TL;DR. The career you are running is longer than the one your parents ran, and it keeps getting longer. Eurostat’s expected duration of working life for someone aged 15 in the EU rose from 32.7 years in 2004 to 37.5 years in 2025; in the Netherlands it is now 44.0 years and in Sweden 43.4. The OECD expects the average normal retirement age for men to rise from 64.4 years for those retiring now to 66.3 years for those starting their careers now. The asset that has to carry all of those years is your health, and it is not expanding at the same pace everywhere. When we set Eurostat’s 2024 working-life figures against its 2024 healthy life years, the gap between the two nearly closes in some countries: in the Netherlands, expected working life is 43.8 years against 44.5 expected years without activity limitation after age 15, a margin of 0.7 years; in Switzerland the margin is 0.2 years. Meanwhile the sprint has a measured price. WHO and ILO estimate that long working hours of 55 or more a week were linked to 745,000 deaths from stroke and ischaemic heart disease in 2016, with a 35% higher risk of stroke and a 17% higher risk of dying from ischaemic heart disease compared with a 35-40 hour week. Long hours are now rare among EU employees, but still reported by 38.5% of self-employed people with staff and 20.0% of managers. The implication is not “work less”. It is “stop treating the body as a consumable in a race that is now four decades long”.
The wrong unit of time
Most performance advice is written for a quarter, a launch, a promotion cycle, or a four-year stretch at a startup. That unit made sense when a career was a shorter arc with a clear end. It is increasingly the wrong unit.
The question worth asking is not “how much can I produce this year” but “what do I need to still be producing well in year 30”. Those two questions frequently have opposite answers. The first rewards spending down capacity. The second rewards protecting it.
To see why this matters now, look at what has happened to the length of the race.
The race got longer: verified data
Verified data: expected duration of working life, 2004 to 2025
| Country | 2004 | 2014 | 2025 | Change 2004 to 2025 |
|---|---|---|---|---|
| EU27 | 32.7 | 34.8 | 37.5 | +4.8 years |
| Netherlands | 37.2 | 39.6 | 44.0 | +6.8 years |
| Sweden | 38.4 | 41.1 | 43.4 | +5.0 years |
| Denmark | 38.9 | 38.5 | 42.6 | +3.7 years |
| Germany | 34.9 | 38.0 | 40.2 | +5.3 years |
| France | 32.7 | 34.7 | 37.5 | +4.8 years |
| Hungary | 28.0 | 31.8 | 37.5 | +9.5 years |
| Spain | 32.1 | 34.8 | 36.8 | +4.7 years |
| Poland | 30.3 | 32.6 | 35.9 | +5.6 years |
| Italy | 29.9 | 30.7 | 33.0 | +3.1 years |
| Romania | 31.8 | 32.8 | 32.7 | +0.9 years |
| Switzerland | 40.1 | 41.9 | 42.8 | +2.7 years |
| Norway | 38.4 | 39.6 | 41.4 | +3.0 years |
Source: Eurostat, dataset lfsi_dwl_a, duration of working life, total, in years. The indicator estimates how many years a person currently aged 15 is expected to be active in the labour market, meaning employed or unemployed, over their lifetime. Changes computed by CEOtudent from the reported values.
Two independent signals point the same way. The OECD’s Pensions at a Glance 2023 reports that the average normal retirement age across member countries will rise from 64.4 years for men retiring now to 66.3 years for men who start their careers now, and from 63.6 to 65.8 years for women. That is roughly two extra years at the end of a career already about five years longer than it was two decades ago in the EU.
Put together, a person entering work in the mid-2020s in much of Europe should plan for something close to four decades in the labour market. The four-year sprint is one tenth of that.
The binding constraint: healthy years, not working years
Longer careers are only half the picture. The other half is how many of those years the body can carry without limitation.
Eurostat publishes healthy life years: the number of years a person is expected to live without long-term limitation in usual activities, based on self-reported limitation in its household survey. Its working-life series and its healthy-life-years series are published separately. We joined them.
CEOtudent analysis: how close working life comes to healthy life, 2024
| Country | Expected working life, 2024 | Healthy life years at birth, 2024 | Healthy years after age 15 (lower bound) | Margin: healthy years after 15 minus working life | Working life as share of healthy years after 15 |
|---|---|---|---|---|---|
| Switzerland | 42.8 | 58.0 | 43.0 | 0.2 | 99.5% |
| Netherlands | 43.8 | 59.5 | 44.5 | 0.7 | 98.4% |
| Denmark | 42.6 | 62.4 (b) | 47.4 | 4.8 | 89.9% |
| Germany | 40.0 | 62.4 | 47.4 | 7.4 | 84.4% |
| Norway | 41.2 | 64.8 | 49.8 | 8.6 | 82.7% |
| Sweden | 43.1 | 67.5 | 52.5 | 9.4 | 82.1% |
| Hungary | 37.5 | 63.3 | 48.3 | 10.8 | 77.6% |
| France | 37.2 | 63.6 (p) | 48.6 | 11.4 | 76.5% |
| EU27 | 37.2 | 65.2 | 50.2 | 13.0 | 74.1% |
| Poland | 35.6 | 64.7 | 49.7 | 14.1 | 71.6% |
| Spain | 36.5 | 68.6 (b) | 53.6 | 17.1 | 68.1% |
| Romania | 32.7 | 65.0 (b, e) | 50.0 | 17.3 | 65.4% |
| Bulgaria | 34.8 | 69.1 | 54.1 | 19.3 | 64.3% |
| Greece | 34.8 | 69.9 | 54.9 | 20.1 | 63.4% |
| Italy | 32.9 | 73.4 (b) | 58.4 | 25.5 | 56.3% |
Sources: Eurostat lfsi_dwl_a (duration of working life, total) and Eurostat hlth_hlye (healthy life years at birth, total), both 2024. Flags from Eurostat: b = break in series, e = estimated, p = provisional. Join, subtraction and ratio computed by CEOtudent.
How to read this table, and what it does not say:
- Why subtract 15. The working-life indicator starts counting at age 15; healthy life years count from birth. Subtracting 15 assumes every year before 15 is a healthy year. Because some children do have activity limitations, the true number of healthy years after 15 can only be equal to or larger than this figure, so the margin shown is a lower bound.
- Both are period measures. Neither predicts what will happen to a specific person. Each describes the pattern implied by current conditions.
- Activity limitation is not the same as inability to work. Many people with a limitation work productively. The table says the two expected spans are converging, not that Dutch or Swiss workers will spend their final working years unable to work.
- Healthy life years are self-reported and sensitive to how people in each country answer the survey question. Cross-country differences, and the flagged series breaks for Denmark, Spain, Italy and Romania in 2024, call for caution. Italy’s high figure in particular sits on a break.
With those limits stated, the pattern is still the most useful single fact in this piece. In the countries where careers have stretched furthest, the expected working span has grown to the edge of the expected span of life without limitation. There is little slack left. The margin is not something you can count on; it is something you have to build.
At the other end of the career, the same Eurostat dataset shows a similar pressure. In the EU in 2024, a 65-year-old could expect 20.2 more years of life, of which 10.3 are expected to be healthy life years. The remaining 9.9 years, a CEOtudent calculation from the two reported figures, are expected to be lived with some activity limitation.
What the sprint actually costs
If careers are long and healthy capacity is the constraint, the relevant question about any sprint is how much of that capacity it spends.
The best-documented cost is long working hours. In 2021 WHO and ILO published the first global estimate of the burden, in a systematic analysis by Frank Pega and colleagues in Environment International.
Verified data: the measured price of long hours, and who still works them
| Measure | Value | Source |
|---|---|---|
| People exposed to working 55 or more hours a week, 2016 | 488 million, 8.9% of the global population | Pega et al., Environment International, 2021 |
| Deaths from stroke and ischaemic heart disease attributable to that exposure, 2016 | 745,194 | Pega et al., 2021 |
| Increase in attributable deaths, 2000 to 2016 | 29% | WHO news release, 17 May 2021 |
| Risk of stroke, 55+ hours versus 35-40 hours a week | 35% higher | WHO news release, 17 May 2021 |
| Risk of dying from ischaemic heart disease, 55+ hours versus 35-40 hours a week | 17% higher | WHO news release, 17 May 2021 |
| EU27 employed people usually working 49+ hours a week, 2004 and 2025 | 11.6% and 6.3% | Eurostat lfsa_qoe_3a2, ages 20-64 |
| EU27 employees working 49+ hours, 2025 | 3.2% | Eurostat lfsa_qoe_3a2 |
| EU27 self-employed with employees working 49+ hours, 2025 | 38.5% | Eurostat lfsa_qoe_3a2 |
| EU27 managers working 49+ hours, 2025 | 20.0% | Eurostat lfsa_qoe_3a2 |
Notes: the stroke estimate refers to the risk of having a stroke; the ischaemic heart disease estimate refers to the risk of dying from it. The Eurostat series has breaks in 2005 and 2021, and the managers series also in 2011, so the 2004 to 2025 comparison indicates direction rather than a precise change.
The Eurostat rows matter as much as the WHO rows. Across the EU, the long-hours week has become uncommon for employees. It has not disappeared; it has concentrated. The people most likely to still run a sustained sprint are those who own a business and those who manage others, which is to say the people most likely to be reading advice about performance.
We have covered what recovery restores and how to structure it in The Recovery Protocol, and why depleted capacity is a design failure rather than a character failure in Burnout Is a Systems Failure. This piece is about the layer above both: deciding what to spend and what to protect over the whole arc.
The 40-Year Ledger
A company that runs for decades does not judge a year by output alone. It separates spending that builds the asset base from spending that runs it down. A career can be managed the same way.
CEOtudent editorial framework: the 40-Year Ledger
| Account | Sprint withdrawal (what a 4-year optimization spends) | Compounding deposit (what a 40-year optimization builds) | Early warning signal | Operating rule |
|---|---|---|---|---|
| Cardiovascular and physical capacity | Sustained 55+ hour weeks, inactivity, skipped training during crunch periods | Regular aerobic and strength training kept through busy periods | Training is the first thing cut every time work gets busy | Movement is a fixed cost, not a discretionary one. See the dose-response data for desk workers |
| Sleep and recovery | Chronic short sleep, recovery postponed to “after the launch” | Protected sleep window, scheduled recovery built into the calendar | Weekend recovery no longer resets the week | Recovery is booked before the sprint starts, not after it ends. See sleep architecture and cognitive performance |
| Skills | Harvesting current skills at maximum rate, no learning time | Deliberate reskilling cycles across the career | Your most valuable skill is the one you learned longest ago | Budget a learning cycle every few years. See Reskilling After 40 |
| Career capital | Title and pay optimised, transferable assets neglected | Reputation, judgement and relationships that travel between roles | You could not describe what you have built that would survive a job change | Trade some short-term pay for durable capital. See Career Capital in the AI Era |
| Optionality | Lifestyle expands to match every raise; no runway | Financial runway and open options | You cannot afford to slow down even if your health requires it | The ability to stop is itself a performance asset. See Optionality as a Career Strategy |
| Attention | Always-on availability as a signal of commitment | Deep work blocks and defended focus | Every hour is reactive | Availability is not output; protect the hours that compound |
This framework is CEOtudent’s editorial synthesis, not the output of a study. It uses the verified data above as its premise: the career is long, healthy capacity is the constraint, and the documented costs of sustained overwork land on precisely the accounts a sprint draws down.
When a sprint is the right call
None of this means never sprint. Some windows reward intensity: a launch, a funding round, a short project with an asymmetric payoff, the first months in a role that will define the next decade. The ledger does not forbid withdrawals. It demands that they be deliberate.
A sprint is defensible when it passes all five of these tests:
- It has an end date you set in advance. “Until things calm down” is not an end date.
- The payoff is asymmetric. The upside meaningfully outweighs the cost to the accounts above, and would not be available later.
- Recovery is scheduled before the sprint starts. If recovery is left for “after”, it rarely happens.
- The cost is reversible. Lost sleep for a few weeks recovers; a pattern of years does not recover on the same terms.
- It is an exception, not a pattern. If you have sprinted in each of the last several quarters, you are not sprinting. You are operating at an unsustainable baseline and calling it a sprint.
These tests are an editorial decision rule, not a clinical threshold.
The CEO and the student over forty years
The CEO view. A chief executive who ran a company’s key asset to failure in pursuit of one strong year would be judged to have failed, whatever that year’s result. Treat your own capacity the same way. The job is capital allocation across a long horizon: deciding what to spend now, what to protect, and what to invest so the asset base is larger in ten years than it is today.
The student view. A forty-year career is not one career. It is several, joined by periods of relearning. The longer the race, the more of those cycles you will run, and each one requires spare capacity: time, energy and health that a permanent sprint has already spent. Staying the student is only possible if the CEO has kept something in reserve.
Frequently asked questions
Is a 40-year career realistic?
For many people entering work in Europe now, it is close to the expected figure. Eurostat’s expected duration of working life was 37.5 years in the EU in 2025 and above 43 years in the Netherlands and Sweden. The OECD projects rising normal retirement ages for people starting work today.
What does “healthy life years” measure?
It is Eurostat’s estimate of how many years a person can expect to live without long-term limitation in usual activities, based on self-reported limitation. It is not a medical diagnosis and it does not mean a person with a limitation cannot work.
Are long hours always harmful?
The WHO and ILO estimate links sustained weeks of 55 hours or more to a 35% higher risk of stroke and a 17% higher risk of dying from ischaemic heart disease compared with 35-40 hours. Those are population-level associations for sustained exposure, not a verdict on any single intense week. The practical lesson is that the pattern, not the occasional burst, carries the documented risk.
Who is most likely to overwork in Europe today?
In 2025, 38.5% of self-employed people with employees and 20.0% of managers in the EU usually worked 49 hours or more a week, against 3.2% of employees.
Why does the table stop at 2024?
Eurostat’s healthy life years series currently ends in 2024, so the join uses 2024 for both datasets to compare like with like.
What is the single most useful change to make?
Move recovery and training from discretionary to fixed. Every other rule in the ledger is easier once those two stop being the first things cut when work gets busy.
Sources
Eurostat, Duration of working life, annual data, dataset lfsi_dwl_a, and its reference metadata, retrieved September 2026.
Eurostat, Healthy life years by sex (from 2004 onwards), dataset hlth_hlye, retrieved September 2026.
Eurostat, Employed persons working usually long hours by sex, age, professional status and occupation, dataset lfsa_qoe_3a2, retrieved September 2026.
OECD, Pensions at a Glance 2023: OECD and G20 Indicators, OECD Publishing, 2023.
World Health Organization, Long working hours increasing deaths from heart disease and stroke: WHO, ILO, news release, 17 May 2021.
Frank Pega and colleagues, Global, regional, and national burdens of ischemic heart disease and stroke attributable to exposure to long working hours for 194 countries, 2000-2016: A systematic analysis from the WHO/ILO Joint Estimates of the Work-related Burden of Disease and Injury, Environment International, volume 154, article 106595, 2021.
This content was compiled with the support of AI following in-depth research, then written and prepared for publication by the CEOtudent editorial team.
This post is also available in:
