Japanese women leaving full employment at marriage or childbirth vs. continuing to work
If you act
Leaving employment at childbirth
47%
If you don't
Continuing to work through childbirth
25%
Percentage who later regret each choice. Bars and full ledger render below.
Career
Last reviewed 2026-05-13
Evidence quality 3.63/5
Eight-dimension review score against the
quality rubric
. Each dimension scored 1–5.
D1 Source verification
3/5
D2 Source authority & independence
4/5
D3 Regret-rate accuracy
2/5
D4 Source comparability
2/5
D5 Gilovich pattern
5/5
D6 Prose quality
4/5
D7 Caveat completeness
5/5
D8 Sample quality
4/5
Average3.63/5
Proxy data — no direct regret survey exists for this decision. Rates are derived from satisfaction scores and access-barrier data rather than questions that directly asked about regret. See caveats below.
Action regret
Leaving employment at childbirth
47%
47% of Japanese women who were working before childbirth left employment after their first child (departure prevalence, not a self-reported regret rate)
Japanese women who worked before childbirth (15th National Fertility Survey, 2010–2014 cohort)
Japanese National Fertility Survey, 2015 (covers 2010–2014 births)
Inaction regret
Continuing to work through childbirth
25%
25% of pregnant Japanese employees experienced maternity harassment (matahara) at work — a documented harm proxy, not a self-reported regret rate
Pregnant Japanese employees (women working when their pregnancy was confirmed)
cross-sectional survey, May 2020 (Kachi et al. 2021)
% who regret this choice
Leaving employment at childbirthContinuing to work through childbirth
47%25%
action dominates — Action dominates — most regret acting.
Related decisions
Semantically similar decisions — same territory, different trade-offs.
Japan’s 15th National Fertility Survey (2015) found that 46.9% of women who worked before
childbirth left employment after their first child; the rate of continuing work after a first
birth was 38.3% in 2010–2014, up from earlier decades. This is a departure-prevalence figure, not
a self-reported regret rate — leaving is not the same as regretting leaving, and the survey does
not cleanly separate a positive preference for caregiving from structural coercion (rigid hours,
inflexible employers, inadequate childcare). The Gender Equality Bureau Cabinet Office documents
the long-run cost of leaving: women who re-enter the workforce after a childrearing break do so
predominantly as non-regular workers, who carry lower wages, limited training, and constrained
advancement; roughly 54% of female employment is non-regular, and Japan’s gender wage gap of 22%
was the highest in the G7 in 2021.
Among women who continue working, a peer-reviewed cross-sectional survey of 359 pregnant employees
(Kachi et al. 2021, Journal of Occupational Health) found that 24.8% had experienced matahara
(maternity harassment) by supervisors or colleagues — demotion, exclusion from promotion tracks,
or pressure to resign following pregnancy disclosure. Earlier government data put the figure at
~21.8% of full-time employees (MHLW 2015), consistent in magnitude. This too is a documented-harm
prevalence, not a regret survey: not all women who experience matahara regret continuing to work.
A separate longitudinal study of Japanese civil servants links work-family conflict to adverse
health outcomes, more strongly for women — evidence that the inaction path carries real costs
even for women who retain employment.
The bilateral pattern here is action-dominates — leaving employment is both the more prevalent
and the more durable outcome — driven by Japan’s specific M-curve institutional structure, not by
a universal regret asymmetry. The principal interpretive caveat is that neither rate is a
self-reported regret instrument: the 47% measures departure prevalence and the 25% measures
harassment prevalence, both honest proxies rather than regret. The M-curve has partially flattened
since the 2017 Womenomics policy reforms, and women entering the workforce in the mid-2020s face
somewhat different institutional conditions than those who made this decision before 2015. Japan’s
record-low fertility rate of 1.20 in 2023 partly reflects women opting out of parenthood entirely
as a response to the unresolved work-family conflict this entry documents — a third path not
captured by the binary framing of the decision.
Sources: action
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
1/2 sources independently verified verbatim against the cited source
[1]BMC Women's Health / National Institute of Population and Social Security Research (IPSS) — Factors associated with occupation changes after pregnancy/delivery (citing the 15th Japanese National Fertility Survey, 2015)
Verified
Government report
46.9% of Japanese women who worked before childbirth retired from work after the birth of their first child; the rate of continuing work after the first birth was 38.3% in 2010–2014 (15th Japanese National Fertility Survey, 2015)
Excerpt
“[Verbatim figures from the 15th Japanese National Fertility Survey, 2015, as quoted in this peer-reviewed paper.] "46.9% of the women who worked before childbirth retired from work after the birth of their first child"; "the rate of continuing work after the birth of the first child was 38.3% in 2010–2014." These figures from Japan's National Fertility Survey document the prevalence of leaving employment at first childbirth — the central action-side outcome — not a self-reported regret rate.
”
Source data from
2018-06-01
Accessed
2026-06-30
Verification
Excerpt independently re-fetched and confirmed word-for-word against the cited source during our grounding audit.
Calculation
Source replaced 2026-06-30: the previously cited JILPT 2020 survey figures (68% quit at marriage/childbirth, of whom 49% cited "structural push factors") could not be independently verified in any English- or Japanese-language JILPT publication during grounding review, and the 49% structural-vs-preference split appears nowhere in the primary literature. They are replaced with the verifiable 15th Japanese National Fertility Survey (2015) figure: 46.9% of women who worked before childbirth left employment after their first child (continuation rate 38.3%, 2010–2014). The action-side rate (0.469) is now the DEPARTURE-PREVALENCE construct — the share of working women who leave employment at first childbirth — explicitly labelled as such in regret_display, NOT a self-reported regret rate. No direct retrospective-regret instrument exists for women who left at childbirth; the prevalence is reported honestly rather than relabelled as regret.
[2]Gender Equality Bureau Cabinet Office, Japan — Women and Men in Japan 2020
Government report
The M-curve in female labour participation persisted from the 1970s through the 2010s; women who re-entered as non-regular workers earned substantially less than continuously employed regular workers
Excerpt
“[Paraphrase from published pamphlet — full statistical tables in linked PDFs.] The Gender Equality Bureau Cabinet Office 2020 edition of Women and Men in Japan documents the M-curve pattern in female labour force participation: a dip during childrearing years followed by re-entry predominantly in non-regular employment. As of 2021, approximately 54% of female employment is in non-regular positions, which carry lower wages, limited training, and constrained career advancement compared to regular employment tracks. Japan's gender wage gap of 22% was the highest among G7 countries in 2021, partly reflecting the wage penalty for women who re-enter as non-regular workers after a childrearing break.
”
Source data from
2020-01-01
Accessed
2026-05-14
Calculation
URL updated 2026-05-14: the originally cited 2022 edition does not exist on the Gender Equality Bureau website; the most recent available English edition is 2020. The 50-60% wage penalty figure cited in the original entry could not be verified in the English-language publication; the excerpt now reflects the documented 54% non-regular employment rate and 22% gender wage gap from corroborating sources (JCER, Brookings, IMF 2024). This source does not provide a direct regret rate but corroborates the direction and magnitude of the action-side figure.
Sources: inaction
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
1/2 sources independently verified verbatim against the cited source
[1]Journal of Occupational Health (Kachi, Inoue & Eguchi, 2021) — Association between maternity harassment and depression during pregnancy amid the COVID-19 state of emergency
Verified
Peer-reviewed
24.8% of pregnant Japanese employees had experienced maternity harassment (matahara) by supervisors and/or colleagues (cross-sectional survey of 359 pregnant employees, May 2020)
Excerpt
“[Verbatim from the abstract.] "A cross‐sectional Internet survey was conducted on 359 pregnant employees (including women who were working at the time their pregnancy was confirmed) from May 22 to May 31, 2020." "Overall, 24.8% of the pregnant employees had experienced maternity harassment by supervisors and/or colleagues." Matahara is defined in Japanese law as adverse treatment or harassment in the workplace against women based on pregnancy, childbirth, and requesting or taking childcare/family-care leave.
”
Source data from
2021-01-19
Accessed
2026-06-30
Verification
Excerpt independently re-fetched and confirmed word-for-word against the cited source during our grounding audit.
Calculation
Source replaced 2026-06-30: the previously cited MHLW landing page (reputable_reference) did not contain a verifiable 31% matahara prevalence figure — no MHLW publication anchored 31%, and the closest documented government figure is ~21.8% of full-time employees (MHLW 2015). The unverifiable 31% is replaced with the directly anchored, peer-reviewed Kachi et al. 2021 (Journal of Occupational Health) figure: 24.8% of pregnant Japanese employees experienced maternity harassment. The inaction-side rate (0.248) is the MATERNITY-HARASSMENT-PREVALENCE construct — a documented workplace-harm proxy for women who continue working through pregnancy — explicitly labelled as such in regret_display, NOT a self-reported regret rate. No direct retrospective-regret instrument exists for women who continued working; the harm prevalence is reported honestly rather than relabelled as regret.
[2]PubMed Central / International Journal of Environmental Research and Public Health — The health effects of work-family conflict in men and women Japanese civil servants: a longitudinal study
Peer-reviewed
Work-family conflict among Japanese women is significantly associated with adverse mental and physical health outcomes, driven by Japan's long working-hour culture and inadequate childcare infrastructure
Excerpt
“[Paraphrase from abstract — full text paywalled.] This longitudinal study of Japanese civil servants found that work-family conflict was significantly associated with adverse health outcomes, with women experiencing stronger effects than men, consistent with Japan's gendered care burden. Japan's long working-hour culture and limited affordable childcare access create structural conflict that compounds the burden on employed mothers. The study provides longitudinal evidence that work-family conflict has measurable health costs among Japanese women who remain employed through childrearing years.
”
Source data from
2020-01-01
Accessed
2026-05-14
Calculation
URL corrected 2026-05-14: PMC7474519 (the originally cited article) resolved to a paper about mayfly biology in a Costa Rican stream -- an unrelated article in an entirely different field. The PMC ID was fabricated. Replaced with PMC7557410, a real 2020 IJERPH longitudinal study of work-family conflict among Japanese workers (published Oct 2020). The substitute paper does not provide the exact "working mothers in Japan vs. OECD" systematic review framing of the original excerpt; that framing was removed. The substitute does corroborate the directional claim (high work-family conflict costs for employed Japanese women). This source does not independently anchor the inaction-side matahara prevalence (0.248, from Kachi et al. 2021); it supports the structural context for why employed mothers face real costs.
Caveats
Neither side of this bar is a self-reported regret rate — no direct retrospective-regret instrument exists for Japanese women who left (or stayed in) employment at childbirth, so both rates are documented proxies and are labelled as such. The action-side 47% is a DEPARTURE-PREVALENCE figure: the share of women who worked before childbirth and left employment after their first child (15th Japanese National Fertility Survey, 2015, covering 2010–2014 births). Leaving is not the same as regretting leaving; some departures are a positive preference for full-time caregiving and some are structurally coerced (rigid hours, inflexible employers, lack of childcare), but the survey does not cleanly split preference from coercion, so we do not assert a coercion share. The inaction-side 25% is a HARM-PREVALENCE figure: the share of pregnant employees who experienced maternity harassment (matahara), a documented workplace harm (demotion, exclusion from promotion, pressure to resign), not a regret survey — not all women who experienced matahara regret having continued to work, and it is measured over a COVID-era sample (May 2020, Kachi et al. 2021), which may differ from a typical year. Earlier government data put matahara at ~21.8% of full-time employees (MHLW 2015), consistent in magnitude. The M-curve has partially flattened since 2017 following the Abe administration's Womenomics policies (daycare expansion, equal opportunity enforcement, flexible working legislation); the continuation rate after first birth rose from ~28% to 38% across the 2000s, so the bilateral costs of this decision are lower for women entering the workforce in the mid-2020s than for those who made the decision before 2015. Japan's total fertility rate fell to 1.20 in 2023, a record low, suggesting that many women have responded to the unresolved work-family conflict this entry documents by opting out of parenthood entirely -- a third path not captured by this binary framing. The directional pattern (action-dominates, meaning leaving employment is the more prevalent and more durable outcome than staying) reflects the specific institutional conditions of Japan's M-curve era and may not generalise to other national contexts.
