The idea
Conventional retirement planning targets a portfolio that lasts forever: withdraw 4%, live off returns, leave the principal intact. That is a sensible default, and it has a consequence nobody states out loud — most people who follow it die with more money than they retired with.
Bill Perkins' argument in Die With Zero is that this is a planning failure, not a success. Money buys experiences, and experiences have an age window. A trekking holiday at 55 is not the same purchase at 85. Money left unspent at death represents hours of your life exchanged for currency that was never converted back into anything.
The financial question that follows: if you are willing to finish at zero, how much more can you spend?
The formula
This is the standard annuity payment formula — the same arithmetic a bank uses to set a mortgage payment, run in reverse. It finds the constant real amount that draws the balance to exactly zero at the end of the horizon.
Working in real terms matters. The answer keeps the same purchasing power every year, so $57,800 in year 30 buys what $57,800 buys today. A nominal calculation would show a bigger number that quietly buys less each year.
A worked example
$1,000,000 at 60, planning to 90, on the default assumptions (7% nominal, 3% inflation, so a real return of 3.88%):
Over thirty years that is roughly $510,000 of additional spending — the difference between a comfortable retirement and a substantially different one. It comes from spending the principal as well as the returns.
The catch, stated plainly
Life expectancy is a median. Planning to hit zero at your life expectancy means, by construction, roughly a 50% chance of outliving your money.
That framing is uncomfortable enough that most presentations of this strategy skip it. The risk bands in the result panel do not: they show what the same spending does over five and ten additional years. On the default numbers, the money runs out well before 100.
And the risk is asymmetric in a way that ordinary financial risk is not:
- Underspending costs you experiences you could have had. Bad, but survivable, and your heirs benefit.
- Overspending means being 90 years old, out of money, with no ability to return to work and possibly needing expensive care. There is no recovery from this.
A rational response is not to reject the idea but to shift the horizon. Planning to 95 or 100 rather than to a median life expectancy captures most of the benefit while removing most of the ruin risk. Try it in the calculator — the annual figure falls, but far less than the extra security costs elsewhere.
What this model does not capture
- Long-term care. The largest practical objection. Late-life care can cost multiples of ordinary living expenses and arrives precisely when a spend-down plan has least left. Most people reserve a separate sum or insure.
- Sequence-of-returns risk. This assumes a constant real return. Real markets are lumpy, and a bad first decade damages a spend-down plan more than a preservation plan because there is less margin — see the stress test.
- Guaranteed income. Social Security, pensions and annuities change the picture fundamentally, because they cannot run out. If you have them, subtract them from expenses and use this tool for the remainder only.
- Spending is not flat. Real retirement spending tends to be high early ("go-go years"), lower in the middle, and higher again late through healthcare. A constant real figure is a simplification.
Why spending is not flat in practice
The formula assumes a constant real spend for the whole horizon. Real retirement spending does not behave that way, and the shape is well documented enough to be worth planning around.
Researchers describe three phases. The go-go years, roughly 60 to 75, are the most expensive: travel, hobbies, and the experiences that motivated retiring in the first place. The slow-go years, around 75 to 85, see spending fall as travel reduces and life narrows. The no-go years beyond 85 are cheapest for lifestyle and potentially the most expensive of all for healthcare.
This actually strengthens the core argument rather than weakening it. If your capacity to convert money into experiences is concentrated in the first fifteen years of retirement, then a plan that spreads spending evenly across thirty is misallocating on purpose. Front-loading spend towards the years you can use it is the whole point.
A common approach is to plan a higher spend for the first decade and step down afterwards, keeping a separate reserve for late-life care. That is more than this calculator models — it solves for a single constant figure — so treat the number here as an average to build around, not a schedule to follow literally.
Giving while you are alive
One genuinely strong point of the book survives all the caveats. If you intend to leave money to children, giving it at 30 — when they are buying homes and raising families — is worth far more to them than inheriting at 60, when they are near retirement themselves. Timing an inheritance is a real decision, and defaulting to "whatever is left when I die" is a choice too, just an unexamined one.
The legacy input above lets you model exactly that: set the amount you intend to pass on, and it is protected from the spend-down rather than being whatever happens to remain.
For the accumulation side, see the Coast FIRE calculator, and if you are retiring before 59½ in the US, the bridge account calculator covers getting at the money in the first place.