By 2030, approximately 1.2 billion women worldwide will either be undergoing the menopause transition or be post-menopausal (Delanerolle et al. 2025). Although menopause is medically defined as the moment when a woman’s menstrual period has been missing for 12 months, hormonal changes and debilitating symptoms typically start years earlier during a period called ‘perimenopause’ and may extend for years after transitioning. Common symptoms, which can be bothersome at best and incapacitating at worst, include vasomotor symptoms (e.g. hot flushes and night sweats), sleep disruption, migraines, bone health deterioration, and emotional symptoms such as depression and anxiety (Monteleone et al. 2018). Despite the disruptive nature of this unavoidable biological process, its labour and health impacts have been understudied, particularly from a causal perspective.

Menopause is having a moment

As female labour force participation increases and the world population ages, middle-aged women are becoming a larger share of the workforce. Consequently, the public discussion about menopause has recently taken off in policy circles, academia, and social media, led primarily by women at this stage in their lives. Such public momentum needs to be supported by rigorous research to effectively trigger social and policy changes, as has been the case with other biological events affecting women.

For instance, research on the effects of menstruation on schooling outcomes, such as drop-out rates and absenteeism, led to campaigns to increase school sanitation and access to menstrual technologies (Oster and Thornton 2011) and interventions to reduce period stigma (Macours 2025). On a different front, the literature documenting the ‘child penalty’ (Kleven, Landais, and Leite-Mariante 2025, Steinhauer et al. 2019) has derived into a policy debate about how to reduce the disproportionate childbearing costs borne by women and close the gender gap with paid leave, affordable childcare, or a reshaping of gender norms (Kleven, Olivero, and Patacchini 2025).

Thus, causal evidence of the impacts of menopause on female work and health outcomes is relevant to inform policymakers and employers, support women during this transition, and potentially mitigate its costs. For instance, a notable and controversial mitigation strategy for menopausal symptoms is the use of hormone replacement therapy (HRT). HRT was commonly prescribed since the 1960s until the (now highly contested) Women’s Health Initiative Study in 2002 linked it to higher risks of heart attacks, strokes, blood clots, and breast cancer. The publication of this study led to an unexpected and abrupt decrease in HRT take-up rates, which Daysal and Orsini (2014) exploit to provide evidence on the positive impact of this medical treatment on women’s employment. However, overall, the evidence on the protective role of HRT on work and, additionally, health outcomes is still scarce.

The emerging evidence on the hidden costs of menopause

Most of the existing evidence on the effects of menopause on labour and health outcomes is correlational. The timing of menopause varies across women, typically occurring during the late 40s or early 50s, with an average age of 51, when women’s careers and family lives may also be undergoing major changes. Consequently, obtaining causal evidence requires disentangling the impact of menopause from other correlated factors.

A few studies use quasi-experimental methods to provide such causal estimates, such as Bryson et al. (2022), who find that early menopause (before age 45) reduces the employment rate of women in their 50s. More recently, Conti et al. (2025) use administrative data from Sweden and Norway and an event study to estimate the ‘menopause penalty’ and find that receiving a menopause diagnosis decreases women’s full-time employment and earnings and increases public transfer receipts. Abrahamsson et al. (2025) combine register and survey data with the date of menopause from Norway and similarly find decreases in earnings for women with more severe menopause-related symptoms.

Our recent research, part of this emerging literature, estimates the causal effects of the menopause transition on labour and health outcomes for US women (Juarez and Marquez-Padilla 2024). We use the National Longitudinal Survey of Young Women (1988-2003) and the Study of Women’s Health Across the Nation (1997-2006), two longitudinal surveys that provide information on the exact timing of menopause (the age of the last menstrual period), together with labour, health, and socio-demographic variables. Leveraging the variation in menopause age between women, we use an event study and a difference-in-difference framework, in the spirit of the child penalty literature, and set our event time as two years before a woman’s last period to capture the point at which hormonal changes and symptoms sharply increase, during perimenopause (Politi et al. 2008).

A key advantage of our research is that we do not need to proxy for the event time (a woman’s last menstrual period) using age, genetic markers, or medical diagnoses. Thus, we can estimate the effects of menopause for all women, independent of menopause severity or access to healthcare and with little measurement error. As impacts may be heterogeneous across cohorts defined by menopause age and over time, we use the imputation procedure proposed by Borusyak et al. (2024) to estimate them.

Our results show that menopause decreases women’s probability of employment, by about 10 percentage points, three years after menopause and increases their probability of working part-time by about 10 percentage points, if employed (Figure 1). We do not find significant effects on monthly earnings for women who remain employed but do find significant increases in the probability of receiving Social Security disability benefits (5 percentage points), and a large fall in the probability of being married or living with a partner (of up to 20 percentage points).

Figure 1 Labour outcomes around the menopause transition

Notes: Each graph shows the estimated average treatment effects by year, using the imputation method by Borusyak et al. (2024), with their 95% confidence intervals. The vertical dashed line indicates the reference period, two years before menopause. The reported mean dependent variable is calculated at the reference period. Data from the National Longitudinal Survey of Young Women (1988-2003). Standard errors are clustered at the individual level.

Regarding health outcomes, we find strong detrimental effects on bone health, such as increases in the probabilities of being diagnosed with osteoporosis and experiencing bone fractures, but no significant effect on other common chronic conditions, such as diabetes or hypertension.

We also find that HRT take-up increases around menopause, as expected, but it is significantly stronger for white and high-socioeconomic-status women, as proxied by their education, than for women who are non-white or of low socioeconomic status, suggesting inequalities in access to medical treatment during this transition.

HRT use may be correlated with the severity of menopausal symptoms or labour market attachment. Thus, to overcome this endogeneity and provide causal estimates of the protective role of HRT on work and health outcomes, we estimate our baseline models separately for women who experienced menopause before and after the 2002 publication of the Women’s Health Initiative Study, which, as mentioned, exogenously lowered HRT uptake (Daysal and Orsini 2014). Our results show that the negative impacts of menopause are mostly significant after the publication of the Women’s Health Initiative Study, suggesting that HRT mitigates the menopause labour penalty and protects bone health.

Changing ‘the change’: Implications for policy

Decades after experiencing the child penalty (the median age at first birth in 1980 was 22.6; see Eickmeyer et al. 2017), the ‘menopause penalty’ comes at a time when women’s labour market outcomes could be peaking: for 1995-2005, the median income for 55-year-old men exceeded that of 45-year-olds, while the reverse is true for women (Smith 2022).

Our findings on the protective role of HRT for work and bone health, along with the inequalities in access to HRT by race and education, underscore the need to expand menopause-related medical care to reduce labour market gaps – both between men and women and among women of different backgrounds.

More broadly, increasing access to healthcare services and medical training about menopause, reducing social stigma, understanding social norms surrounding women’s retirement (Wang 2024), and adapting workplace and employer practices to support women’s labour market attachment are all potential avenues to mitigate the costs of menopause. Furthermore, supporting women’s labour and health during the menopausal transition could also lead to a range of long-term benefits, such as better pensions and less economic dependency in old age. Our results on the protective role of HRT on bone health suggest that expanded access to menopause-related care can protect women against frailty and physical dependency as well, which is costly for them, their families, and society as a whole.

Editors’ note: This column is published in collaboration with the International Economic Associations’ Women in Leadership in Economics initiative, which aims to enhance the role of women in economics through research, building partnerships, and amplifying voices.

References

Abrahamsson, S, M Barschkett, and M Flatø (2025), “Beyond hot flashes: The career cost of menopause”, IZA Discussion Paper No. 17789.

Borusyak, K, X Jaravel, and J Spiess (2024), “Revisiting event-study designs: Robust and efficient estimation”, Review of Economic Studies 91(6): 3253–85.

Bryson, A, et al. (2022), “The consequences of early menopause and menopause symptoms for labour market participation”, Social Science and Medicine 293: 114676.

Conti, G, et al. (2025), “The menopause ‘penalty’”, NBER Working Paper No. 33621.

Daysal, N M, and C Orsini (2014), “The miracle drug: Hormone replacement therapy and labor market behavior of middle-aged women”, IZA Discussion Paper No. 7993.

Delanerolle, G, et al. (2025), “Menopause: A global health and wellbeing issue that needs urgent attention”, The Lancet Global Health 13(2): e196-e198.

Eickmeyer, K J, K K Payne, S L Brown, and W D Manning (2017), “Crossover in the median age at first marriage and first birth: Thirty-five years of change”, Bowling Green State University, National Center for Family & Marriage Research.

Juarez, L, and F Marquez-Padilla (2024), “Midlife crisis: The labor and health impacts of the menopause transition”, SSRN 5264120.

Kleven, H, C Landais, and G Leite-Mariante (2025), “The child penalty atlas”, Review of Economic Studies 92(5): 3174–207.

Kleven, H, G Olivero, and E Patacchini (2025), “Child penalties and parental role models: How classroom exposure shapes gender gaps”, VoxEU.org, 8 February.

Macours, K (2025), “Ending period stigma in schools”, VoxEU.org, 5 March.

Monteleone, P et al. (2018), “Symptoms of menopause – global prevalence, physiology and implications”, Nature Reviews Endocrinology 14(4): 199–215.

Oster, E, and R Thornton (2011), “Menstruation, sanitary products, and school attendance: Evidence from a randomized evaluation”, American Economic Journal: Applied Economics 3(1): 91–100.

Politi, M C, M D Schleinitz, and N F Col (2008), “Revisiting the duration of vasomotor symptoms of menopause: A meta-analysis”, Journal of General Internal Medicine 23(9): 1507–13.

Smith, T (2022), “Life cycle income profiles”, Research Highlights, American Economic Association, 9 November.

Steinhauer, A, C Landais, J Posch, and H Kleven (2019), “Child penalties across countries: Evidence and explanations”, VoxEU.org, 14 May.

Wang, X (2024), “The second glass ceiling for women”, International Economic Association, 12 December.