The short version
- In any single year, UK shares fell in value about one year in four.
- Over ten-year periods they never ended lower in pounds, but about one in eight left you poorer after inflation.
- Over twenty years, shares lost buying power in just 2 of 131 periods and beat cash 98% of the time.
- Cash isn't risk-free either: it lost buying power in about a third of all periods, however long it was held.
This is the starting guide in Worries and myths. Every guide in the series is listed at the end of this page.
- 1 in 4years in which UK shares fell in value (23%)
- 1 in 8ten-year periods in which shares lost buying power after inflation (12%)
- 98%of twenty-year periods in which shares beat cash
Two ways to lose money
There are two kinds of loss. Your investment can fall in pounds. Or it can grow more slowly than prices, so it buys less than it did. The second kind is quieter, and it’s the one that catches people who keep everything in cash.
So the honest answer to “will I lose money?” needs both. This page uses 150 years of UK data, from 1871 to 2020, to show how often each kind of loss happened for shares and for cash.
A fall in price is also only a paper loss until you sell. When do you actually lose money? shows what that meant for £1,000 in a FTSE 100 tracker during the 2020 crash.
Time changes the odds
In any single year, UK shares lost money in pounds about one year in four (23.3%). After inflation it was closer to three years in ten.
Hold for longer and the picture changes. Over ten-year periods, shares never ended below where they started in pounds. After inflation, 12.1% of ten-year periods still left you poorer. Over twenty years that fell to 1.5%: two periods out of 131.
The longer shares were held, the less often they lost buying power. Cash didn't improve with time.
Share of all holding periods, 1871–2020, that ended worth less after inflation
- UK shares
- Cash (government bills)
Source: Jordà-Schularick-Taylor Macrohistory Database (R6), UK, 1871–2020. Total returns with dividends reinvested, before fees and tax. Analysis: Invest in ETFs. Chart licensed CC BY-NC-SA 4.0.
Show the data
| Holding period | Periods | Shares lost buying power | Cash lost buying power | Shares lost money before inflation |
|---|---|---|---|---|
| 1 year | 150 | 28.0% | 33.3% | 23.3% |
| 5 years | 146 | 18.5% | 30.8% | 5.5% |
| 10 years | 141 | 12.1% | 34.8% | 0.0% |
| 20 years | 131 | 1.5% | 28.2% | 0.0% |
Cash behaves differently. It never falls in pounds, but it lost buying power in roughly a third of periods whatever the length: 33% over one year and 28% over twenty. Time doesn’t rescue cash the way it has rescued shares. What inflation does to cash looks at this in detail.
To test your own dates, try any two years in the returns explorer.
How often shares beat cash
The same pattern shows up when you compare the two directly. Over one year, shares came out ahead 65% of the time: about two years in three. Over twenty years, they came out ahead 98% of the time.
Over 20 years, UK shares beat cash 98% of the time
Share of all holding periods, 1871–2020, in which shares returned more than cash
Source: Jordà-Schularick-Taylor Macrohistory Database (R6), UK, 1871–2020. Total returns with dividends reinvested, before fees and tax. Analysis: Invest in ETFs. Chart licensed CC BY-NC-SA 4.0.
Show the data
| Holding period | Periods | Shares beat cash |
|---|---|---|
| 1 year | 150 | 64.7% |
| 5 years | 146 | 80.8% |
| 10 years | 141 | 91.5% |
| 20 years | 131 | 97.7% |
Over one year, shares are unpredictable. Over twenty, the odds have been heavily on your side, though never certain.
The worst cases
Long periods improve the odds. They don’t guarantee anything.
- The worst ten years started in 1911: shares lost 8.0% a year after inflation, as the First World War and the inflation that followed eroded their value.
- The worst twenty years started in 1901 and lost 1.4% a year after inflation.
- The worst single year was 1974, when shares lost 57% of their buying power. The next year they more than doubled. Every big UK crash is listed here.
What this means for you
- Money you’ll need within about five years has usually been safer in cash. Over five years, shares still lost buying power 19% of the time.
- Money you can leave for ten years or more has historically had much better odds in shares, but expect falls along the way.
- This is one country’s history. A global ETF spreads your money across many countries’ markets, each with its own good and bad decades.
| Held for | Periods | Typical yearly return after inflation | Worst | Best |
|---|---|---|---|---|
| 1 year | 150 | 6.5% | −57.0% (from 1974) | 103.4% (from 1975) |
| 5 years | 146 | 6.2% | −17.9% (from 1970) | 21.8% (from 1932) |
| 10 years | 141 | 5.7% | −8.0% (from 1911) | 17.8% (from 1975) |
| 20 years | 131 | 5.6% | −1.4% (from 1901) | 13.8% (from 1975) |
“Typical” is the median: half of periods did better, half did worse. Yearly figures for longer periods are averages, compounded.
For comparison, the US stock market’s S&P 500 returned about 10% a year from 1928 to 2025, with dividends reinvested, before inflation. The UK figures on this page are after inflation, so the two aren’t directly comparable, and the US average hides long stretches well below it.
About the data
- Source: the Jordà-Schularick-Taylor Macrohistory Database (release 6), which economists use for long-run returns. These figures use its UK series for 1871 to 2020, the latest year it covers.
- Shares means the UK stock market’s total return: price changes plus dividends reinvested, before fees and tax. An index ETF tracks something similar, minus its small annual fee.
- Cash means the interest on short-term government bills, a standard stand-in for savings. Real savings accounts have paid more or less than this at different times.
- After inflation means adjusted using the UK consumer price index.
- Periods are whole calendar years, overlapping: 1871–1880, 1872–1881 and so on. Yearly data hides falls that recovered within the same year.
- This is the UK market only, and the past is no guarantee of the future.
The analysis script and the derived data are kept with the site’s source files, so anyone can check the sums.
Every guide in Worries and myths
This page describes what has happened in the past. It isn't a forecast or advice about what you should do. Investments can fall as well as rise, and you may get back less than you put in. Read the full disclaimer.
Sources
- Macrohistory Database, release 6, Òscar Jordà, Moritz Schularick and Alan M. Taylor. Consulted 13 October 2026. Licensed CC BY-NC-SA 4.0.
- Jordà, Knoll, Kuvshinov, Schularick and Taylor (2019), "The Rate of Return on Everything, 1870–2015", Quarterly Journal of Economics 134(3).
- Jordà, Schularick and Taylor (2017), "Macrofinancial History and the New Business Cycle Facts", NBER Macroeconomics Annual 2016, volume 31.
- Historical returns on stocks, bonds and bills, 1928–2025, Aswath Damodaran, NYU Stern School of Business, updated January 2026. S&P 500 average worked out from the value of $100 invested at the start of 1928.
Analysis run 13 October 2026.