‘Like second-hand smoke’: the toxic effect of workplace flexibility bias for workers’ health

ABSTRACT

As many workers struggle to reconcile their personal and work responsibilities, scholars have called on organizations to redesign and redefine workplace structures and cultures. One pernicious feature of today’s organizations is workplace flexibility bias, workers’ sense that they will face career consequences for making schedule adjustments for family or personal reasons. Expanding the theoretical and empirical understanding of this bias, we ask whether and how flexibility bias affects workers’ health. Analysis of 2008 National Study of the Changing Workforce data shows that workplace flexibility bias is related to minor health problems, poor self-rated health, poor sleep quality, depression symptoms, and more frequent sick day use. Mediation analysis illustrates that this is due in part to the increased stress and negative work-life spillover that accompanies flexibility bias. This bias is also linked to alcohol use, exercise frequency, and treatment for high blood pressure and mental health issues indirectly via stress and spillover. Importantly, these deleterious effects are net of caregiving responsibilities and flexible work arrangement use, suggesting that current workplace structures and cultures are not just problematic for workers with family responsibilities but can make all types of workers sick. These results help provide more compelling scholarly and business cases for work redesign and redefinition.

RESUMEN

Mientras que muchos trabajadores luchan por reconciliar sus responsabilidades personales y laborales, los académicos han exigido que las organizaciones rediseñen y redefinan sus estructuras y culturas laborales. Una característica perniciosa de las organizaciones de hoy es el prejuicio por la flexibilidad laboral, es decir, la sensación que tienen los trabajadores de que si hacen ajustes en el horario por razones familiares o personales deberán afrontar consecuencias laborales. Para aumentar el conocimiento teórico y empírico de este prejuicio, nos preguntamos si el prejuicio por la flexibilidad afecta la salud de los trabajadores y, en caso afirmativo, cómo se manifiesta. El análisis del Estudio Nacional de la Población Activa del 2008 demuestra que el prejuicio por la flexibilidad laboral está relacionado con problemas menores de salud, una autopercepción pobre de la salud, peor calidad del sueño, síntomas de depresión y ausencias laborales debidas a enfermedad. Un análisis de mediación ilustra que esto se debe en parte al aumento del estrés y a un equilibrio negativo entre la vida laboral y personal que acompaña al prejuicio por la flexibilidad. Este prejuicio está asociado también con el uso de alcohol, la rutina de ejercicio y el tratamiento para la hipertensión y los problemas de salud mental provocados indirectamente por el estrés y el desequilibrio entre la vida laboral y personal. Es importante recalcar que estos efectos negativos son independientes de que se tengan responsabilidades de cuidado de otros familiares y de que se hagan arreglos para tener un horario laboral flexible, lo cual sugiere que las estructuras y culturas laborales actuales no son problemáticas solo para los trabajadores con responsabilidades familiares, sino que pueden causar enfermedades a todo tipo de trabajadores. Estos resultados ayudan a ofrecer argumentos convincentes a nivel académico y empresarial para el rediseño y la redefinición laboral.

KEYWORDS:

• Workplace flexibility bias
• work redesign
• health

PALABRAS CLAVE:

• Sesgo de flexibilidad en el lugar de trabajo
• rediseño del trabajo
• salud
• equilibrio trabajo-vida

Acknowledgements

The authors would like to thank Lynn Fahey, Michelle Pham, Phyllis Moen, and the anonymous reviewers for their insightful feedback on an earlier draft of this manuscript. This paper was presented at the 2016 Work and Family Researchers Network in Washington, DC.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Erin A. Cech joined the department of sociology at the University of Michigan as an assistant professor in 2016. Prior to that she was a Postdoctoral Fellow at the Clayman Institute for Gender Research at Stanford University and was on faculty at Rice University. Cech’s research seeks to uncover cultural mechanisms of inequality reproduction – particularly around gender, sexual identity, and racial/ethnic inequality in STEM; cultural definitions of ‘good work’ and ‘good workers’, and cultural logics in popular explanations of inequality. Her research appears in the American Journal of Sociology, American Sociological Review, Social Forces, and Social Problems.

Lindsey Trimble O’Connor is an assistant professor of sociology at California State University Channel Islands. Before that, she was a postdoctoral fellow at the Clayman Institute for Gender Research at Stanford University. O’Connor is interested in the factors upholding gender inequality in the labor market. One area of research examines the gendered effects of workplace structures and cultures that proliferate the ideal worker norm. The second area of research investigates the role of sponsorship and networks in the maintenance of gender inequality. Her research appears in Gender & Society, Social Science Research, Work & Occupations, and Research in the Sociology of Work.

Notes

1 Despite its documented success, Best Buy discontinued this work redesign initiative in 2013 because it was not in line with the then new CEO's leadership vision (Perlow & Kelly, 2014).

2 We liken this distinction to the difference between personally being the target of gender or racial discrimination and recognizing that gender or racial bias exists in one's workplace culture. We argue that workers need not be the direct target of flexibility stigma to recognize – and be affected by – this general bias in their workplace.

3 Supporting our argument, previous research shows, for example, that women and men's job satisfaction and organizational attachment are affected by vicarious exposure to sexism at work (Miner-Rubino & Cortina, 2007; Settles, Cortina, Buchanan, & Miner, 2013). That is, even when workers are not the direct targets of sexism, their workplace experiences are negatively impacted by the presences of bias in their workplace.

4 In using the second-hand smoke metaphor, we do not mean to imply that workers with family responsibilities deserve their poor health in the same way that many people feel smokers deserve poor health for engaging in an activity that is widely thought to be harmful. We use this metaphor instead to facilitate the readers’ conceptualization of workplace flexibility bias as something hidden and harmful that percolates ‘in the air’ of a workplace, and seeps into workers in ways that affect their health.

5 A key stress buffer is a social support – one's access to persons or groups that provide support to deal with the complexities of one's life (Pearlin et al., 1981). Working in an organization with workplace flexibility bias may also undermine workers’ sense of social support because the presence of this bias may mean that workers do not feel comfortable openly discussing work-life challenges with their coworkers or their supervisors. Future research should examine this possibility.

6 See Idler and Benyamini (1997) and Krause and Jay (1994) for discussions of self-rated health as a broad-based indicator of a range of physical health issues.

7 This measure taps into respondents’ personal view that penalties for workers who make schedule adjustments or take leave for personal or family reasons exist in their workplace. It is a subjective, individual-level measure of workers’ belief, not an organizational-level measure.

8 We estimate indirect effects in SEM models using the estat teffects post-estimation command in Stata 14. We calculated indirect effects in GSEM models post-estimation following the instructions in Stata's GSEM documentation: www.stata.com/manuals13/semexample42g.pdf (Stata, 2013).

9 Because of the need to account for the complex design of the NSCW survey through the use of ‘svy’, Stata is unable to calculate standard goodness-of-fit indices based on standard likelihood theory. We provide CD estimates for the SEM models. At the time of writing, standard goodness-of-fit measures for GSEM models does not yet exist.

10 In other words, the reason negative spillover impacts self-rated health and treatment for mental health is because spillover may stress out workers. The total indirect effects of workplace flexibility bias through both negative spillover and stress is significant in both instances.