The most common version of phased retirement is a polite fiction: the same job, the same meetings, the same inbox, at eighty percent of the hours and eighty percent of the pay. Six months in, the person is working full weeks again, now at a discount, and both sides quietly conclude that "phasing doesn't work."
What failed was not the phase. It was the assumption that a role can be reduced by decree, without redesigning what the role contains.
What the working-time evidence actually says
The best recent evidence on reducing working time comes from an adjacent field: the four-day week trials. In 2025, Fan, Schor and colleagues published in Nature Human Behaviour the largest intervention study to date, covering 2,896 employees across 141 organizations in six countries over six months. Well-being improved on burnout, job satisfaction, and mental and physical health, a pattern absent in comparison companies. Two findings matter directly for phased retirement.
First, there was a dose-response relationship at the individual level: the larger a worker's own reduction in hours, the larger the well-being gain. Reduction is not symbolic; the hours themselves carry the effect.
Second, and more important, the intervention was never just cutting a day. Organizations reorganized the work before the trial, trimming low-value activity so that the reduced schedule held. The measured benefits reflect the package of redesign plus reduction, not reduction alone. The authors are explicit about this, and it is precisely the step that failed phased-retirement schemes skip. (The study is quasi-experimental, with self-selected firms and non-randomized controls, so the numbers deserve the usual caution; the design lesson survives the caveat.)
Translated to the last decade of a career: a phase that removes hours but keeps every accountability is a pay cut with extra steps. A phase that removes accountabilities, hands them to named successors, and concentrates the remaining time on high-judgment work is a different job, and that is the version worth defending.
The retention math, and the training gap
Why should an employer bother? The OECD's Employment Outlook 2025 gives the macro answer: member countries that reduced the exit rate of older workers to the level of the best-performing decile would gain at least 0.2 percentage points of annual per capita GDP growth, in about half of the OECD. Inside a single firm the same logic applies at smaller scale. Every senior person who leaves on a cliff takes context that has no backup, and replacement markets for thirty-year veterans do not exist.
But the same report exposes the habit that undermines phased arrangements: organizations stop investing in people they expect to leave. Only about one third of adults aged 60 to 65 participated in training in 2023, against more than half of those aged 25 to 44. A phase treated as a runway to the exit, with no learning attached, becomes self-fulfilling: the person's tools age, their scope shrinks for the wrong reasons, and the arrangement confirms the stereotype that funded it. The OECD's blunt addition is that negative stereotypes about aging harm firm performance, and it recommends age-inclusive hiring and training as an economic measure, not a courtesy.
Practice: designing the phase
The workable pattern has three properties.
Named subtraction. List what the role stops doing, and who picks each piece up. "Do the same with less time" is not a phase; it is a squeeze. The four-day-week organizations redesigned first and reduced second; the order matters.
A real remaining job. The retained days should hold the work where thirty years of pattern recognition beats any alternative: the hardest client conversations, the judgment calls, the apprenticeship of successors. If the remaining job could be done by anyone, the phase is severance in slow motion.
Continued investment. Training, new tools, inclusion in what is changing. The 60-to-65 training gap is a choice organizations make, and phased workers feel it within weeks. A person being phased out of investment will phase themselves out of engagement to match.
Phased retirement is often sold as a kindness to the departing. The evidence suggests it is closer to an operating discipline: a forced, explicit decomposition of a senior role that the organization should have written down years earlier. The kindness is real. The documentation is the point.
Sources
- Fan, Schor, Kelly & Gu (2025). Work time reduction via a four-day workweek (Nature Human Behaviour, DOI 10.1038/s41562-025-02259-6). Six-month trial, 2,896 employees in 141 organizations across six countries; improved burnout, satisfaction, mental and physical health; individual-level dose-response between hours reduced and well-being; effects reflect work redesign plus reduction, not reduction alone. https://www.nature.com/articles/s41562-025-02259-6
- OECD (2025). OECD Employment Outlook 2025. Cutting older-worker exit rates to best-decile levels worth at least 0.2pp annual per capita GDP growth in about half of OECD countries; only ~1/3 of adults 60-65 trained in 2023 vs more than half of 25-44; negative age stereotypes harm firm performance. https://www.oecd.org/en/publications/oecd-employment-outlook-2025_194a947b-en.html