Redesigning Work: Findings from the Work, Family, and Health Network and Implications for Gender Inequality

Social welfare policies in the United States are unique in that they are often provided by employers – think of health care, retirement benefits, and particularly work-family policies. How do we formulate effective work-life integration policies that work for employers and for employees?

This week’s WAPPP seminar featured Erin L. Kelly, Professor of Work and Organization Studies at the MIT Sloan School of Management. Professor Kelly’s early work focused on this hybrid social welfare system in the United States, and she has continued to investigate organizational policies that may address work-family and gender issues. In particular, her work focuses on the consequences of workplace policies for employees and organizations.

Work-life conflict, job control, and flexible schedules

Specific work environments affect work-life conflict and strains. As we’ve discussed in past WAPPP seminars, job control can affect a range of health-promoting behaviors. With more control over where and when you do your work, you may have more time to exercise, cook healthy meals, get enough sleep, and reduce stress. Professor Kelly’s research as part of the Work, Family, and Health Network tests promising workplace initiatives with rigorous methods to determine what workplaces can change to improve the health and well-being of employees and their families while maintaining or improving organizations’ key outcomes.

While flexible work arrangements (like telecommuting and flex schedules) are generally considered a good thing, they may have unintended consequences. Generally, flexible schedules are decided through an “accommodation model,” in which individual employees negotiate with their managers directly. However, there are all kinds of possible confounding factors: who is willing to ask for flexible scheduling? Who gets it? Is there a “flex stigma,” such that flexible workers are perceived as less committed or less competent? Are gendered inequalities exacerbated because flexible work schedules are associated with women and families?

STAR: Normalizing flexibility

Professor Kelly’s work tries to sidestep some of these issues by making flexibility the new normal. She presented the Work, Family, and Health Network’s randomized controlled trial of flexible work in an IT division of a Fortune 500 company. Half of the work units were enrolled in STAR, a program in which work groups reconsider when, where, and how work is done. These decisions are collective and broadly framed to focus on all employees, not necessarily family needs or women’s struggles. 

The other half of the work units remained under existing company policy. Official policy allowed employees to work from home with a formal request that went all the way up to the VP. Existing practice, on the other hand, was more informal – some employees were able to work from home or shift hours, but this practice was deliberately under the radar. 

The researchers collected baseline survey and health data, along with daily diaries and spouse and child surveys. They repeated survey and health data collection at 6, 12, and 18 months. One complication arose midway through the study: the company announced that it was being acquired. The merger was implemented partway through the study.

Effects of STAR on work-family outcomes

Participating in STAR was associated with increased control over work time, greater supervisor support for personal and family life, and employees saying they had “enough time” with family. There was no effect on work hours or overall intensity. While some employers may have hoped for increased productivity, the researchers see this  as a positive sign – flexibility shouldn’t turn into pressure to be “always on.” Significantly, there was no difference in work hours with STAR based on gender or parental status. 

In terms of work outcomes, STAR increased job satisfaction at 12 months and reduced turnover intentions at 12 months. STAR also reduced voluntary exits at 3 years (7.6% of STAR participants left the company, compared to 11.3% of control group members). A related paper estimates  that STAR had a positive ROI of 1.6 for the firm. The employer reduced turnover costs and the lost productivity that comes with training new workers. 

Effects of STAR on health outcomes

STAR decreased burnout, stress, and psychological distress at the 12-month follow up for participants who began STAR before the merger was underway. For other employees, anxiety about the merger (and the future of STAR!) may have confounded these effects. Participants in STAR showed modest increases in sleep duration and quality at the 12-month follow up. STAR also increased parents’ time with children over 12 months. 

Interestingly, many of these benefits transferred to STAR participants’ children. Adolescents whose parents were in STAR showed better sleep quality and more consistent sleep duration, along with more positive affect and less emotional reactivity to daily stressors. Professor Kelly believes that this effect is partially due to parents being more available and able to stick to routines that work for their families with flexible schedules. 

The STAR experiment shows the benefits of flexible work schedules on work-family and health outcomes. However, Professor Kelly brought up a number of other questions for future study. How did behaviors actually change under STAR? Did groups differ in their utilization of flex time? We look forward to hearing more from Professor Kelly and the Work, Family, and Health Network!