Goldman Sachs trademarked the term "Returnship" in 2008. The program was designed for what the firm called a "non-traditional cohort" — typically women who had stepped off the finance ladder for childcare and were trying to return — and offered a 10-week paid stint with the possibility of a permanent offer. The program produced visibly successful placements and gave the model legitimacy. Seventeen years later, the corporate returnship has become a documented HR product category: iRelaunch maintains a directory of more than 250 employer programs; Path Forward, the largest U.S. returnship nonprofit, has placed several thousand returners since 2016; and roughly half of the largest U.S. employers run some form of structured re-entry program.
The argument here is that the returnship model works — the placement data, the wage data, and the retention data all support it — but its current scale is much smaller than the addressable population, and the structural reforms that would meaningfully expand re-entry options (caregiver tax credits, federal paid family leave, employer cost-sharing for re-entry training, salary-history-ban enforcement) have advanced more slowly than the corporate-program model has. The returnship works as a corporate-program category; it is not yet a workforce-policy category.
The scale of the addressable population
The Census Bureau's Current Population Survey and supplemental work by the Bureau of Labor Statistics consistently identify a large U.S. workforce population that has been out of paid employment for a year or more and would prefer to be working: roughly 5–7 million workers depending on the measure, with a heavy concentration of college-educated women in their 30s and 40s.
The Sylvia Ann Hewlett research at the Center for Talent Innovation, including her co-authored 2010 book Off-Ramps and On-Ramps, was the first to quantify the off-ramp pattern at scale. The Hewlett survey work found that 37% of highly qualified women had voluntarily off-ramped at some point in their careers, the average off-ramp duration was about 2.2 years, and the majority of off-rampers reported difficulty returning at salary levels comparable to their pre-departure roles. Subsequent work by McKinsey, LeanIn.org, and Catalyst has tracked the pattern through the post-2020 period, where the pandemic produced a measurable additional spike in women's labor-force exit.
The Goldin Career and Family (2021) framing identifies the structural mechanism: "greedy work" — careers that demand unpredictable and excessive hours and that penalize flexibility — produces the off-ramp pressure that returnship programs then attempt to remediate. The returnship is, in this framing, a downstream patch for an upstream design problem.
What returnship programs actually produce
The empirical evidence on returnship outcomes, drawn primarily from program-administered data and the iRelaunch tracking, supports several findings.
Conversion rates from returnship to permanent role typically run between 60% and 85% depending on the employer, with the higher end among firms that designed the program with clear permanent-hire intent. Goldman Sachs' own published numbers fall in this range; Microsoft's LEAP Engineering Acceleration Program (originally a returnship-adjacent re-entry program) has had similarly high conversion rates.
Wage outcomes for returnship completers are typically below their pre-departure compensation but within 10–15% of comparable peers — a meaningful improvement over the typical off-ramper's return wage, which the older Hewlett data and subsequent research found could be 30–40% below pre-departure levels.
Retention rates for returnship hires are documented as comparable to or modestly better than peer hires in most evaluated programs. The MetLife and JPMorgan Chase returnship programs have both published retention data supporting this.
The Wharton Center for Human Resources research, including work by Peter Cappelli and others, has identified the structural features that distinguish high-performing returnship programs: paid (not unpaid) participation, clear and quantified hiring conversion targets, structured re-onboarding curricula that update technical skills, sponsorship rather than just mentorship pairings, and integration with employee-resource groups for ongoing support.
What the model does not solve
The returnship model has visible limits, and being honest about them is important.
First, eligibility is structurally biased toward college-educated workers with strong prior corporate-career histories. The vast majority of advertised returnship slots target professional-services, tech, finance, consulting, and engineering roles. The model has not extended to the workers who would benefit most from structured re-entry — long-unemployed workers, lower-wage workers, workers with disability-related employment gaps, and workers returning from incarceration. The Vera Institute's research on reentry from incarceration documents how thin the structured-program landscape is for that population.
Second, the program scale is small relative to the addressable need. Even at 250+ employer programs with thousands of slots annually, the total returnship placements per year is in the low five figures. Against a five-to-seven-million addressable population, the model is a niche solution.
Third, returnships do not address the structural mechanism that produces the off-ramp in the first place. The greedy-work structure of senior professional careers, the absence of universal paid family leave, the high cost of childcare, and the absence of caregiver-friendly career-track design are all upstream. The returnship is the airbag, not the seatbelt.
The policy reforms that would matter
The federal and state policy reforms that would scale returner outcomes are the same ones that would close the gender wage gap and the motherhood penalty, and they are well-studied.
Universal paid family leave is the highest-leverage single reform. The U.S. is one of two OECD countries without a national paid-family-leave entitlement; the OECD Family Database puts the OECD average at 39 weeks of paid leave with substantial variation. The empirical literature on California's 2004 paid family leave law (now followed by similar laws in nine other states and DC), including work by Maya Rossin-Slater and Christopher Ruhm, finds positive effects on women's labor-force attachment and on infant and maternal health.
Subsidized childcare expansion. The Hamilton Project and Brookings analyses, including work by Diane Whitmore Schanzenbach, consistently find that the U.S. underinvests in early-childhood care and education relative to OECD peers, and that childcare cost is a primary driver of women's labor-force exit.
Salary-history-ban enforcement and pay-transparency expansion. As noted in our salary-negotiation coverage, the Massachusetts 2018 evaluation by Jesse Rothstein and others found measurable wage-gap narrowing for job-changers, with the largest effects for candidates with non-linear histories.
Caregiver tax credits and Social Security caregiver credits, which would compensate caregivers for the foregone wages and Social Security contributions during their off-ramp.
For the broader treatment of the structural reforms that would make modern careers compatible with care work, see our flagship Women, Work, and the Future → and The Caregiver Workforce →.
The returnship is the airbag, not the seatbelt. It works as a corporate-program category. The structural reforms that would prevent the off-ramp in the first place have moved more slowly than the corporate-program model has.
The returnship has earned its place in the corporate HR toolkit. The empirical record on conversion rates, wage outcomes, and retention is strong enough that programs are now standard at most large U.S. employers and many mid-sized ones. The honest evaluation is also that the model is small relative to the addressable population, structurally biased toward already-advantaged workers, and downstream of the policy choices that produce the off-ramp. The returnship revolution, taken at face value as the corporate marketing presents it, is real. As a substitute for the broader structural agenda — paid leave, subsidized childcare, salary-history bans, caregiver compensation — it is insufficient. Both can be true.
Updated May 21, 2026. This piece was substantively rewritten as part of NWLB's 2026 editorial refresh.
