When cash pays 5%, why hold shares at all?

For years, cash had to apologise for itself. It was safe, certainly, but the return was miserable.

Cash can now look investors in the eye with a promise of return that beats inflation.

The best one-year savings accounts pay about 5%, with no stock market drama and no risk of waking up to find your investment has fallen sharply. It is a fair question to ask: are shares worth the risk at all, or is saving better than investing?

The answer starts with time. Cash is for money you may need soon and shares are for money you can leave invested for years, accepting short-term falls in pursuit of greater long-term growth.

What does the evidence show?

Some of the best-known historical evidence gives surprisingly different answers. Barclays Private Bank says UK shares beat cash in 91% of rolling 10-year periods since 1899, meaning every possible 10-year stretch, each starting a month or year apart.

Paul Lewis, presenter of BBC Radio 4’s Money Box, found something very different in research published in 2016. Across rolling ten-year periods between 1995 and 2016, a FTSE 100 tracker beat the best savings accounts only half the time.

Lewis set shares a harder test than Barclays does. He used the actual returns from a FTSE 100 tracker after charges and compared them with the best one-year savings account available each year. Barclays uses three-month Treasury bills instead, partly because they provide a consistent measure of cash stretching back more than a century.

Using the best available savings rates, rather than Treasury bills, gives cash a stronger showing. Timing matters too. For investments starting on the first of any month from October 1999 to September 2001, cash beat the tracker over every holding period Lewis could measure, up to 15 or 16 years.

What happens with more recent data?

I applied a similarly demanding test over a different period. My own analysis starts in December 2000, as far back as the relevant MSCI sterling total-return data go. I compared the MSCI UK and MSCI World indices, after a 0.2% annual charge, with cash rolled into a new one-year fix each year. For cash, I used the Bank of England’s average one-year fixed savings rate plus an allowance for best-buy accounts.

Over rolling ten-year periods to August 2026, UK shares beat cash 91% of the time. Global shares did so 90% of the time, so going global barely changed the odds.

The period contains only two non-overlapping decades, the same limitation as Lewis’s data. The analysis confirms Barclays’ result on fresh data rather than replacing it.

Why does time matter?

Ten years is only one horizon. Barclays found that UK shares beat cash in 70% of rolling two-year periods since 1899, rising to 91% over ten years. A 70% chance of beating cash is also a 30% chance of falling behind, which is exactly the risk that makes cash the safer home for money with a date on it.

Even ten years offered no guarantee. Cash still came out ahead in about one period in 11. The percentages are historical frequencies, not forecasts.

And cash has a time problem of its own.

In July 2008, the Bank of England‘s average one-year fixed savings rate was 6.06%. A one-year fix guarantees that rate for one year only. After that, you take whatever rate is available when you renew.

For money held over 15 or 20 years, that means renewing again and again at rates nobody can know in advance. The same Bank of England average stayed below 2% every month from January 2013 to August 2022. Best-buy savers did better, but faced the same broad decline.

Rates may rise as well as fall. Today’s rate cannot be locked in for decades.

How much cash do you need?

When Lewis published his research in 2016, I was sceptical. Cash paid next to nothing then, and I saw little reason to hold more of it than necessary.

My view has shifted. My wife and I are very likely to move house within the next two years, probably to a more expensive area. We also want to travel much more than we have and already have several destinations in mind. Those are foreseeable calls on our money, even if we cannot put exact dates or amounts on them yet.

So I now keep more in cash, though most of my wealth remains invested in shares. This is not a retreat from equities so much as a clearer division of labour.

MoneyHelper, a UK government-backed financial guidance service, suggests three to six months’ essential outgoings as a rule of thumb, while wealth manager Hargreaves Lansdown says money you may need within five years should generally be kept in savings rather than invested. Investing that money is a real risk: from December 2000, global shares fell 41% and did not regain their starting level until March 2006.

For retirees, Hargreaves Lansdown suggests keeping one to three years’ essential spending in cash. These rules of thumb give you two useful buckets: money for emergencies, and money for foreseeable spending. The harder question is what to do with everything left over.

What is the cost of staying in cash?

In my test, £10,000 invested in a global tracker from December 2000 to August 2026 grew to about £67,600 after charges. Rolled through one-year fixes over the same period, including an allowance for best-buy rates, it reached about £25,000. The global tracker ended with more than two and a half times as much as cash.

A UK tracker reached about £41,800 – or around 1.7 times as much as cash despite no exposure to US shares. US strength helped widen the gap between the UK and global results, while sterling weakness also gave the global result a modest boost.

The lesson is not that cash is bad. It is that once your needs are covered, holding more of it has an opportunity cost that compounds.

Work out what you are likely to need soon and keep that in cash. For money you will not touch for many years, accept the short-term falls that come with shares.

Then look at what is left. If it is still sitting in cash, ask why. If the only answer is that today’s rate feels reassuring, that is not enough.

The mistake is letting a one-year rate decide where long-term money belongs.