Understanding the gender gap in sickness absence

The project's main ambition has been to find explanations for why we observe a gender difference in sickness absence in Norway (a phenomenon in many countries with high female labor force participation). The project is empirical, quantitative and has especially investigated the relation between work, family and sickness absence. The project has resulted in five articles. One article is descriptive and decomposes the gender gap in sickness absence over time (1995-2013) and over the life-cycle (20-60 years). Three articles investigate the effect of having children - on sickness absence and other labor market outcomes. One last article investigates how increased labor market competition after the EU expansion in 2004 affects Norwegian manual workers and their wives' work, income and sickness absence. "What can (not) explain the gender gap in sickness absence over time - a decomposition analysis". Sara Cools, Marte Strøm og Kjersti Misje Østbakken. Working paper - The article is a descriptive analysis of the development in gender differences in sickness absence over the period 1995-2013. This is a period where women's educational level has increased relative to men's and women increasingly participate in the labor market similar to men. We may expect that these developments would lead to a convergence in sickness absence rates for women and men. The opposite has happened: over the period, the gender gap in sickness absence has increased. We use register data for the full population aged 20-60 years and use decomposition methods to investigate whether differences in education, industry, pregnancy and family can explain the gender gap in sickness absence. They explain parts of the difference, but not the increase over time. "Sickness absence among men and women: What role do children play?" (2015) Sara Cools, Simen Markussen og Marte Strøm. Published in Søkelys på Arbeidslivet. - Sick leave among women increases dramatically at childbearing age and remains higher than among men long after the children have grown up, which suggests a causal relationship between childbearing and the long-term health of women, possibly through a change in total workload. Using Norwegian registry data, we instrument number of children using the sex mix of the first two children and estimate the effect of family size on various measures of health. We find no adverse effects on health, except for a short-term rise in pregnancy related sickness absence. In the long run (20-30 years), however, there are indications that health is affected favorably. "Children and Careers: How Family Size Affects Parents? Labor Market Outcomes in the Long Run" (2017). Sara Cools, Simen Markussen and Marte Strøm. Published in Demography. - We estimate the effect of family size on various measures of labor market outcomes over the whole career using data from Norwegian administrative registers. We instrument parents' number of children using the sex mix of their first two children. We find that having additional children causes sizable reductions in labor supply for women, which fade as children mature. Among women with a college degree, there is evidence of persistent and even increasing career penalties of family size. Some of the career effects persist long after labor supply is restored. We find no effect of family size on any of men's labor market outcomes. "Children and parents' labor market outcomes". Simen Markussen and Marte Strøm. Revise and resubmit - We use miscarriage as a biological shock to fertility in order to estimate the effect of children on male and female labor market and the probability of receiving a health-related welfare benefit. We apply an event study framework and IV-estimations to estimate the effects. We find that each of the three first children negatively affects women's labor supply in the short run, and the marginal effect of each child is equally large. Having the first child also increases the probability of receiving a disability insurance - for both men and women. "International labor market competition and household dynamics". Pål Schøne and Marte Strøm. Working paper. - The last decades, manual workers in Norway and other high income countries in EU/EØS have been increasingly exposed to labor market competition from low income countries, especially after 2004. In this article, we investigate how increased labor market competition affects household income, work and sickness absence. We compare non-licenced manual workers (like carpenters, painters) to manual workers who need a licence to perform their occupation (like plumbers, electricians) and show that non-licensed manual workers experience a relatively poorer wage development and a higher probability of not working. We also find that the wife compensates for some of the earnings loss by working. We do not find, however, that the husband's poorer labor market prospects leads to higher sickness absence for the husband nor the wife.

Resultatene forventes å ha nytte for videre forskning på kjønnsforskjeller i sykefravær, men også for forskning på kjønnsforskjeller i arbeidsmarkedet mer generelt. Resultatene forventes også å ha nytte for offentlige myndigheter som utformer politikk for sykefravær ved å identifisere faktorer som påvirker ? og ikke påvirker ? kjønnsforskjeller i sykefravær. Resultatene kan også være relevante for politikk på likestillingsområdet. Prosjektet har i tillegg bidratt til å heve forskerkompetansen til prosjektleder og samarbeidspartnere/samarbeidsmiljøer både når det gjelder problemstillinger i prosjektet og kompetanse på ulike empiriske metoder.

In the project I plan to explore three possible mechanisms behind the gender gap. They are related to: (1) family formation (2) investments in health (3) statistical discrimination Project 1 on the role of family formation will compare behavior before a nd after children. One article will give an overview of the life-cycle patterns in sickness absence for men and women. My work with Simen Markussen shows that this dimension is an important one. There is a shift in the level of women?s sickness absence af ter first birth that does not disappear even 30 years later. I wish to do a detailed decomposition of sickness absence along the life-cycle dimension exploiting the high frequency and precision of Norwegian register data. Another article in the same proj ect will approach the problem of identifying the causal impact of children on sickness absence by using miscarriage as an instrumental variable for number of children. Miscarriage gives random variation in the number of children (women with a miscarriage have children later) which is not determined by unobservable factors correlated to labour market outcomes and sickness absence. This allows us to estimate the causal impact of number of children on sickness absence. Project 2 is about the seemingly parado x that women have higher sickness absence than men but at the same time live considerably longer. A possible explanation for this is that women are more risk-averse and take better care of their health. Another possible explanation is that there are biolo gical differences between men and women in the sort of illnesses they get and how fast they recover or not. I will investigate the gender difference in behavior following the same health shock to separate out the biological factor.