The U.S. stock markets have been soaring – with a few hiccups – ever since the financial crisis. The S&P 500 index reached a new intraday high on Wednesday, breaching 3,000 for the first time before closing at 2,993 – and now stands at more than four times its level when it bottomed out in March of 2009. (The Canadian markets have not done nearly as well.)
To truly appreciate how well U.S. stocks have done, you have to consider what the returns have been over the longer term. The average return on U.S. stocks from 1938 to 2018 was 7.5 per cent per annum. This figure is a real return, meaning that I have netted out the effects of inflation. To put this into context, one dollar invested in stocks in 1938 would now be worth more than 6,000 times as much in real terms.
As good as that is, a 7.5-per-cent real return pales in comparison with the past decade. Over the 10-year period ended Dec. 31, 2018, the S&P 500 produced an average real return of 12.6 per cent. But is this sustainable, especially when all the potential good news (such as future rate cuts) seems to have been priced in? In other words, should we be thinking about selling off our equity investments?
As an actuary, I would normally advise long-term investors to remain fully invested at all times, especially if they are saving for retirement. Trying to time the market has proven to be a mug’s game. No one has demonstrated they can do this consistently and the more successful investors such as Warren Buffett don’t even try. Nevertheless, a real annual return of 12.6 per cent seems so high that it is natural to wonder whether normal investment rules should be thrown out the window.
With this thought in mind, I sifted through historical data to answer the question: Is a decade of above-average returns typically followed by a decade of returns that are below average? This seems to be a reasonable premise given that market returns have a tendency to revert toward their long-term averages. Put another way, the elastic band that we know of as stock prices can only be stretched so far before it snaps.
The historical data are decidedly mixed on the question of reversion toward the mean. Whenever the average real return in a given 10-year period was above the long-term average (7.5 per cent), the corresponding return for the subsequent 10 years was below average only half the time. The other half of the time saw the subsequent 10-year period producing above-average returns.
A chart may make this finding a little easier to digest. The table shows every 10-year period since 1938 in which the real return was above the long-term average. For each such period, the table also shows the average real return in the subsequent 10-year period. Really good returns in the following 10 years are shown in green and exceptionally poor returns are shown in red.
High returns no reason
to run for the hills
A mixed result on reversion toward the mean
Periods above
avg. returns
Subsequent
10-year avg.
Periods above
avg. returns
Subsequent
10-year avg.
1941-1950
13.1%
1975-1984
11.1%
1942-1951
15.5%
1976-1985
11.1%
1943-1952
13.1%
1977-1986
11.8%
1944-1953
15.3%
1978-1987
16.0%
1945-1954
12.1%
1979-1988
19.5%
1946-1955
9.8%
1980-1989
18.6%
1947-1956
8.4%
1982-1991
14.7%
1948-1957
11.5%
1983-1992
9.6%
1949-1958
8.6%
1984-1993
8.9%
1950-1959
6.3%
1985-1994
8.4%
1951-1960
5.8%
1986-1995
5.1%
1952-1961
3.4%
1987-1996
4.5%
1953-1962
5.4%
1988-1997
0.1%
1954-1963
0.9%
1989-1998
-5.6%
1955-1964
-4.6%
1990-1999
-6.1%
1956-1965
-3.0%
1991-2000
-4.6%
1957-1966
-0.1%
1992-2001
-3.5%
1958-1967
-2.7%
1993-2002
0.5%
1959-1968
-2.8%
JOHN SOPINSKI/THE GLOBE AND MAIL
SOURCE: fred Vettese
High returns no reason to run for the hills
A mixed result on reversion toward the mean
Periods above
avg. returns
Subsequent
10-year avg.
Periods above
avg. returns
Subsequent
10-year avg.
1941-1950
13.1%
1975-1984
11.1%
1942-1951
15.5%
1976-1985
11.1%
1943-1952
13.1%
1977-1986
11.8%
1944-1953
15.3%
1978-1987
16.0%
1945-1954
12.1%
1979-1988
19.5%
1946-1955
9.8%
1980-1989
18.6%
1947-1956
8.4%
1982-1991
14.7%
1948-1957
11.5%
1983-1992
9.6%
1949-1958
8.6%
1984-1993
8.9%
1950-1959
6.3%
1985-1994
8.4%
1951-1960
5.8%
1986-1995
5.1%
1952-1961
3.4%
1987-1996
4.5%
1953-1962
5.4%
1988-1997
0.1%
1954-1963
0.9%
1989-1998
-5.6%
1955-1964
-4.6%
1990-1999
-6.1%
1956-1965
-3.0%
1991-2000
-4.6%
1957-1966
-0.1%
1992-2001
-3.5%
1958-1967
-2.7%
1993-2002
0.5%
1959-1968
-2.8%
JOHN SOPINSKI/THE GLOBE AND MAIL, SOURCE: fred Vettese
High returns no reason to run for the hills
A mixed result on reversion toward the mean
Subsequent
10-year avg.
Periods with above
avg. returns
Periods with above
avg. returns
Subsequent
10-year avg.
1941-1950
13.1%
1975-1984
11.1%
1942-1951
15.5%
1976-1985
11.1%
1943-1952
13.1%
1977-1986
11.8%
1944-1953
15.3%
1978-1987
16.0%
1945-1954
12.1%
1979-1988
19.5%
1946-1955
9.8%
1980-1989
18.6%
1947-1956
8.4%
1982-1991
14.7%
1948-1957
11.5%
1983-1992
9.6%
1949-1958
8.6%
1984-1993
8.9%
1950-1959
6.3%
1985-1994
8.4%
1951-1960
5.8%
1986-1995
5.1%
1952-1961
3.4%
1987-1996
4.5%
1953-1962
5.4%
1988-1997
0.1%
1954-1963
0.9%
1989-1998
-5.6%
1955-1964
-4.6%
1990-1999
-6.1%
1956-1965
-3.0%
1991-2000
-4.6%
1957-1966
-0.1%
1992-2001
-3.5%
1958-1967
-2.7%
1993-2002
0.5%
1959-1968
-2.8%
JOHN SOPINSKI/THE GLOBE AND MAIL, SOURCE: fred Vettese
For the eternal optimists, the many green periods in the chart give some hope that we may be looking at another 10 good years while conservative investors may be more inclined to focus on the red periods. The point, though, is that no future result is preordained solely on the basis of past returns.
That said, there are two important caveats. First, 81 years of data may seem like a lot but remember that the 10-year periods shown are highly overlapping. Second, my analysis was purely technical and totally ignores the potential of fundamental factors such as climate change, an aging population and trade wars to reverse any trend, good or bad. All we can conclude is that the high returns of the past 10 years are not sufficient reason on their own to prompt investors to sell off their U.S. stock portfolio.
Frederick Vettese is the former chief actuary of a major consulting firm and the author of Retirement Income for Life.