Gender, Work Burden, and Mental Health in Post-Reform China

ABSTRACT

This study investigates how total work burden, including paid work and unpaid care work, affects the mental health of prime-age, employed women and men in urban China. Based on the 2010 China Family Panel Studies (CFPS), regression results indicate that total work burden is negatively related to the mental health of both men and women, consistent with the idea that additional work hours reduce time available for rest and leisure. Women have longer working hours and are more likely to be time poor than men, and this gender inequality in total work burden contributes to the gender gap in mental health. The relationships between the components of total work burden – paid and unpaid work – and mental health shed further light on the strength of gender norms and the barriers to redistribution of unpaid work from women to men necessary to reduce the gender gap in mental health.

KEYWORDS:

• Work burden
• paid work
• unpaid work
• time poverty
• mental health
• China

JEL: Codes::

• J16
• J22

ACKNOWLEDGMENTS

The authors would like to thank the anonymous referees and the guest editors for their instructive comments, as well as Professor Xiaoyan Lei for helpful comments on an earlier version of the study. The work was carried out with the aid of a grant from the International Development Research Center of Canada (IDRC; Project no. 107579). The views expressed in this paper are those of the authors and do not in any way implicate the IDRC.

Notes

1 While mental health is defined in terms of positive attributes, it is typically measured by the presence of mental disorders or mental illnesses (WHO 2014). Women experience higher rates of anxiety and depression, and men experience higher rates of social psychosis (see WHO [2002] for China; see Seedat et al. [2009] for other countries).

2 While some paid and unpaid work activities have positive impacts on mental health (such as stimulating tasks or playing with children), for already employed workers, increases in total work burden are expected to negatively impact mental health.

3 Evidence from several countries indicates that unpaid caregiving reduces mental well-being, particularly for women. These studies used different measures of the caregiving work burden, and samples included people who are employed as well as those not in the labor force (MacDonald, Phipps, and Lethbridge 2005; Ho et al. 2009; Glavin and Peters 2015).

4 We could not construct a longitudinal sample from the CFPS because the information on time use is available only in the 2010 survey (see CFPS [2010] for a description of the survey).

5 The regions excluded from the survey are Xinjiang, Tibet, Qinghai, Inner Mongolia, Ningxia, Hainan, Hong Kong, Macao, and Taiwan.

6 Urban labor force participation rates (LFPR) from the CFPS are similar to countrywide LFPR estimates of 82.9 and 69.4 percent, for men and women, respectively (for ages 15–64; International Labour Organization [ILO] 2014).

7 The CFPS organizer recommended the sum of the scores on the six questions as a valid measure of mental health on the basis of principal factor analysis (see CFPS [2010, 2012]).

8 The overall prevalence of mental illness in China is reportedly lower than in other countries, although this has been attributed to the possibility of greater stigma associated with mental illness resulting in lower participation in surveys and underreporting in responses in China, as well as in insufficient nuance in the translation of mental health concepts and definitions derived from the US (WHO World Mental Health Survey Consortium 2004; Huang et al. 2008).

9 It is difficult to define the cut-off score for mental health assessment, as the interpretation of MHS depends upon cultural and contextual factors. The MHS from the CFPS is compatible with MHS drawn from other population-based surveys in China, such as the Third Survey of Chinese Women's Social Status, conducted jointly by the All China Women's Federation and China's National Bureau of Statistics in 2010, which used an eight-item mental health questionnaire, with the response measured on a scale of 0–3.

10 For Equation 1(1) , we estimated a model with a squared term of TH, and this variable was statistically insignificant.

11 In the Chinese residential registration system, those who are born in rural areas hold a rural hukou and those who are born in cities hold an urban hukou. Hukou determines a person's access to social security and services and privileges urban over rural hukou holders. In the 2010 CFPS, people with a rural hukou residing in urban areas include both migrant workers from rural areas and local suburban residents with a rural hukou, as many suburban areas used to be classified as rural areas.

12 The regression models do not control for household composition variables in order to avoid multicollinearity problems, as the variables on household composition are important determinants of paid and unpaid work hours. We did estimate models that include the variables on household gender and age composition, however, and the estimates of primary interest are substantively similar to those presented in Table 3.

13 We collapse the five categories of self-reported health in the CFPS into two categories: Poor health captures the three categories of relatively unhealthy, unhealthy, and very unhealthy. Good health captures the two categories of healthy and fair health.

14 The model for lower quantiles is not estimated because 35 percent of the workers score 0. See Roger Koenker and Kevin F. Hallock (2001) for quantile regression methods.

15 To check the robustness of OLS estimates, we estimated two sets of models, with one controlling for community fixed effects and another for province fixed effects. The two sets of estimates are substantively similar. To streamline the presentation, we report only the OLS estimates of the models that control for province fixed effects. The quantile regression is not feasible for the model with community dummy variables as there are no observations at the median and the third quantile of the full sample in many communities.

Additional information

Notes on contributors

Lan Liu

Lan Liu is Associate Professor in the Institute of Population Research at Peking University, China. Her main research interests are in the field of labor economics and gerontology; her research focuses on eldercare, gender equality, and public policy in China.

Fiona MacPhail

Fiona MacPhail is Professor of Economics at the University of Northern British Columbia and Book Review Editor for Feminist Economics. Her research focuses upon gender, work, and policy in Canada as well as several Asian countries; she has a particular research interest in migration.

Xiao-yuan Dong

Xiao-yuan Dong is Professor in the Department of Economics at the University of Winnipeg, Canada and Adjunct Professor of the National School of Development, Peking University, China. She has served on the Board of Directors of the International Association of Feminist Economics and is Associate Editor of Feminist Economics. Her research focuses on China's economic development, with an emphasis on labor and gender issues.