We measure lifespan. We've started measuring healthspan. The forty-to-fifty-year working life sits between them. Others have named it; almost nobody has designed it. We inherited it from a life expectancy that no longer exists.
Lifespan is how long you live. Healthspan is how long you live well. Workspan is how long you work, and it's the only one of the three that is almost entirely a design decision: school leaving age, career ladders, retirement age, pension rules. Those decisions were made for a life that ended not long after the work did.
Switch the horizon and watch what the old design leaves behind.
Illustrative arithmetic: education to 22, retirement at 65, bars drawn to scale on a 150-year axis. The point is the shape, not the decimal.
Workspan isn't a new word, and it would be dishonest to present it as one. It sits in a family of "span" terms that people in ageing research have coined to measure what lifespan misses.
The gerontologists John Rowe and Robert Kahn are generally credited with first using "health span" in a 1987 article in Science, arguing that after the rise in lifespan, extending the years of full function should be the next goal.
The longevity and future-of-work writer Avivah Wittenberg-Cox, who works on the consequences of 60-year careers, has urged people to extend their workspans to match lengthening lifespans. A 2021 World Economic Forum piece on older workers called for policies to accommodate an extended workspan, and the US Health Action Alliance lists "extend workspan" next to "invest in healthspan" in its 2024 guidance for employers. Stanford's Center on Longevity makes the same argument in its New Map of Life work, which says longer lives will mean longer work spans.
Researchers keep adding spans. In late 2025 Alex Zhavoronkov and Dominika Wilczok proposed "peakspan", the age range in which someone keeps at least 90% of their peak performance in a given domain. Others have proposed "brainspan" for the years of sound cognitive function.
What this page adds is the framework: treating workspan as a design problem on par with lifespan and healthspan, the five pillars below, and the durability test.
Workspan breaks the problem into five parts. Each has evidence behind it.
Retirement ages, super, and career structures assume a three-stage life: learn, earn, stop. Some governments are already redesigning. Singapore raised its retirement age to 64 and its re-employment age to 69 on 1 July 2026, with a target of 65 and 70 by 2030.
In a 2021 survey of 604 Australian HR professionals and business leaders by the Australian HR Institute and the Australian Human Rights Commission, 46.7% said their organisation would be reluctant to recruit workers over a certain age. Sixty per cent said the departure of older workers had cost them key skills. Fewer than a third consulted older workers on issues that concerned them.
The 2023 edition of the same survey found one in six organisations would not consider hiring anyone over 65 in any circumstance. Yet many respondents who had actually employed older workers reported their performance, adaptability, energy and creativity were no different from younger colleagues.
Employers surveyed for the World Economic Forum's Future of Jobs Report 2025 expect 39% of workers' core skills to change by 2030. A single education at the start of a sixty-year working life can't cover that. Singapore's answer is a S$4,000 SkillsFuture top-up for citizens aged 40 and over, plus a training allowance of half their income, capped at S$3,000 a month for up to two years.
The idea that there's a fixed number of jobs, so one generation must leave for the next to enter, is old and wrong. Alicia Munnell and April Wu at Boston College's Center for Retirement Research studied US data from 1977 to 2011 and found no evidence of crowding out. A one-point rise in older workers' employment was associated with lower youth unemployment, higher youth employment and more hours worked.
One question sorts most skills: can the job be fully specified and scored? If you can write down exactly what good looks like and mark it, a machine can eventually be trained to do it. If you can't, because it depends on judgement, context and trust, the skill tends to get more valuable with time instead of less.
A long working life that stays on the left side of that line gets cheaper every year. A long working life that moves right gets more valuable. That movement is what a second-stage career is for.
The usual story is that the big bets belong to the young. The data says otherwise.
Economists from MIT, Northwestern and the US Census Bureau studied 2.7 million American founders. The average founder was 41.9. Among the fastest-growing one in a thousand new companies, the average founder was 45. A 50-year-old founder was about 1.8 times as likely to build one of those companies as a 30-year-old.
The strongest predictor wasn't youth. It was prior experience in the same industry, which roughly doubled the rate of top-tier success. That's compounding skill, cashed in.
The catch: it's one large US study, and founding a company is only one kind of second stage. But it demolishes the idea that the window closes at 35.
BMW expected the average age at its Dingolfing plant to rise from 39 to 47. Instead of pushing people out, it staffed a pilot gearbox line with 42 workers averaging 47, and asked them how to redesign it. They made about 70 small changes: wooden floors, adjustable chairs, special shoes, tilted screens, job rotation. The line's productivity rose 7% in a year, matching lines with younger workers, and absenteeism fell below the plant average.
The whole redesign cost tens of thousands of euros. It was initially mocked as the "pensioners' line".
The catch: one production line, one company, a decade and a half ago. It's the classic case because so few employers have published anything since.
Why the long plan isn't science fiction: the evidence file →