Most early-retirement arithmetic treats your wealth as one pot: add everything up, divide by a withdrawal rate, read off a date. In Europe that fiction breaks on a legal fact: a large share of most people's wealth sits in a pension that cannot be touched until an access age. If you want to stop working at 50 and your pension opens at 65, the fifteen years in between are the bridge, and the bridge can only be paid from money you can actually reach.
The usual advice stops at "make sure you have accessible savings too". We can do better than advice: the free calculator on this site models a locked pension directly, so the question "when does the lock actually change my date" has a computable answer. We ran it six ways. The results are less obvious than the advice.
The short version
- A pension lock does not delay your retirement date by itself. It binds only when the accessible side of your wealth is too thin to cover the bridge years on its own.
- While you are still saving with a decade or more of runway, even a badly skewed split often changes nothing: your new savings land on the accessible side and quietly build the bridge for you.
- Close to the finish line it is a different story. In our 50-year-old scenario, moving €500,000 of €700,000 behind a lock pushed the earliest stop date back by more than three years, with identical total wealth.
- Once the bridge is the constraint, income that starts after the pension opens changes nothing. A state pension makes the plan richer later, not possible sooner.
Two pots, one plan
The access age depends on where you are and what kind of scheme it is. Typical cases for private and occupational pots: 57 in the UK from April 2028, 62 for newer private contracts in Germany, the legal retirement age for a French PER (64) and for Belgian supplementary pensions (66 now, 67 from 2030). State pensions arrive later still and are income rather than a pot. Every scheme has its own small print, so check yours; the point here is only that the age exists and is far from 50.
That age splits an early retirement into two phases. From your stop date to the unlock, spending comes entirely out of accessible money: brokerage accounts, funds, cash, whatever you can sell. From the unlock onward, both pots are available. So the plan has to pass two tests, not one: the total has to be big enough for the whole retirement, and the accessible side has to be big enough for the bridge. The second test is the one a single FI number never sees.
The calculator runs both tests every month of the plan. Its Could stop tile reports the earliest month the whole plan survives, locked pot and all, so the numbers below are read straight off it.
A 35-year-old, still saving: the lock costs nothing yet
Take someone born in 1991 with €200,000 saved, netting €4,200 a month, spending €2,600 of it, aiming to spend €2,800 a month (in today's money) once retired. Withdrawal rate 3.5%, inflation 2%, stocks at 7% nominal. To keep the comparison clean we gave the pension the same 7% growth (the calculator's default for pensions is 5%), sent all new saving to the accessible side, and used the calculator's default realistic tax settings (10% capital gains over a €10,000 annual exemption, 30% on yield). The only thing that changes between the three runs is where the existing €200,000 sits; the pension, where there is one, unlocks in January 2056, the year this person turns 65.
| Scenario | Could stop | Stop-now progress |
|---|---|---|
| All €200,000 accessible | September 2038, at 47 | 30% |
| €130,000 accessible, €70,000 locked | September 2038, at 47 | 25% |
| €70,000 accessible, €130,000 locked | September 2038, at 47 | 13% |
The date does not move. Not by a month, even in the version where two thirds of the wealth is untouchable for thirty years. Twelve more years of saving €1,600 a month into accessible stocks builds a bridge fund big enough for the gap between 47 and 65 in every split, so the lock never becomes the binding constraint.
What does move is the Stop-now progress tile: how much of a stop-today plan is already funded. 30% against 13% is the honest difference between those two people today. If the saving stopped, if the job disappeared, the pension-heavy version is much further from safety than the identical total suggests. The lock is not costing this person time yet. It is costing them optionality.
A 50-year-old, nearly there: the lock costs three years
Now someone born in 1976: €700,000 saved, netting €4,500 a month, spending €3,200, wanting €3,000 a month in retirement. Same growth and tax settings as above; the pension, where there is one, unlocks in January 2041, at 65. Again the only difference between the two runs is the split.
| Scenario | Could stop | Stop-now progress |
|---|---|---|
| All €700,000 accessible | March 2027, at 51 | 97% |
| €200,000 accessible, €500,000 locked | May 2030, at 54 | 54% |
Same €700,000, same salary, same spending. The all-accessible version can stop next spring. The pension-heavy version works three more years, because €200,000 plus a few years of saving has to carry every month of spending from the stop date to January 2041 unaided, and it cannot do that from 51. The plan is not short of money. It is short of reachable money, which for the bridge years is the only kind that counts.
This is the same person the one-pot arithmetic congratulates. Total wealth divided by a withdrawal rate gives identical answers for both rows of that table. The three-year gap between them is invisible to the FI-number way of thinking, and it is real.
The state pension does not rescue the bridge
The obvious objection: this 50-year-old will also get a state pension. True, so we added one: €1,200 a month starting January 2043, at 67. One input added to the previous scenario, nothing else changed.
The earliest stop date stays May 2030. The end-of-horizon balance grows by more than three million euros, and the date does not move a single month.
That is the bridge logic working as it must. The constraint on this plan is the years from 54 to 65, and income that begins at 67 cannot pay for a single one of them. Once the bridge binds, everything that happens after the unlock is already good enough, and making it better is irrelevant. The only levers that move the date are the ones that act on the bridge itself: more accessible saving, lower spending, or an earlier access age if your scheme allows one.
Where the next euro goes
None of this says pensions are a bad deal. A pension contribution is usually made from pre-tax salary, which makes it the cheapest euro of investing available to most Europeans, and in these scenarios the locked pot compounds untouched the whole time and funds the decades after 65 handsomely. (The calculator's salary field is net take-home, so when you model your own version of this, remember that a sacrificed euro costs you less than a euro of net savings; the playground guide covers the mechanics.)
What it does say: once early retirement is the goal, the split starts to matter as much as the total, and it matters more the closer you get. Far out, tax efficiency wins and the bridge builds itself. Close in, every euro added to the locked side makes the after-years richer without making the stop date earlier, and someone chasing tax relief hard can end up pension-rich and bridge-poor: congratulated by their FI number, unable to act on it. Where the line sits for you is exactly what the runs above compute, and your own version is one scenario link away.
We wrote about the withdrawal-rate half of this problem, including what European taxes do to the 4% rule, in an earlier piece; its pension section reached the same conclusion from a different angle.
What this leaves out
- Pension withdrawals are usually taxed as income, and this model does not do that. The calculator applies one capital-gains regime to all withdrawals. In reality UK pension drawdown beyond the tax-free quarter is income-taxed, and most continental schemes tax payouts one way or another, so the after-unlock need is somewhat higher than the chart shows. The bridge-side arithmetic, which is the subject here, is unaffected. The modelled state pension income is likewise not income-taxed.
- We set pension growth equal to stocks on purpose, so access would be the only difference between the runs. Real pension funds often hold more bonds and grow slower, which makes a locked-heavy split somewhat worse than shown here, not better.
- One expected path is not a guarantee. These are straight-line runs at 7% nominal against 2% inflation. The calculator's Monte Carlo mode runs 500 simulated markets around the same plan and reports how often it survives.
- Schemes have small print. Early-access provisions, protected ages, transfer rules and the tax treatment of taking money early vary by country and contract. The unlock ages above are the typical cases, not yours until you have checked.
As always: these are estimates from your own assumptions, for planning. Not financial advice.