Start with the problem, not the portfolio
A retirement target such as £500,000 or £1 million sounds precise, but on its own it tells you very little.
Two people can retire with the same portfolio and need completely different amounts from it. One might own their home outright, spend £25,000 a year and receive pension income later. Another might spend £50,000, continue paying housing costs and want to stop working years before any pension begins.
The useful question is therefore not:
How large should my portfolio be?
It is:
What will I need the portfolio to pay for, and for how long?
Step 1: estimate the spending you want to fund
Start with annual spending in today’s money.
Current spending is often a better starting point than a percentage of salary, because retirement does not automatically cost 70% or 80% of what you earned while working.
Some costs may disappear or fall:
- commuting;
- pension contributions;
- mortgage payments, if the mortgage is repaid.
Others may increase:
- travel;
- time spent at home;
- hobbies;
- healthcare, or care later in life.
Suppose you estimate that you would want £30,000 a year in today’s money. That is the starting point, not yet the portfolio requirement.
Step 2: subtract income that does not need to come from the portfolio
You may expect income from pensions, annuities, rental property or other sources. Suppose £12,000 a year of your £30,000 spending is eventually covered elsewhere. The portfolio-funded gap is:
Once that other income has started, the portfolio needs to support £18,000 a year rather than the full £30,000. That distinction matters enormously, and so does when the other income starts.
- From the portfolio
- Other income
Spending of £30,000 a year from retirement at 55. Other income of £12,000 a year from 67. Everything in today’s money; shown to age 90.
Show the numbers
| Age | From the portfolio | Other income |
|---|---|---|
| 55–66 | £30,000 | £0 |
| 67–90 | £18,000 | £12,000 |
For the first 12 years the portfolio pays for all of the spending; after that, for 60% of it. The years before other income begins are the bridge period.
Step 3: estimate the long-term portfolio requirement
A common shortcut is the withdrawal-rate approach. If a portfolio needs to provide £18,000 in the first year and you use a 4% initial withdrawal rate:
This is where the familiar 25× spending rule comes from:
so:
The arithmetic is simple. The difficult question is whether 4% is an appropriate assumption.
What does the 4% rule actually mean?
The 4% rule is not the claim that a portfolio reliably earns 4% every year.
It is a withdrawal framework: withdraw 4% of the initial portfolio in the first year, then adjust that cash amount for inflation in later years.
Where the figure comes from
The number is usually traced to a 1994 study by the financial planner William Bengen. Using historical US stock and bond returns, he tested retirements starting in each year from 1926 onwards: withdraw a percentage of the portfolio in the first year, then raise that cash amount each year with inflation. Starting at about 4%, the money lasted at least 30 years in every period he tested, for portfolios holding a mix of US shares and intermediate-term US government bonds.
Three details of that test matter here. It covered retirements of about 30 years. It used one country’s markets over one period of history. And it is evidence about what would have happened in the past, not a guarantee about what will happen next.
Source William P. Bengen, “Determining Withdrawal Rates Using Historical Data”, Journal of Financial Planning, October 1994, vol. 7, no. 4, pp. 171–180.
It should therefore be treated as a planning assumption rather than a guarantee. The result depends on factors including:
- how long the portfolio needs to last;
- the investment mix;
- fees and taxes;
- inflation;
- the order in which good and bad market years occur;
- whether spending can be reduced after poor returns.
A retirement beginning at 65 and one beginning at 40 are therefore not the same problem, even if annual spending is identical.
Why the order of returns matters
Average return alone does not describe a retirement portfolio.
If two portfolios experience the same set of annual returns in a different order, they can finish with very different values when withdrawals are being made.
Poor returns early in retirement are particularly damaging, because money is being withdrawn from a portfolio that has already fallen. There is then less capital left to take part in a later recovery. This is known as sequence-of-returns risk.
- Bad years first£460,300
- Same returns, reversed£1,042,900
Starts with £500,000. £20,000 is withdrawn at the start of each year and kept level in today’s money, so the returns are real returns. Hypothetical returns, in order: −18%, −12%, +4%, +9%, +7%, +11%, +6%, −3%, +10%, +8%, +13%, +5%, +7%, +9%, +2%, +12%, +6%, +8%, +11%, +4%, +9%, +7%, +14%, +6%, +10% (an average of 5.8% a year). The second line uses exactly the same list, reversed.
Show the numbers
| Year | Bad years first | Same returns, reversed |
|---|---|---|
| 0 | £500,000 | £500,000 |
| 5 | £329,800 | £644,200 |
| 10 | £327,200 | £823,700 |
| 15 | £342,200 | £1,035,000 |
| 20 | £381,700 | £1,282,600 |
| 25 | £460,300 | £1,042,900 |
After 25 years one portfolio holds £460,300 and the other £1,042,900. With no withdrawals, both would end at £1,921,900: the same returns multiplied in a different order give exactly the same result. It is the withdrawals that make the order matter.
Early retirement adds a bridge
Suppose you want to stop working at 55, but another source of retirement income does not begin until 67. For 12 years the portfolio has to support more of your spending: the bridge period in the first chart.
A simple calculation such as:
shows the scale of the spending during that period, but it is not the same as saying you need an additional £360,000 on the day you retire. The portfolio remains invested, withdrawals happen over time, and returns and inflation matter.
This is exactly why a year-by-year projection becomes more useful than multiplying one annual number by 25.
Where FIRE fits into this
FIRE, Financial Independence, Retire Early, uses the same underlying question. The difference is mostly the date.
If your portfolio can support all of your spending for the rest of the period you are planning, paid work can become optional. That does not require an all-or-nothing definition.
Full financial independence
The portfolio and other non-employment income are projected to support all planned spending.
Partial financial independence
The portfolio covers part of your spending, reducing the amount you need to earn from work.
Coast FIRE
You have enough invested that, under your assumptions, the portfolio could reach a later retirement target without further contributions. You still need income for today’s spending, but no longer need to contribute towards that particular future target.
These are different points on the same timeline rather than different financial philosophies.
Why savings rate matters
For financial independence, spending affects both sides of the calculation. If you spend less:
- more of your current income can be saved; and
- the future portfolio has less spending to support.
That is why the savings rate can have such a large effect on the time it takes. With a take-home pay , a savings rate and a real return , starting from nothing and saving at the end of each year, the number of years to reach 25 times annual spending satisfies:
Income appears on both sides and cancels out: only the savings rate and the return are left. Solving it for different savings rates gives the chart below.
Starting from nothing. Saving and spending are fixed shares of the same take-home pay; each year’s saving is invested at the end of the year at a 5% real return. Target: 25 × annual spending. A simplified mathematical illustration, not a forecast.
Going from saving 10% to 20% shortens the wait by 14.6 years. Going from 50% to 60% shortens it by 4.2 years: the same ten points, taken off a much shorter wait.
So how much do you need?
There is no useful universal answer. A better retirement calculation is:
- estimate spending;
- identify other income and when it starts;
- calculate what the portfolio needs to fund in each period;
- choose and test withdrawal and return assumptions;
- include any early-retirement bridge;
- account for tax where relevant;
- compare the required portfolio with what you have today;
- rerun the plan under less favourable assumptions.
The result is not one magic retirement number. It is a plan showing when different sources of money are expected to fund your life.
That is also why planning backwards is useful. Instead of projecting your current savings for thirty years and hoping the final number is enough, you define what the money eventually needs to do and work back from there.