Conventional wisdom is that stocks are the best investment in the long run - made famous by Dr. Jeremy Siegal’s book, “Stocks for the long run”, first published in 1994 and currently in its fifth edition.
It makes sense on the face of it. The S&P 500 index has an average annual return of 9.5% over the past 91 years (1928–2018)*.
Now you would have had to sit through some pretty steep drawdowns over this period. Including seeing your investments fall 65% during the Great Depression in the 1930s and cut more than half during the Great Financial Crisis of 2008.
Nevertheless, $1 invested in the stock market at the end of 1927 would’ve grown to $3,828.50 by the end of 2018. Put a few more zeroes after that $1 and you are talking some serious money.
But most of us do not invest over 90+ years. Saving for retirement typically tends to be one of our longest goals and even then we’re talking about 40–50 years (assuming you start investing in your 20s and retire in your late 60s).
Several of our goals are shorter term, on the order of 3 to 10 years. For instance, saving for a new home or a car. Goals like saving for college are only slightly longer, between 10 and 18 years, depending on when the parent starts saving.
So does it still make sense to invest in stocks if you’re saving money for a short-term goal?
Given that the maxim ‘stocks for the long run’ was based on historical evidence, let’s do that same and look at how stocks have performed over shorter periods. With the obvious caveat that past performance is not indicative of future results.
I took historical returns for the S&P 500 stock index between 1928 and 2018 and calculated rolling 3, 5, 7, 10 and 20-year returns. For example, the first three-year period is 1928–1931, followed by 1929–1932, …. and finally ending with 2016–2018.
On the other end, the first twenty-year period is 1928–1947 and the last one is 1999–2018.
The following chart shows the percentage of periods that were positive for each investment horizon, along with the actual number of periods that were positive.Positive periods over various investment horizons for an all-stock portfolio (S&P 500 index).
One year horizons have the most uncertainty: 66 out of 91 years saw positive returns (73%), while 25 years saw stocks fall.
The frequency of positive periods rises for three-year horizons, to 74 out of 89 (83%). Of the 15 negative three-year periods, the worst one occurred during the Great Depression when saw stocks fell 62% (1930–1932).
Amongst five-year periods, there were 9 that saw negative returns. The worst of these occurred also came about amid the Great Depression (1928–1932) when the S&P 500 index fell 49%.
If you were investing $100,000 for a home down payment over five years, you wouldn’t want to find out that the account is worth less than $50,000 when you’re ready to buy the home.
Moving the investment horizon up to seven and ten years reduces the likelihood of negative periods. But they still occurred: 5 out of 82 ten-year periods saw stocks fall. Two of these include:
It’s only when you extend the investing time horizon to twenty years do you not get a single negative return period. The “worst” twenty-year period (1929–1948) saw an overall return of almost 60%, which is equivalent to a 2.4% annual return over that period. Not great but still positive.
In fact, the most recent twenty year period, 1999–2018, was amongst the five worst. Yet despite two market crashes (the tech bubble and the financial crisis), stocks still gained 195% — that translates to about 5.6% a year.
Hence stocks for the long run.
Though as we saw above, the uncertainty rises as the investing time frame shrinks.
If a 100% stock portfolio is not ideal for short term goals, the question is whether balancing stocks out with bonds works better. Bonds are typically expected to zig when stocks zag, and vice versa. So mixing the two should, in theory, lead to less frequent negative periods.
Similar to the earlier exercise, I looked at rolling 3, 5, 7, 10 and 20-year returns for a portfolio containing an equal 50%–50% mix of stocks and bonds**.
The following chart shows the percentage of periods that were positive for each investment horizon, along with the actual number of periods that were positive.Positive periods over various investment horizons for a balanced 50–50 stock/bond portfolio (S&P 500 index and 10-year US treasuries).
The 50–50 stock/bond portfolio fared better than a 100% stock portfolio over one-year periods. For one thing, 72 out of 91 years are positive (81%), compared to 66 positive years for just the stock portfolio. A bigger deal is that the worst year for the 50–50 portfolio saw a -23% return, compared to -44% for the all-stock portfolio (both occurred in 1931). So you can see how diversifying the portfolio with bonds helped in the past.
The 50–50 portfolio continues to do better as the investment horizon expands. Though even at the three and five-year time frames, a balanced portfolio of stocks and bonds did not have a 100 percent success rate. The worst negative periods for three and five-year horizons would have seen $1,000 cut down to $667 and $830, respectively.
Beyond seven years the balanced portfolio has never had a negative period. Of course, this is the past and there’s no certainty that the same will be true in the future.
So even a diversified 50–50 stock-bond portfolio may not be ideal for short term goals that are in the range of three to five years. Let alone an all-stock portfolio.
The probability of seeing your investment portfolio shrink may be small but there is an outside chance that it will happen, exactly when you need the money.
Instead, a liquid high-yield savings account may be the best option when putting aside money for a short-term goal.