) -0.02 (0.03) 0.00 (0.04) 0.02 (0.03) 0.01 (0.06) -0.09∗∗ (0.04) 0.03 (0.04) -0.27∗∗ (0.12) 0.00 (0.06) 0.01 (0.03) 0.06 (0.04) -0.04 (0.04) 0.09 (0.07) -0.09 (0.06)

Controls

none

2-digit industry dummies

basic controls

basic controls+ product market/ownership

LL N Pseudo R2

-699.303 1,280 0.037

-652.587 1,275 0.100

-663.184 1,269 0.082

-643.561 1,264 0.107

Family Leave Career Leave Special Leave Unpaid Leave Flex. Leave Sick Leave Annual Leave Seniority Leave Other Leave Paid Maternity Leave Paid Paternity Leave Child Care Center Child/Elder Care Support Welfare Equal Opportunity Safety Fraction Female PT Fraction Female FT

Notes: The binary dependent variable measures profitability in the workplace based on the scale: ’Profit’ compared to ’Break Even’ or ’Loss’. Huber-White corrected standard errors in parentheses. ∗ Statistically significant at the .10 level; ∗∗ at the .05 level (two-tailed test); ∗∗∗ at the .01 level (two-tailed test).

Table 7: Relative Labor Productivity: Marg. Effects of Workplace Practices (Employee Survey) Variable

(1)

(2)

(3)

(4)

0.09∗∗ (0.04) -0.02 (0.05) -0.04 (0.11) -0.02 (0.06) 0.01 (0.06) 0.13∗∗ (0.06) 0.00 (0.03) 0.01 (0.03) -0.05 (0.04)

0.09∗ (0.05) -0.04 (0.05) -0.03 (0.11) 0.01 (0.06) 0.02 (0.06) 0.06 (0.06) 0.00 (0.03) 0.00 (0.04) -0.03 (0.04)

Employees Not Surveyed

0.19 (0.20)

0.17 (0.22)

0.03 (0.04) -0.05 (0.05) -0.06 (0.11) -0.04 (0.06) 0.02 (0.06) 0.07 (0.06) 0.01 (0.03) 0.01 (0.04) -0.04 (0.04) 0.01 (0.06) 0.06 (0.06) 0.16 (0.19)

0.01 (0.05) -0.04 (0.05) -0.08 (0.11) -0.03 (0.06) 0.02 (0.06) 0.06 (0.06) 0.00 (0.03) 0.00 (0.04) -0.03 (0.04) -0.01 (0.07) 0.06 (0.06) 0.11 (0.20)

Controls

none

2-digit industry dummies

basic controls

basic controls+ product market/ownership

LL N Pseudo R2

-1245.440 1,817 0.011

-1210.818 1,817 0.038

-1187.989 1,804 0.050

-1157.639 1,800 0.072

Fraction Permanent Part Time Fraction Maternity/Paternity Leave Fraction Subsidized Child Care Fraction Access to Family Phone Fraction Work-from-Home Possible Fraction Flexible Schedule Welfare Equal Opportunity Safety Fraction Female PT Fraction Female FT

Notes: The binary dependent variable measures higher labor productivity in the workplace relative to main competitors based on the scale: ’a lot higher’ or ’a little higher’ compared to ’about the same’, ’a little lower’, or ’a lot lower’. HuberWhite corrected standard errors in parentheses. ∗ Statistically significant at the .10 level; ∗∗ at the .05 level (two-tailed test); ∗∗∗ at the .01 level (two-tailed test).

Table 8: Profitability: Marginal Effects of Workplace Practices (Employee Survey) Variable

(1)

(2)

(3)

(4)

0.02 (0.04) 0.03 (0.04) -0.16 (0.10) -0.04 (0.06) 0.03 (0.06) -0.03 (0.06) 0.02 (0.03) 0.08∗∗ (0.03) -0.04 (0.04)

-0.04 (0.05) 0.01 (0.04) -0.19∗ (0.10) -0.02 (0.06) 0.04 (0.06) -0.05 (0.06) 0.05 (0.03) 0.06∗ (0.03) -0.03 (0.04)

Employees Not Surveyed

0.22∗∗ (0.08)

0.21∗∗ (0.09)

0.01 (0.04) 0.01 (0.05) -0.17 (0.11) -0.05 (0.06) 0.00 (0.06) -0.02 (0.06) 0.00 (0.03) 0.07∗ (0.04) -0.04 (0.04) 0.12∗ (0.06) -0.08 (0.06) 0.21∗∗ (0.09)

0.01 (0.04) 0.02 (0.05) -0.21∗ (0.11) -0.05 (0.06) 0.03 (0.07) -0.04 (0.06) 0.01 (0.03) 0.07∗ (0.04) -0.04 (0.04) 0.11 (0.07) -0.09 (0.06) 0.19∗ (0.10)

Controls

none

2-digit industry dummies

basic controls

basic controls+ product market/ownership

LL N Pseudo R2

-706.847 1,281 0.027

-655.584 1,276 0.096

-668.174 1,270 0.075

-647.360 1,265 0.102

Fraction Permanent Part Time Fraction Maternity/Paternity Leave Fraction Subsidized Child Care Fraction Access to Family Phone Fraction Work-from-Home Possible Fraction Flexible Schedule Welfare Equal Opportunity Safety Fraction Female PT Fraction Female FT

Notes: The dependent variable measures profitability in the workplace based on the scale: ’Profit’ compared to ’Break Even’ or ’Loss’. Huber-White corrected standard errors in parentheses. ∗ Statistically significant at the .10 level; ∗∗ at the .05 level (two-tailed test); ∗∗∗ at the .01 level (two-tailed test).