Jim re ".. over the past 100 years the stock market has delivered 9% to 10% average annual returns..."
That's a quote I've seen many times and its very misleading. The statistic is usually derived by looking at the S&P 500 or the Dow average. As we know, those averages are adjusted continuously to add up-and-coming companies and to delete those that are failing.
Jim:
Excellent points...
Even with the pruning of companies from the major indexes an investor putting money in S&P and Dow index funds would have benefited from the deletion of the failing companies and the addition of the up and comers...
Food for thought:
[[[Asset Allocation, Rebalancing and Long-Term Investment Returns
Predicting returns for passive and index portfolios, determining proper asset allocation mixes, and keeping portfolios in balance are frequently discussed issues in the world of passive and index investing. This article examines these issues.
Are Past Returns a Reliable Predictor of Future Returns?
A fully diversified passive and index portfolio should contain, according to many practitioners, broad coverage of all asset classes, and include a mix of large and small cap stocks, value and growth stocks, U.S. and international stocks, emerging markets, REIT’s (commercial real estate), and short to intermediate bond indexes or bonds. An asset class is a group of securities which have similar expected long-term returns and volatilities which are at least somewhat uncorrelated with other asset classes. Asset class diversification is both logically and empirically supportable, and passive portfolios spread across thousands of securities are certainly the best means to obtain it. Yet, there are frequently overlooked problems with asset class data and its application in portfolio design.
First, past performance data for many equity asset classes is not reliably available for more than two or three decades, and for many international classes, data goes back ten years or less. Investors should be cautious about assuming that such limited samples can be used to reliably predict future return rates and volatility (standard deviations) for individual asset classes, let alone mixes of asset classes where allocations are shifted up or down and hypothetical returns are correspondingly altered. To be statistically trustworthy, longer time frames, fifty or one-hundred years, must be sampled. Without getting into the statistics involved, it is important that investors understand that the relatively small differences in long-term returns between different equity asset classes, and high standard deviations within each asset class, make predictions of future comparative returns difficult.
It is virtually certain that over very long time-frames, say one-hundred years, stocks outperform bonds which outperform cash, and real estate performs, with somewhat less volatility, a little under stocks. Very approximately, after inflation, stocks can be expected to return 6-7%, bonds, 1-2%, and T-bills, 0.1-1%. Returns in the 1980’s and 1990’s were much higher than this, as interest rates came down from record double digit highs, boosting bond prices, and stocks averaged 18% per year, before inflation. Also, there is persuasive data suggesting that, over very long time-frames, small stocks outperform large stocks, value stocks outperform growth stocks, and international stocks in various categories smooth out long-term returns, if not adding to them.
But, as Keynes pointed out, over the long run we are all dead, and I know only one investor who has been in the market fifty years or longer. So, real investors must deal with the unpredictability of markets and returns over shorter time frames. And, the predictability of returns and volatility over shorter time frames, even twenty-five year time frames, involves uncertainty. Let’s take a look at some numbers.
Twice in the last century, 1900-1999, the S & P 500 lost approximately 65% of its value, adjusted for inflation and/or deflation, and took 15 years to produce a mere breakeven in returns. This occured during 1929-1943, and 1968-1983. Few investors hung around to hit the bottom, that’s why stocks lost so much of their value; too many sellers and not enough buyers. Once in the last century, 1929-1932, the Dow lost 86% of its value. And, whole generations learned to stay out of the stock markets. The situation was very much reversed as of January, 2000. The S & P 500 had averaged over 27% annually after inflation for 1996-1999, over three times the historical average return. Surveys done at the time indicated that most investors expected future stock market returns in the 15%-20%+ range, and faith in stocks, as well as price/earnings ratios, had never been higher.
Real returns differ substantially depending upon time of market entry, and, for the typical investor with a twenty or twenty-five year time frame, this can make very large differences in monies available for or during retirement. An investor in the S & P from 1929 through 1949 received an inflation adjusted return of 4.5%. Yet, an investor entering the market in 1932 at the bottom, and holding until 1951, received an inflation adjusted annual return of 10.84%, over 6% greater per year. Our 1929 investor received a compound total return of just 84.4%; our 1932 investor received a compound total return of 818.1%, ending up with almost ten times money as much as our unfortunate 1929 investor. Time of market entry matters, yet a lot of research suggests that markets cannot be timed.
What long-term returns should an investor realistically expect from a well diversified passive and index portfolio 100% invested in stocks? An honest answer would be somewhere between 4% per year and 7% per year after inflation, with a probable bonus of 1-2% per year for overweighting value and small equity asset classes. And, it could be better or worse, depending upon time frame and time of market entry and exit.
Past data also provide us with a rough rule of thumb for allocation decisions. We know that, adjusted for inflation and very long-term, stocks have returned about 6-7%, bonds 2%, and T-bills 0.1-1%. We also know that the worst likely multi-year decline in a stock portfolio is somewhere around 85%, that of the Dow from 1929-1933. If we compare returns for various allocations, starting with 100% in T-bills yielding 1%, each 20% increase in allocation to stocks adds about 1-1.20% to expected long-term returns, until we reach 100% in stocks, and an expected long-term inflation adjusted return of 7%. If we compare risk for various allocations, and define it as how much our portfolio might decline, with T-bills having no risk of decline, each 20% increase in allocation to stocks adds about 16%-17% to our worst possible portfolio decline, eventually reaching a worst possible decline of -85% for a portfolio 100% invested in stocks. Each investor needs to decide how much additional units of risk they are willing to endure for each additional increment of return. And, if future equity premiums decline, as several well-respected academic researchers in 2001-2 have suggested, returns relative to risk will decrease.
Given these sobering facts about relatively limited data on many equity asset classes, the unpredictability of equity returns for a typical investor's investment lifetime, and the problem with time of market entry, even passive investors need to be cautious about attaching excessive significance to past asset class data when constructing or comparing diversified passive portfolios. Diversification is smart, passive strategy is smart, long-term time frames are essential, but investors initiating portfolios today are probably best prepared if they exercise caution in projecting long-term returns. This issue is examined in additional detail elsewhere on this site in an article entitled, "Stocks for the Long Run?".
How Critical are Asset Class Allocation and Rebalancing to Long-term Returns?
Many advisors specializing in passive and index management promote the idea that frequent rebalancing is essential, and that it is one of the ways they add value for investors. Just how critical are shifting portfolio allocations across asset classes in determining long-term returns? Just how often should portfolios be rebalanced? Again, some hypothetical comparisons are helpful in answering these questions.
For a baseline portfolio, we’ll allocate 10% each to U.S. Large Growth, U.S. Small Growth, U.S. Large Value, U.S. Small Value, International Large Growth, and International Small Growth. The remaining 40% will be split equally between a one-year fixed bond portfolio and a five year fixed bond portfolio, producing the classic "normal" or 60/40 portfolio. These asset classes were chosen because data is available back to 1973, and include the 1973-1974 bear market. Now, let’s see what happens to returns for the period of 1973-1998 as we allow first the S & P (U.S. Large) to become overweighted, then multiple growth sectors, then value stocks, then small stocks.
A well allocated, balanced baseline portfolio provided a compound annual return of 13.28% with a standard deviation of 13.81%. As we let the S & P 500 become overweighted, ramping it up in 3% increments from 10% to 25% of total portfolio value, returns decline slightly to 12.83%, but so does risk, with the standard deviation dropping from 13.81% to 13.29%. Overweighting the three growth asset classes produces a return of 12.73% and standard deviation of 13.26%. Overweighting the two value asset classes produces a return of 13.63% and standard deviation of 14.16%. Overweighting the small caps produces returns of 13.65% and standard deviation of 14.26%.
As these portfolios became increasingly out of balance, little significant change occured. The range of returns was between 12.73% and 13.65%, with standard deviations (risk) increasing slightly as returns increased slightly. Considering the relatively limited data samples, I wouldn’t want to conclude that anything of statistical significance for an investor has occured here. And yet, 26 years is a long investment time frame for most investors, what I call an "investment lifetime".
Another way of looking at the rebalancing issue is to compare the effect of different rebalancing frequencies on compound total returns. Let's take a portfolio which is 70% in equities and 30% in fixed income, a 70/30 mix, with six different equity asset classes including growth and value, large and small, Reits, and U.S. and international asset classes. For the period of January, 1975 through December, 2000, monthly rebalancing produced a compound total return of 3923%, quarterly, 3959%, yearly, 3971%, and every other year, 4233%. This period was chosen because it allowed Reits and other equity asset classes to be included that could not be included if the analysis began in 1973. Thus, for this 26 year time-frame, more frequent rebalancing actually reduced long-term returns.
I then reran the original 60/40 portfolio data using Jan. 1972 through Dec. 2000, since that goes as far back as the data for this asset class mix is available, and adds in the peaky and very narrow, unbalanced markets of 1999 and 2000. Once again, portfolio returns and volatility differed very little as a function of rebalancing frequency. Contrary to some opinion, rebalancing frequency mattered little, and didn't increase volatility. In this case, frequent rebalancing actually increased volatility slightly for this 29 year period, 10.71% for quarterly rebalancing compared to 10.31% for biyearly rebalancing.
Truman Clark, of DFA, examined the issue of rebalancing in detail in three papers published online in Fall of 2001. He concluded that, "the proposition that a rebalancing strategy can increase expected return is dubious," and that "rebalancing costs definitely reduce expected returns." He cautions advisors that rebalancing "simulations with historical returns may provide misleading estimates of the benefits...," "don't employ naive, mechanical rebalancing rules," and "don't rebalance just to appear to be doing something."
Probably the best rule of thumb on rebalancing is to look at the overall stock/bond ratio quarterly, since it is the primary determinant of expected returns, and examine individual equity asset classes once a year, or so. Rebalance only when asset classes, and particularly, the equity/fixed ratio, gets out of balance far enough to produce a significant expected difference in returns.]]]
evansonasset.com
Jim |
