How often have you read that 7 years is considered a “long-term” investment timeframe? Here’s a typical example, taken directly from the Betashares website:
The idea that 7 years qualifies as long-term is widely accepted across the investment industry – and it’s highly problematic.
Framing it this way implies that investors can expect the average market return simply by staying invested for 7 years. But this assumption glosses over a key truth: markets don’t deliver average outcomes on demand.
This article explores the wide range of real investment outcomes across different time horizons, using nearly 100 years of historical data. It shows that even over longer periods, investing remains risky – and that averages can be misleading.
The Dataset
This analysis uses historical data from the S&P 500 – the index tracking the largest 500 companies in the U.S. It was chosen for its long data history, covering nearly a century of monthly returns from January 1926 onward.
The returns are nominal and exclude inflation adjustments or fees. Notably, the dataset includes extreme periods such as the Great Depression.
Across the full dataset, the average annual return was 10.32%.
Spread of Potential Outcomes
Using S&P 500 data from 1926 onward, we calculated annualised returns over every possible rolling 5, 7, and 10-year period. For each starting month in the dataset, we measured the return an investor would have received over the following period – creating hundreds of real, overlapping investment experiences – not theoretical averages.
In other words, it answers the question: “What kinds of returns could you have experienced if you invested for 5/7/10 years, starting in any given month for the past 100 years?”
These rolling periods are grouped by 2% return intervals (e.g. 9%–11%, –3% to –1%), illustrating the range of outcomes that investors historically experienced. The columns represent the number of times a return fell within a specific range of return.
As the investment horizon extends from 5 to 10 years, the distribution narrows: extreme outcomes become less common, and returns appear more consistent. However, a few key patterns stand out:
- Many outcomes still fall below the long-term average return of around 10%.
- Extremely high returns become less frequent over longer timeframes.
This highlights a crucial point:
Long-term investing reduces the range of possible outcomes, but it doesn’t guarantee the average, or even a positive return for that matter. In fact, there were 45 historical instances where a 7-year investment returned worse than –1% p.a.
These results challenge the assumption that time alone smooths returns to the average. Even over a decade, the “average” is more elusive than many investors are led to believe.
Table
The key observations are:
Returns Are Frequently Below the “Average”
Despite the long-term average return of 10.32% p.a., a large share of rolling periods delivered less than this:
- 41.17% of 5-year periods
- 44.22% of 7-year periods
- 47.01% of 10-year periods
Interestingly, the proportion of below-average outcomes increases as the timeframe extends. This suggests the average is distorted by a handful of very strong return periods – pulling the overall average up, even though most results were lower.
Long Timeframes Don’t Guarantee Positive Returns
This is the central point of the article. Even over 7 years, about 5% of rolling periods delivered a negative annualised return.
The same is true for 10-year periods – although none were worse than –5% p.a.
Time reduces volatility but doesn’t eliminate the chance of disappointment. The belief that “long-term” means safe and positive is not fully supported by historical data.
You Are More Likely to Outperform the Average
This is the opposite side of the first observation. In all periods studied, it was more likely to exceed the 10% p.a. average:
- 58.83% of 5-year periods
- 55.78% of 7-year periods
- 52.99% of 10-year periods
Notably, over 10% of 5-year periods produced returns over 20% p.a. These outlier years become less common as the timeframe extends, reinforcing the idea that very high recent returns often revert toward the average over time.
In other words, strong short-term performance shouldn’t be used as a guide for future expectations.
Lessons for Portfolios
The key takeaway from this analysis is simple: you’re unlikely to experience the average return – even over long periods. While long-term investing does reduce the range of outcomes, it doesn’t eliminate the chance of underperformance. In fact, history shows that nearly half of all long-term periods delivered returns below the average.
This has important implications for financial planning – particularly for modelling long-term outcomes like retirement. A reliance on average return metrics not only ignores the effect of sequencing risk but also underestimates how likely it is for portfolio returns to be lower than the average.
Since future returns are unknowable, a more prudent approach is to use conservative assumptions when projecting outcomes. Planning based on slightly lower-than-average returns helps create a margin of safety. While this historically would have led to an understatement of potential wealth, that’s a better problem to have than falling short.
