[ER-07] Coast FIRE and Barista FIRE

One of these is genuinely immune to sequence risk and the other is early retirement wearing a name tag. The difference is whether anything leaves the portfolio.

Pillar: Early Retirement Mechanics · Applies to: Anyone in their 30s or 40s deciding whether they can stop saving, downshift, or go part-time before a full retirement is funded Last verified: August 2026 · Refresh cadence: Evergreen (the arithmetic); healthcare figures via FN-02 Related: ER-01 Safe Withdrawal Rates · ER-06 Sequence of Returns Risk · ER-05 Accessing Money Early · HC-01 ACA Bridge · HC-03 Medicaid Floor · [IN-04] Part-Time Work · SS-04 Earnings Record · TR-04 Practice Retirement

Not advice. The arithmetic here is exact; the return assumption underneath it is not, and §1 shows that assumption swinging the answer by 2.3x. Treat any coast number — including one from a calculator that sounds confident — as a projection whose input you chose.

---

TL;DR

• Coast FIRE: you have enough invested that compounding alone reaches your number by traditional retirement age. You stop saving; you do not start spending the portfolio.
• Barista FIRE: part-time income covers part of your spending and the portfolio funds the rest. That is a withdrawal, which makes it early retirement with a wage subsidy, not coasting.
• The distinction is not vocabulary. ER-06 established that sequence-of-returns risk requires cash flows. Coast FIRE is the only strategy in this wiki genuinely immune to it. Barista FIRE has it in full, at the worst possible age.
• The coast number is an opinion about returns. For $1,500,000 at 65 from age 35: $462,478 at 4% real, $347,066 at 5%, $261,165 at 6%, $197,051 at 7% — a 2.3x spread from a single unverifiable input.
• Coasting on 7% and getting 5% leaves you with $851,642 against a $1,500,000 goal — $648,358 short, 43% of the target — and the error is invisible for thirty years. Caught at 35 it costs 8 years of continued saving; caught at 55 it cannot be saved away at all.

---

1. The coast number is an opinion about returns

The formula is trivial: divide the target by compound growth over the years remaining.

Target $1,500,000 in today’s dollars at 65, coasting from 35:

Assumed real return	Coast number at 35
4%	$462,478
5%	$347,066
6%	$261,165
7%	$197,051

Same person, same goal, same age — and a 2.3x spread between the most and least optimistic defensible assumption. Nothing about the arithmetic is uncertain. The input is, and it decides everything.

Every Coast FIRE calculator hides this behind a default, usually 7%, and usually a nominal 7% quietly compared against a real target. The single most useful discipline on this page is to run your number at 4% and ask whether you would still stop saving.

The number grows at your assumed return

Holding the return at 5% real and varying only the age:

Age	Years to 65	Coast number
30	35	$271,935
35	30	$347,066
40	25	$442,954
45	20	$565,334
50	15	$721,526

The bar rises at exactly the assumed real return — 5% a year — because one year of forgone compounding is precisely what a year of waiting costs. Waiting five years to start coasting raises the number by 28%, and no market view is required to know that; it is the same arithmetic read backwards.

That reframes the decision usefully. “Should I coast now or in three years?” has an exact answer in dollars, independent of any forecast, and it is usually larger than people expect.

A note on which return to use. These are real returns, net of inflation and net of fees. A 7% real return is above the long-run realised US equity return before costs, and ER-01 §2 is the reason to be careful: the US record is the best of the developed markets, and using it as the central case is a choice, not a neutral default.

2. What happens when the optimistic assumption is wrong

The uncomfortable property of coasting is that the error is silent.

Coasts at $197,051 on a 7% assumption, actually gets 5%
Balance at 65	$851,642
Target	$1,500,000
Shortfall	$648,358 — 43% of the goal

Nothing signals this along the way. The portfolio rises every good year, the plan looks fine at 45 and at 55, and the bill arrives at the one moment when no lever remains. Compare it to the failure mode of a withdrawal plan, which announces itself continuously in the withdrawal rate (TX-07) — coasting has no equivalent dashboard, and needs one imposed by hand.

Caught early it is cheap. Caught late it is not. Continuing to contribute $25,000 a year at a 5% real return, the 7%-assumption coaster needs 8 years of further saving to get back on track. The same discovery at 55 cannot be fixed by saving at all — only by working longer, spending less, or accepting a smaller retirement.

Hence the one operational rule this page insists on: re-run the coast number annually against your actual balance. If it has drifted below the target path, you are no longer coasting; you are behind, and you have the option of noticing.

3. The distinction that actually matters

	Coast FIRE	Barista FIRE
Contributions	Stopped	Stopped
Withdrawals	None	Yes — the portfolio funds the gap
Sequence-of-returns risk	None ✅	Full, and at a young age
What it needs from a job	Enough to cover living costs	Enough to cover part of living costs
Failure mode	Silent under-compounding (§2)	A bad decade early, with no wage to stop the selling

ER-06 §2 proved that order of returns is arithmetically irrelevant without cash flows — two portfolios with the same returns in opposite order ended at the identical dollar. Coast FIRE is that case exactly. No withdrawals, no sequence risk, and a bad decade early is genuinely good for a coaster who is still holding — the same market that damages a retiree leaves an untouched portfolio to recover on the full share count.

Barista FIRE is the opposite, and the name conceals it. The moment the portfolio funds any part of spending, sequence risk applies in full — and it applies at 45 rather than 65, with more withdrawal-years ahead and no Social Security floor for two decades (ER-01 §3). Barista FIRE is early retirement with a partial wage subsidy, and it should be planned with ER-06’s tools, not this page’s.

The practical consequence: a Barista FIRE household needs the buffer, the guardrail rule, and the flexible spending plan that ER-06 and TX-07 describe. A true coaster needs none of them and should not carry their cost.

4. What the part-time job is really for

For most Barista FIRE households the wage is not the point. Health coverage is, and the 2026 restoration of the 400% FPL subsidy cliff made that sharper (HC-01, FN-03).

Three ways the job interacts with coverage, in descending order of value:

1. Employer coverage. Removes the ACA question entirely and is the historical reason the strategy is named after a coffee chain — a small number of employers offer benefits at genuinely part-time hours ◻️. Confirm the hours threshold before resigning anything.
2. Marketplace coverage with earned income shaping MAGI. Part-time earnings are MAGI, which is the dial that sets your subsidy — and the target is a band, not a minimum. Too little income is its own failure: below 138% FPL in an expansion state you land on Medicaid, and below 100% in a non-expansion state you land in the coverage gap with no subsidy at all (HC-03).
3. Neither — earning enough to cross the cliff and buying unsubsidised coverage, which is the case where the job has to be much larger to be worth taking at all (HC-05 §3).

The second is where most Barista households actually live, and it inverts the usual advice about earning more. [IN-04] owns the part-time-work analysis; what belongs here is the warning that the job’s value is not its wage — it is the wage plus the coverage minus the subsidy the wage destroys.

One more effect, easy to forget: stopping work early leaves zeros in the 35-year earnings average that sets your Social Security benefit (SS-04). Part-time years replace zeros with small numbers, which is worth more to a short career than the headline wage suggests.

5. Two stacked assumptions, not one

§1 treated the target as given. It isn’t. The coast number is derived from the FIRE number, which is itself derived from a withdrawal rate — and ER-01 found credible rates spanning 2.31% to 5.50%. The uncertainties multiply.

Spending $60,000 a year, coasting from 35 at a 5% real return:

Withdrawal rate	FIRE number	Coast number at 35
2.31%	$2,597,403	$600,980
4.00%	$1,500,000	$347,066
5.50%	$1,090,909	$252,412

A 2.4x spread — and that is before §1’s return assumption applies its own 2.3x. The two are independent choices and they compound rather than cancel.

This is why a coast number quoted to the dollar is the least trustworthy figure in FIRE, and why the useful output is a range with an annual review rather than a target with a party. State both assumptions out loud whenever you state the number; if you cannot, you have not made the decision, you have accepted someone’s defaults.

6. The coast years are the cheapest conversion years you will have

A coaster earning just enough to live on usually has the lowest taxable income of their adult life: the high salary has stopped, contributions have stopped, and withdrawals have not started.

That is precisely the profile the Roth conversion ladder wants (ER-02, TX-02) — and a coaster has an advantage a retiree does not: they can pay the conversion tax from wages rather than from the converted balance, which is strictly better and is the single most under-used feature of the whole strategy.

The binding constraint is the subsidy band, not the tax bracket. A conversion is MAGI, and MAGI sets the premium tax credit — so the ceiling is the cliff at 400% FPL and the floor is 138% (or 100% in a non-expansion state), not the top of the 12% bracket (HC-01, HC-03, ER-05). A coaster who converts to the top of a bracket without checking the band has made the expensive version of a good decision.

Three Scenarios

A. 36, $400,000 invested, spending $55,000, hates the job. At 5% real her coast number is $347,066, so she is past it — she can stop saving today and reach $1,500,000 at 65 without another contribution. What she cannot do is stop working, because nothing is funding the $55,000. Coast FIRE bought her freedom from saving, not from earning, and the distinction is the whole strategy.

B. 44, $650,000 invested, wants to drop to three days a week at $38,000 against $70,000 of spending. The $32,000 gap comes from the portfolio, so this is Barista FIRE and sequence risk is live from day one (ER-06). The right structure is not the coast number — it is a buffer, a written guardrail rule (TX-07), and a MAGI target set against the subsidy band (HC-01, HC-03).

C. 52, coasting since 44 on a 7% assumption. Returns have run about 4.5% real. He is $200,000 behind the path and has never once checked, because the balance kept rising and nothing in the strategy demanded a review. Re-running the coast number annually would have caught it at 46, when eight years of contributions still fixed it. At 52 the honest options have narrowed to working longer or spending less.

💡 Pro-Tips

1. Run your coast number at 4% real before you resign anything. If the decision only works at 7%, it is a bet, not a plan.
2. Check whether your calculator used nominal or real returns — comparing a nominal 7% against a real target overstates the outcome by roughly a third over thirty years.
3. Re-run the number every year against your actual balance. Coasting has no built-in warning system; this is the substitute.
4. Know which strategy you are running. If the portfolio funds one dollar of spending, you are Barista, not Coast, and you need ER-06’s toolkit.
5. Price the part-time job as wage + coverage − subsidy lost, never as wage alone (HC-01, [IN-04]).
6. Target a MAGI band, not a minimum — under the floor is as damaging as over the cliff (HC-03).
7. Keep contributing to tax-advantaged space even while coasting if you have any earned income; a coaster with a small 1099 still has an IRA and possibly a solo 401(k) ([IN-02]).
8. Coast in the accounts you can actually reach later. A coast number locked entirely in a 401(k) does not solve a 45-to-59½ access problem (ER-05, ER-02).

⚠️ Common Pitfalls

1. Accepting the calculator’s default return. It is usually 7% and usually the whole answer (§1).
2. Mixing nominal returns with a real target, which quietly inflates every projection.
3. Never re-checking. The §2 failure is invisible for decades and then unfixable.
4. Calling Barista FIRE “coasting.” One has no sequence risk; the other has it in full at a young age (§3).
5. Assuming a bad market hurts a coaster. Without withdrawals it does not — and may help (ER-06 §2).
6. Ignoring where the money sits. Coasting to 65 in accounts you cannot touch before 59½ solves the wrong problem (ER-05).
7. Taking a part-time job that pays just enough to cross the subsidy cliff and losing more than it pays (HC-01 §4).
8. Optimising MAGI downward into the Medicaid floor or the coverage gap (HC-03).
9. Forgetting the Social Security earnings record. Early stopping leaves zeros in the 35-year average (SS-04).
10. Treating the coast number as a finish line. It is a projection with an annual review attached, not an achievement.

✅ Actionable Checklist

Before downshifting

• Compute the coast number at 4%, 5%, 6% and 7% real and look at the spread
• Confirm whether your target is stated in today’s dollars and your return in real terms
• Decide honestly whether you are Coast (no withdrawals) or Barista (withdrawals)
• If Barista: build the buffer and write the spending rule first (ER-06, TX-07)
• Check the account mix against your access ages (ER-05)

Choosing the work

• Price coverage first: employer plan, or marketplace with a MAGI target (HC-01, HC-03)
• Confirm the employer’s hours threshold for benefits in writing ◻️
• Compute wage + coverage − subsidy lost, and compare that to the hours

Every year, without exception

• Re-run the coast number against the actual balance and the years remaining
• If behind the path, decide now between saving again, working longer, or a smaller target
• Re-check the MAGI band against the current year’s FPL figures (FN-02)
• Contribute to any tax-advantaged space your earned income allows

Sources & further reading (verified August 2026)

• tools/er07_worked_examples.py — the coast-number table, the shortfall arithmetic, and the recovery calculation. Compound interest only; every figure is reproducible by running the file.
• ER-06 Sequence of Returns Risk — §2 establishes that order of returns is irrelevant without cash flows, which is the foundation of the Coast/Barista distinction in §3 and is the reason that distinction is structural rather than semantic.
• ER-01 Safe Withdrawal Rates — §2 on why a US-calibrated real return is a choice rather than a neutral default, and §3 on horizon arithmetic for anyone who will eventually draw on this portfolio.
• HC-01 for the premium tax credit and the 400% FPL cliff that shapes the part-time decision, HC-03 for the floor below it, and FN-02 for the current-year FPL figures.
• [IN-04] owns the part-time-work analysis this page only touches · SS-04 owns the earnings-record effect · TX-07 owns the spending rule a Barista household needs · TR-04 owns trialling the downshift before committing to it.