I’ve heard people make statements like this: “I’m 50 and I haven’t saved anything for retirement. I wanted to send my kids through college and pay off all my debt first. I’m starting to put money in the stock market. I’m not worried because my house is paid for.”
The stock market is for the long run. If you’re 50 and you’re planning to work until you’re 80, past performance suggests you’ll be fine relying on stocks. Since most people aren’t going to work until 80, and for that matter won’t have the choice to work until 80, let’s consider shorter horizons. If you’re 50 and you plan to retire at 70, you don’t have a 20-year horizon if you’re starting to save now. Only the initial savings (a tiny part of your overall contributions before retirement) will have 20 years to grow. That shrinks if you plan to retire by 65, by which time most people have retired. Or 62, when you can take Social Security. Or 58, when you might retire involuntarily due to health or unavailability of jobs.
Let’s take a quick look at how $100,000 will perform at different time horizons if it’s 100% in the S&P 500. I’ll use FICalc with data starting in 1950. Here’s the range of inflation-adjusted outcomes at different horizons without taking any withdrawals.
| Horizon (Years) | Median | Low | High |
|---|---|---|---|
| 30 | 740,207 | 343,708 | 1,256,817 |
| 20 | 301,831 | 115,717 | 1,156,196 |
| 17 | 283,450 | 93,724 | 973,973 |
| 15 | 288,545 | 90,573 | 815,778 |
| 12 | 246,189 | 78,991 | 586,626 |
| 10 | 209,928 | 66,538 | 462,711 |
| 5 | 152,529 | 63,459 | 304,676 |
| 2 | 116,784 | 50,379 | 199,953 |
If you’re five years or less from retirement, you don’t want to rely too heavily on the stock market to help you out. The worst outcome was a loss of 37% of your initial investment. At ten years, you should expect your money to more than double, but you should prepare for a significant loss. At a 30-year horizon, even the worst outcome is very, very good. The median outcome is that your portfolio is 7.4x the starting value.
Let’s not be overly pessimistic here. At the longer horizons, the low outcomes are due to high inflation. For instance, at a 17-year horizon, there were two cases where the real value of your stocks declined, and those were for people that made the contribution in 1965 or 1966. We’re probably not going to see anything like those inflation rates in the absence of a major change in US institutions. The worst 17-year outcome not including any years from 1965 to 1979 was 2000-2016. You’d have put your \$100,000 in at the height of the 1990s tech bubble, immediately experienced a deep, years-long bear market, then a few years later the financial crisis, and your \$100,000 would have still turned into an inflation-adjusted \$149,952. Past performance is no guarantee and all that, but 17 years was a good time horizon in the past.
On the other hand, at a 10-year horizon, you would have had to worry about market crashes. 1999-2008 and 2000-2009 captured the worst parts of the lost decade and you’d have been in trouble if that was when a health issue caused you to stop working. Moving to a five-year horizon makes it even worse. There were lots of bad five-year periods with low inflation. Only about half the time would your portfolio gained 50%, and about 20% of the time it would have lost value. Stock market returns are not reliable over five years.
If you’re 40 or older and postponing saving for retirement, you’re losing your ability to leverage the stock market’s high average returns. If you’re 50 and you’re going to retire at 65, you don’t have a 15-year time horizon, you have a 15-year time horizon for the money you’re putting in the market today. You have a 10-year horizon for money you put in five years from now and a 5-year horizon for money you put in ten years from now.