The risk nobody mentions
Why the order matters more than the average
Two people, same money, same average return over thirty years. One is comfortable at 90 and one ran out at 79. The only difference is which years were the bad ones.
Two people, same money, same average return over thirty years. One is comfortable at 90 and one ran out at 79. The only difference is which years were the bad ones.
If two people get the same average return over thirty years, you'd expect roughly the same outcome. While you're still saving, that's about right — the order barely matters.
Once you start taking money out, it stops being true, and the difference is enormous.
When you're saving, a bad year means you buy at lower prices — it can even help. When you're drawing, a bad year means you sell at lower prices, and those units are gone. They aren't there to recover when the market does. The loss becomes permanent in a way it never was while you were building up.
Two people stop at 62 with £500,000 each. Both take £30,000 a year, rising with inflation. Both experience exactly the same set of annual returns over the next twenty-five years — the same numbers, in a different order.
Three poor years right at the start. They draw £30,000 from a falling pot, selling more units each time to raise the same money. By 70 the pot is far smaller than it should be, and the good years that follow are working on much less capital. The money runs out in their late seventies.
The same three poor years arrive in their mid-seventies instead. By then the pot has had a decade of growth behind it and can absorb the hit. They still have money past 90.
Exactly the same numbers. The only variable is the order they arrived in, and it produced a difference of more than a decade.
That's the uncomfortable part. It wasn't a bad decision or a bad investment. It was when they happened to be born.
Most calculators — including the free one on this site — use one flat growth rate. They effectively assume every year is average, which is a year that has never actually happened.
That produces a smooth, reassuring line and tells you nothing about how fragile the plan is. Two plans can show the same answer on a flat-rate projection while one survives almost any sequence and the other only works if the first five years behave.
This is why the Financial Plan runs your numbers two thousand more times with the good and bad years reshuffled, and tells you in how many of them the money lasted. That percentage isn't a forecast. It's the answer to one question: how much does this plan depend on luck?
There's no way to eliminate it. There are recognised ways of reducing how exposed you are, and they're trade-offs rather than solutions.
Keeping a year or two of spending in cash means a bad market at the wrong moment doesn't force you to sell into it. It costs growth in every year it isn't needed. Whether that insurance is worth its premium is a genuine decision.
Someone who can cut 10% in a bad year is dramatically more robust than someone whose outgoings are fixed. Flexibility is worth more than almost any product feature.
Even modest income in the first few years reduces how much you have to sell at the worst time. This is why two days a week at 61 does more for a plan than the money alone suggests.
Where the State Pension and any defined benefit scheme cover the non-negotiable costs, market falls affect your choices rather than your survival — a completely different kind of bad year.
How you structure investments to manage this is a regulated advice question and not ours to answer. What we can do is show you how exposed your plan is, and what each of the above does to the numbers — so if you do go to an authorised adviser, you arrive knowing what problem you're asking them to solve.
The practical consequence of sequence risk is not that you need a better forecast. It is that you need a plan that survives being wrong.
A plan built to work on one flat growth rate is optimised for a year that has never happened. Every real retirement is a specific sequence, and nobody gets to choose theirs. So the question worth asking about any plan is not whether it produces a comfortable answer, but how far the inputs have to move before the answer stops being comfortable. A plan that fails when growth is half a percent lower, or when the first two years are poor, is a plan resting on precision it cannot possibly have.
This is also why the answers that come out of the stress tests are given in years rather than percentages. Being told your terminal wealth falls by 12% conveys nothing useful. Being told a bad first three years costs you seven years at the end is a fact you can act on, and it makes the value of the responses above obvious. Holding two years of spending in cash, or working two days a week until 63, or being able to cut spending by a tenth in a bad year, each buy back a measurable amount of that.
Guaranteed income covering the non-negotiable costs, so a bad market changes your choices rather than your survival. Some genuine flexibility in what gets spent, because the ability to take less for two years is worth more than most product features. And a stop date with a little give in it, because the single most powerful response to a bad opening sequence is not stopping quite yet.
Related, but a distinct thing. Market risk is the risk that investments fall in value. Sequence risk is the risk that they fall at a particularly damaging moment, which for a retirement means just as you begin taking money out. The two can come apart completely. You can experience a perfectly respectable long-run average return across thirty years and still be badly hurt, purely because the poor years arrived at the start rather than in the middle. Equally, a retirement that begins with a good decade can absorb a severe fall later without much lasting harm. Same investments, same average, entirely different outcome, and the only variable is timing.
Roughly the five years either side of the day you stop work. That is when the pot is at its largest, so a percentage fall costs the most in actual pounds, and it is also the point at which you are about to start selling rather than buying. A fall in your first drawing year does damage in two directions at once: the pot is smaller, and you are converting units into income at the worst available price. Before that period, a fall is an opportunity, because contributions buy more. Long after it, the pot has usually had enough growth behind it to absorb the hit. It is that narrow window that decides an unusual amount.
No, and moving everything to cash would swap one serious problem for another. Cash reliably loses purchasing power across a retirement that might run thirty or forty years, and inflation does that damage quietly rather than visibly, which makes it easier to ignore and no less real. The useful question is not cash or investments as a binary. It is how much of the first few years you want insulated, which is a matter of degree and comes with a cost in growth you give up in every year the insulation is not needed. Where to hold it and in what is a regulated advice question, and not one we can answer.
That is exactly what running the same plan thousands of times tells you. Reshuffle the good and bad years, keep everything else identical, and count how often the money lasted. If it lasts in 95% of the sequences, the plan is robust and does not need luck. If it lasts in 45%, the plan works when the next few years are kind and fails when they are not, which is a very different proposition from the single smooth line you were shown. Neither number is a prediction about markets. Both are measurements of how much your particular plan depends on timing you cannot control, and that is worth knowing beforehand.
No, and the page says so rather than pretending otherwise. The free projection runs on one flat growth rate, which is precisely the simplification this guide exists to warn about. It is genuinely useful for the question it does answer, which is roughly what age your money runs out on a set of assumptions, and for most people that is a sensible first step. What it cannot do is tell you how fragile that answer is. Two plans can produce an identical result on a flat-rate projection while one survives almost any sequence and the other only works if the first five years behave. Seeing which one you have needs the full version, and that is one of the main things the Financial Plan is for.
The projection runs in your browser in about a minute. Nothing is sent anywhere and no email is needed to see the answer.
Thirty minutes with a planner to work out whether you need a Plan at all. Some people leave that call having been told they don't.
One written document — every year to 100 with the tax done properly, what breaks it priced in years, and the ninety days after. £1,500.