The five years before retirement.

A good retirement is built, not waited for.

Most people treat retirement as a date in a diary. It is closer to a handover. For thirty or forty years your income has arrived whether or not you thought about it. After the handover it arrives only because you arranged for it to, out of money you have already earned. The five years before that date are when the arranging gets done, and they are the last stretch in which a decision can still change the outcome by much.

They are also the first stretch in which a poor run of markets does lasting damage. Both things are true at once, which is what makes this window awkward, and worth taking seriously rather than leaving to the last quarter.

i. What makes these five years different.

Earlier on, almost every mistake is recoverable. Contribute too little in your thirties and there are decades in which to make it up. A market fall in your forties turns out, in hindsight, to have been a period of buying at lower prices.

Five years out, both of those cushions have gone. There is not much time left to add to the pot out of earnings, and there is not much time left for a fall to recover in before you start selling. The decisions get smaller and their consequences get larger. That is the whole reason this window deserves a plan rather than a rough intention.

It is worth saying what does not change. Five years is still a long time to be invested, and a plan that has to fund thirty years does not become a cash plan simply because you are standing near the beginning of it. How long the money needs to last is a separate question, and for most people the answer is longer than they assume.

ii. Start with the income, not the pot.

The first question is nearly always “have I got enough?” It cannot be answered as asked, because “enough” is not a property of a pot. It is a relationship between the pot and the income you want to take from it, for as long as you need to take it.

So start at the other end. What does a year actually cost you? Not the aspirational figure — the real one. The bills that arrive whether or not you feel like paying them. Then the things that make the year worth having. Then the occasional larger item: the car, the roof, the help to a child.

Three tiers, roughly: essential, comfortable, and the extras. Most people find the essential tier is smaller than they had feared, and that the comfortable tier is where the real conversation lives.

Then set the floor against it. The full new State Pension is £241.30 a week in the 2026/27 tax year — a little under £12,550 a year — having risen 4.8% in April. It is taxable, it is not paid until State Pension age, and you may not be entitled to the full rate: you need at least ten qualifying years on your National Insurance record to receive any new State Pension at all. Getting your own forecast takes about two minutes and is the most useful thing most people can do in an afternoon.

Check the age as well as the amount. State Pension age is 66 and rises to 67 between 2026 and 2028. If you were born between 6 April 1960 and 5 March 1961, yours lands somewhere in the middle, a month at a time. Born on or after 6 March 1961 and it is 67. People plan around the age they assumed when they were forty, and are caught out at sixty-four.

iii. Find everything you own.

Almost anyone who has worked for more than three employers has mislaid a pension. Not lost as in gone — lost as in nobody has written to the current address in a decade.

This is dull work and it pays better than most of the clever work. Track the pots down, list them, and note what kind each one is, because the kinds behave nothing alike.

A defined contribution pension is a pot with your name on it. A defined benefit pension is a promise of an income for life, usually rising with inflation, and it is not a pot at all — even though the transfer value quoted on the statement makes it look like one. Giving up a promised, largely inflation-linked income for a capital sum is one of the few decisions in financial planning that is effectively irreversible, which is why regulated advice is required by law where the transfer value exceeds £30,000. The promise is backed by the scheme, and by the Pension Protection Fund up to its limits if the employer fails. Strong — but not the same word as guaranteed.

There is a date worth marking while you are listing things. From 6 April 2027, unused pension funds count towards inheritance tax. For some people that changes the order in which it makes sense to spend things — and the order is exactly the sort of decision that is easy to change five years out and hard to change afterwards.

The decisions get smaller and their consequences get larger. That is the trade these five years ask you to make.

iv. Why late contributions often carry the most relief.

Money going into a pension in the last working years often attracts relief at a higher marginal rate than money going in earlier, for a fairly boring reason: earnings tend to peak late. That is a point about tax relief, not about what the money then goes on to do.

The annual allowance is £60,000, and it counts employer contributions as well as your own. Tax relief on your own contributions is a separate limit: 100% of your relevant UK earnings, or £3,600 gross if you earn less than that. Unused annual allowance from the previous three tax years can generally be carried forward, which matters in a year when a bonus, a sale or a final push makes a larger contribution possible. The allowance tapers for higher earners, reducing where adjusted income exceeds £260,000 and threshold income exceeds £200,000, down to a floor of £10,000.

The band that catches most people out is lower down. Between £100,000 and £125,140 of adjusted net income the personal allowance is withdrawn at a rate of £1 for every £2 earned, which produces an effective rate of 60% across that stretch for taxpayers in England, Wales and Northern Ireland — Scottish rates differ. Pension contributions are one of the few things that reduce adjusted net income. Whether that is the right use of the money in your case is a question about your circumstances, not about the arithmetic — but it is worth knowing the band exists before the tax year ends rather than after. There is more on this on the tax-efficient planning page.

v. Tax-free cash is not a prize for reaching the minimum age.

You can usually take a quarter of a pension free of income tax. The cap is the lump sum allowance, £268,275 — a quarter of the old lifetime allowance, and unchanged since it replaced it. The earliest age you can normally take it is 55, rising to 57 on 6 April 2028 — which falls inside the five-year window for anyone reading this in their early fifties.

The mistake is treating it as something to be collected. Taking it because you have become eligible, with nothing particular to do with it, moves money out of a sheltered account into a taxable one and shortens the life of the plan for no benefit.

The order also matters, and this is the part that catches people. Taking only the tax-free cash and leaving the rest invested does not, by itself, restrict what you can contribute in future. Taking taxable income flexibly generally does: it triggers the money purchase annual allowance, which cuts what you can pay into defined contribution pensions to £10,000 a year, and it cannot be undone. If you intend to carry on working and carry on contributing — and plenty of people who "retire" at sixty do both — then doing these things in the wrong order can cost more than the cash was worth.

vi. The order of returns matters more now than it ever will again.

Over a long stretch of contributing, what matters is the average return. Once money is coming out, the order the returns arrive in matters as well — and around the point where contributions stop and withdrawals start, it matters most.

The reason is mechanical rather than mysterious. While you are paying in, a fall means you buy at lower prices. Once you are drawing out, a fall means you sell more units to raise the same income, and those units are no longer there to recover when the market does. Two portfolios earning the same average return over the same period can end up in quite different places purely because of the order in which those returns turned up.

That is sequencing risk, and it is why the years either side of a last payday get treated differently from the years before them. The usual answer is not to abandon investing — across a retirement measured in decades, cash has its own way of shrinking. It is to hold enough in cash and steadier assets to cover the near-term income, so that you are never forced to sell a falling asset to pay for the shopping.

How much is enough depends on the income you need, the shape of the rest of the portfolio, and how much variability you can live with without changing your mind at the worst moment. It has a real answer. The answer is different for each person, which is the honest reason it is not printed here as a rule of thumb.

vii. Write it down before you need it.

What these five years really offer is quiet. Everything above can be done calmly now, or urgently later, and the calm version is better in every case. Nothing here has to be settled in one sitting, and none of it improves for being left.

When I work through a retirement with someone, the output is not a recommendation on a page. It is a written plan: the income, where each part of it comes from and in what order, what happens if the first few years of markets are poor, and what happens to whoever is left. The work is careful and slow, and I write all of it down.

A plan you can read is a plan you can hold someone to. Including yourself.

— Andrew

Andrew Daw is an independent financial planner based in Bath, working with clients across Bath, Bristol and the South West. More about Andrew.