Copyright 2020 Tom Madell, PhD, Publisher
Jan 2020. Published Jan. 6, 2020.
By Tom Madell
Article Summary: Based on 27 years of continuous data, five year periods of strong stock fund returns for two highly representative funds are surprisingly strongly negatively related to subsequent five year returns. If the relationship carries forth into the future, one can anticipate mid-single digit returns as the most likely average broadly diversified stock fund return for both US and international stocks.
As we start a new year, or even the upcoming decade, it may be an appropriate time to reflect on your investments.
As you undoubtedly know, the prior decade was an outstanding time to be a stock investor, especially if the majority of your investments were in US stocks as opposed to international stocks; the latter also did well over the prior decade, although a little more so over the first half of the decade than the second.
Are there any lessons that can be learned by examining the behavior of stocks over such extended periods, such as over the previous five years?
As I have continually emphasized on my website, no one, no matter how expert, can consistently predict correctly the future performance of stocks, or the factors that seem to influence whether they go up or down. But we do know that returns show that, in the past, stocks have gone up each year about two-thirds of the time. This makes an investment in stocks at the beginning of any upcoming year to have a better than average chance of turning out positively. When you multiply the two-thirds odds over five to 10 years, investing in stocks, more than likely (but never guaranteed), will turn out to be a good choice. Of course, this means that one-third of the time, you may expect to see losses over the short term.
While you should therefore disregard any predictions of where stocks will go over the next year (except while acknowledging the two-thirds chance they will rise to some degree), it might be the case that if one focuses on successfully predicting the longer term, say five years, using multi-year intervals of past performance might turn out to actually predict multi-year future performance. If so, long-term investors might be afforded a much better idea of what to expect if they continue to hold their stock positions beyond a single year going forward, or if they decide to reduce their positions.
The data I present below analyzes US and international stock returns going back to 1993 up through the end of 2019 in order to see if there is a relationship between the prior five years of returns and the subsequent five years.
It is pretty common knowledge that past returns cannot predict what the next 12 months will be. There are too many variables and unknowns to successfully predict what is likely to happen over the next year.
But it is possible that a strong decade of returns, or even a half decade, suggests that the next decade, or half of one, will lead to the opposite results, that is, much lower returns than we experienced between 2010 and 2019. Why? For at least two reasons. Since stocks, on average over long periods, tend to produce returns of about 9 to 10%, when stocks have done even better over the prior 10 years, it is statistically more likely that they will return to their average, or even lower to restore this long-term average.
A second reason is this: the strength of the economy, which has a lot to do with overall stock market performance, tends to run in up and then down cycles.
Typically, a "good" economy tends to last somewhere around 5 or even 6 years, but then is followed by a not-so-good economy. The not-so-good economy, or even a recession, typically hurts stock prices and may bring average overall stock performance down for several years before returning to a good economy again.
A final reason why stocks go for a number of years doing quite well but then seem to suffer may be that as stock investors enjoy excellent returns, psychology may dictate that if troubling signs emerge, these investors begin to fear that they will lose the gains they have made. As a result, they have a tendency to sell some (or all) of their positions more readily than in an up period absent such signs. A vicious cycle can emerge causing previously unfearful investors to get nervous and sell too. But as time progresses, selling becomes depleted and astute investors begin to buy what are now perceived as bargain prices.
The following two tables show the returns of, first, an unmanaged index fund of the entire US stock market, the Vanguard Total Stock Market Index Inv (VTSMX) fund, and next, a managed fund of international stocks, the Vanguard International Growth Inv (VWIGX) fund, starting the first full year of the former's inception in 1993 through the end of 2019. (All returns described in this article are annualized.)
Rather than presenting the returns chronologically, I have listed the five year back-looking returns from highest to lowest and then compared them to what each fund earned over the following 5 years.
At the very bottom of each table is shown the return for the just completed five years, 2015 to 2019. However, no subsequent returns can be known until five years from now, that is at the close of 2024.
5 year span Return 5 years later Return
1995-1999 26.8 2000-2004 -1.4
1994-1998 21.5 1999-2003 0.5
1993-1997 18.9 1998-2002 -0.8
2009-2013 18.7 2014-2018 7.8
1996-2000 16.7 2001-2005 2.0
average = 20.5 average = 1.6
2010-2014 15.6 2015-2019 11.1
2003-2007 13.8 2008-2012 2.2
1997-2001 9.7 2002-2006 7.4
2002-2006 7.4 2007-2011 0.2
average = 11.6 average = 5.2
2006-2010 2.9 2011-2015 12.0
2008-2012 2.2 2013-2017 15.4
2001-2005 2.0 2006-2010 2.9
2005-2009 0.9 2010-2014 15.6
1999-2003 0.5 2004-2008 -1.8
2007-2011 0.2 2012-2016 14.5
1998-2002 -0.8 2003-2007 13.8
2000-2004 -1.4 2005-2009 0.9
2004-2008 -1.8 2009-2013 18.7
average = 0.5 average = 10.2
2015-2019 11.1 2020-2024 ???
5 year span Return 5 years later Return
2003-2007 21.9  2008-2012 -1.4
2009-2013 15.8  2014-2018 3.5
1995-1999 15.2  2000-2004 -0.5
1993-1997 14.9  1998-2002 -2.1
2002-2006 13.8  2007-2011 -2.0
average = 16.3 average = -0.5
1994-1998 10.1  1999-2003 0.7
1996-2000 10.0  2001-2005 4.2
2010-2014 6.8  2015-2019 10.6
2006-2010 5.7  2011-2015 3.6
2005-2009 5.6  2010-2014 6.8
2001-2005 4.2  2006-2010 5.7
average = 7.1 average = 5.3
1997-2001 2.7  2002-2006 13.8
2004-2008 1.9  2009-2013 15.8
1999-2003 0.7  2004-2008 1.9
2000-2004 -0.5  2005-2009 5.6
2008-2012 -1.4  2013-2017 10.9
2007-2011 -2.0  2012-2016 7.1
1998-2002 -2.1  2003-2007 21.9
average = -0.1 average = 11.0
2015-2019 10.6 2020-2024 ???
To summarize what these tables show is that for both funds, initial periods of high five year periods of returns are followed almost always over the next five years by low returns. On the other hand, initial low periods of returns are often, but not always, followed by high returns.
The average subsequent five year returns for initial high, medium, and low performance five year periods are shown below the actual returns. As can be seen for both VTSMX and VWIGX, after periods of high returns, the subsequent five year returns were only 1.6% and minus 0.5% respectively. For the lowest returns over 5 years, the subsequent 5 year returns averaged 10.2 and 11.0% respectively. When the initial 5 year returns were in between high and low (medium), the returns were 5.2 and 5.3% respectively. While the groupings of returns as high, medium, and low were defined somewhat arbitrarily, the fact remains the both sets of subsequent returns were rather surprisingly strongly inversely related. In fact, in a statistical test of the chance such a relationship could have occurred by chance, the results showed that this degree of negative relationship was strong enough not to have occurred most likely just by chance.
While these previous returns cannot guarantee that returns in the years 2020 to 2024 will follow the same inverse relationship, that is, in this case be similar to medium 2015-2019 returns of 11.1 and 10.6 respectively, it would appear likely from the data that subsequent 5 year returns will be more likely in the mid-single digits than either particularly high or low. You should remember too that these results are based on returns from a half decade or 5 years. Therefore, the return for 2020, may still turn out to be either high or low. But when the entire 5 year period of 2020 to 2024 is considered, it seems quite likely than the returns will average out as reflected in the above tables. This means that since for both funds the 2015 to 2019 returns were near the 9 to 10% range, subsequent returns will also fall in a moderate range, neither very high nor very low.
My recommended funds had another good year. Click here for details.
To sign up to receive notification of new Newsletter postings, click here.
If you find this month's Newsletter helpful, consider pasting this link to it in an email to a friend: