Coast FIRE Isn't a Permission Slip. For Us It's a Deadline — and We Won't Downshift When We Hit It.

Every January my wife and I have the one money meeting that actually decides anything. Last January she was pregnant, and we spent most of that evening on a number I had been avoiding.

Not our FIRE number. Our coast number, which the internet calls coast FIRE: the point where what's already invested compounds to the target on its own, and everything we add after that is optional.

I went in expecting a boring arithmetic evening. It turned into the most eye-opening conversation we have ever had about money, and it rearranged how we look at the next twenty years. Ours lands around 40, six to eight years out. And before anything else: I haven't crossed it. This is not a milestone post. It's me writing down what I intend to do on a day that hasn't arrived yet, while saying it is still cheap.

Because the internet already has a plan for that day. You hit coast, you quit the stressful job, you go part-time or seasonal or barista, you buy your freedom back while the kid is small. Nick Wolny writes the cleanest coast FIRE explainer out there, and even he frames the payoff as a chance to "downshift to a less stressful career." Anders Skagerberg hit coast at 29, had a daughter the same year, moved to a flexible seasonal-and-freelance setup, and wrote ten good reasons why coast is the best flavour of FIRE. Same trigger as mine. Opposite answer.

I like my job a great deal. When we hit coast I don't plan to leave it and I don't plan to stop saving. I plan to buy a four-day week and carry on.

Coast is the most useful number in a FIRE plan and the most misread one. It's a floor, not a finish line — the month your plan stops being fragile, not the month you're allowed to walk out.

The coast FIRE number everybody calculates and nobody defends

The mechanics are genuinely simple: take your FIRE number, discount it back by however many years you have left, and what falls out is the pile that gets you there on autopilot. Wolny's worked example has a 25-year-old who wants $50,000 a year and retires at 60 needing $1.25M eventually, but only about $390k by 40 to coast into it.

My problem with every one of those pages, including the good ones, is that the exponent is doing all the work and almost nobody sources it.

I went looking. WalletBurst assumes 4% real. Wolny uses 6% and calls it slightly conservative. Nick Maggiulli 4%, White Coat Investor 5%, firenum 5%. None of them defends the choice; firenum is the only page that even gestures at a dataset. So the whole genre rests on a number picked by feel, and a coast target built on 4% real is a different universe from one built on 6%.

The best answer I could find is the Dimson-Marsh-Staunton data behind the UBS Global Investment Returns Yearbook: world equities returned 5.2% a year in real terms over the 125 years from 1900 to 2024, across 35 markets. Bonds 1.7%. Bills 0.5%. As a modern cross-check, MSCI World has compounded at roughly 9.1% a year nominal since the end of 1987, which nets out somewhere above 6% real — except that's a gross index with no withholding tax and no fund fee, so it's a ceiling, not a plan.

I use 5%. It sits below the century figure, well below the US-only figure, and if it's wrong it's wrong in the direction I can live through.

What the worst two decades on record do to a coast FIRE plan

Averages are the wrong tool here anyway, and this is where the genre goes quiet.

Every coast FIRE page has a risks section. Four bullets, usually: markets might underperform, expenses are unpredictable, you might lose your job, lifestyle creep. All true, all weightless, because not one of them tells you how bad "underperform" actually gets.

So I pulled Robert Shiller's dataset and worked it out: rolling 20-year real total returns on US stocks, 1871 to 2024.

Data: Robert J. Shiller, US equity real total returns 1871–2024.

The median is 6.8% a year, comfortably above my planning assumption. But 22% of those windows came in under 4% real, and the worst of them, mid-1901 to mid-1921, annualized slightly below zero across the whole twenty years. Second worst, 1962 to 1982, managed 0.35%.

Now run that against a coast number. Somebody who crosses coast at 40, stops contributing and gets handed the 1901 sequence arrives at 67 with about 35% of what they planned for. Not 90%. Thirty-five.

That is the whole reason I can't read coast as a permission slip. The number assumes an average, and then you remove your only remaining lever right before the sequence gets a vote. Kitces's site has the pairing that makes it concrete: two paths with an identical 8% average return, one favourable early and one favourable late, produce an eight-year spread in when you actually get to stop. A coaster with no contributions has nothing to do about it except wait.

Buying a day is cheap. Buying an exit is not.

The arithmetic that settled it for us is this. Same 5.2% real as the century figure — a touch above the 5% I actually plan with, so if anything the gaps below are understated. Same 25× target, three behaviours from 40 onward.

Illustration, not our balances. Return assumption: Dimson-Marsh-Staunton / UBS Global Investment Returns Yearbook.

Stop saving at coast and full financial independence arrives at 67, by construction. That's what the number means. Move to a four-day week, keep investing what's left, and it arrives around 51. Stay full-time and it's 50.

One year. Buying back a day a week costs roughly one year of full FI. Stopping costs sixteen.

I've read a lot of coast FIRE content this year and not one page runs that comparison, which is strange, because it's the only decision actually on the table. Everywhere the framing is binary: keep grinding, or step off. The interesting option sits in the middle and it's absurdly cheap.

Most people don't want out of their job. They want out of Friday.

The downshift story has a premise underneath it that nobody checks, which is that the reader wants out.

Eurostat put job satisfaction into the 2021 labour force survey. Across the EU, 43.9% of workers report high satisfaction and 5.8% report low. In Czechia it's 37.2% high, 4.6% low, and a huge 57% parked in the middle. Gallup's engagement figures are harsher, 12% engaged in Europe, though engagement and satisfaction are not the same animal.

So I'll concede half of it. Loving your job is a minority position and I'm in that minority, which means I shouldn't build advice out of my own luck. But the other half of the FIRE internet's premise, the miserable workforce desperate for an exit, isn't in the data either. Most people are neither in love nor in flight.

Then Eurofound's 2024 working conditions survey, 36,644 workers across 35 countries, hands over the actual answer. Thirty-three percent would prefer to work fewer hours. Fifty-six percent want the hours they already have. What people are trying to buy is not a different job. It's Friday.

Which is what I've been saying badly at home for years. My wife never got sold on FIRE as a concept; what she agreed to was a free Wednesday. The coast number is only the date that becomes affordable.

The part the American blogs can't tell you

Now the bit that makes this a European article rather than a translated one.

All of the above assumes you can buy the four-day week. In Czechia that's a real question, not a formality.

Part-time work was 17.7% of EU employment in 2024. Netherlands 42%, Germany 29%, Austria 30%. Czechia: 7.7%. And IDEA CERGE-EI's 2025 study is blunt about why. Among Czech women the reasons for working part-time are spread roughly evenly across health, caring duties and simply not finding full-time work, so it's a supply problem rather than a taste. The levy structure punishes low earnings, per-employee costs are fixed, the economy leans on shifts and industry. The arrangement I'm planning to buy is the one Czech employers are least set up to sell.

There's also a gap between what employers say and what workers get. Grafton's Q1 2026 Czech labour survey asked both sides: 88% of employers say they offer flexible hours, and about a third of employees agree that they do.

Germany legislated around this in 2019. Brückenteilzeit lets you cut your hours for one to five years, at employers with more than 45 staff, with a guaranteed right to return to the old ones afterwards — precisely because open-ended part-time was leaving people stuck. Czech law gets halfway there. A parent of a child under 15 can ask for shorter hours and the employer has to agree unless there are serious operational reasons, in writing. Going back up is a request the employer can simply decline. So I can buy the day. I can't buy the option to hand it back.

Barista FIRE, the other permission-slip option, does not survive contact with Czech numbers at all. The 2026 minimum wage is 22,400 Kč a month against an average gross wage of 51,966 Kč in the second quarter of this year. Half-time at minimum wage is about 22% of an average full-time salary. Worse, the contract thresholds — 12,000 Kč a month on a DPP, 4,500 Kč on a DPČ — mean the cosy little coffee-shop job either carries the full 31.9% social and 13.5% health contributions or sits below the line paying nothing at all, and a month below that line builds no pension entitlement whatsoever. Czechia asks for 35 insured years.

An 0.8 contract turns out to be a completely different animal, and it's worth saying out loud: it doesn't shorten the insured period by a single day. A year worked over four days counts exactly like a year over five. Only the assessment base your pension is later calculated from goes down. That's the sort of detail that makes a foreign coast FIRE guide untranslatable.

Where I might be wrong about this

I'd rather put the strongest arguments against me in my own article than meet them in the comments.

The best one is a paper I keep coming back to. Whillans and co-authors, in PNAS, surveyed thousands of people including 850 millionaires, and found almost half of those millionaires spent nothing at all outsourcing tasks they disliked; of 98 working adults handed a hypothetical windfall, 2% would have spent it on saving time. Buying time makes people measurably happier and almost nobody does it. If that holds, my answer is right and my timing is wrong. Why wait for coast at all? Why not buy 0.9 FTE this year?

I don't have a clean answer. The honest one is that I'm still more comfortable with a bigger margin than a shorter week, which is a preference wearing prudence as a costume.

Second, my instinct that downshifting is a trap is much weaker for me than for most people. Westhoff's work on EU-SILC data finds the part-time pay penalty concentrated at the bottom of the wage distribution, 9 to 23% for men in the lower deciles, and statistically insignificant or even slightly positive at the top, because skilled part-timers tend to be retention hires. A senior product manager going to 0.8 is not the person being punished per hour. My case has to rest on savings capacity, not on unfairness.

Third, Jim Dahle's warning lands: cutting back may not fix burnout, and part-time work can turn out to feel just as unsatisfying as full-time. And Bronnie Ware's palliative patients, which is anecdote rather than study and she is careful to say so, put "I wish I hadn't worked so hard" second on the list.

Two go my way. Van Ours's review of retirement and health found cognitive skills deteriorate after retirement, with the mental-health gains showing up mostly in blue-collar and voluntary cases, which is not my profile. And the Killingsworth-Kahneman collaboration found that for people who are already happy, wellbeing keeps climbing with income instead of flattening out. The line about marginal saving years buying nothing is not well supported for someone who likes his life.

What I actually want on the day it lands

The daughter changed the shape of this, not the date. Remodelling everything around her moved our full FI number by a year or two, which after the six-month version of that exercise I've stopped finding alarming. What she changed is what I want the freedom for.

We were going to do the standard young-couple thing: hit FIRE, travel indefinitely, see how far it goes. Now we want to be a boring, stable family that travels in the school holidays and lets her have an ordinary teenage life, with friends and disappointments and everything that comes with those. There's a second reason and we're both stubborn about it. We want her growing up around two parents who work. Not two parents home at eleven on a Tuesday morning. Everyone contributes something to keep the thing moving, and I'd rather she watched a household handle pickup duty and weeknights and the occasional bad week than be told about it later.

Both of us know the coast number, by the way. It isn't a figure living quietly in my head. We open the life plan in MFFT most months to watch the distance shrink, which is more satisfying than it sounds.

So, on the record. When it lands I take a week off. I inflate nothing. I buy the day back instead: school pickup, afternoons with her, the yard, cooking, and a genuinely quiet Wednesday in the middle of the week. Then I go back to the job I like, on four days, and keep investing, because the coast FIRE number was never permission to leave. It was permission to stop being fragile. Teaching maths is a separate ambition with its own date.

And if I get there and catch myself wanting to bank two more years first, I'd like this paragraph to be sitting here waiting for me.