The first decade of the 21st century was lost for many stock investors. For the 10-year period beginning December 31, 1999, the typical S&P 500 Index fund, including reinvested dividends, generated annualized returns of about -1 percent.
Two major periods of losses occurred during that time. The first lasted from August 2000 until September 2002 and saw the S&P 500 Index decline by nearly 50 percent. It took four years to get back to even. The more severe decline began at the end of October 2007, taking the index down almost 57 percent by last March. Despite the strong run-up since, the market is still 30 percent below its peak.
In contrast, bond investors were well rewarded for the decade. One purchasing a 10-year Treasury bond at the end of December 1999 would have locked in an annual yield of about 6.4 percent. Bond fund investors did equally as well. Here are 10-year annualized returns at the end of December 2009 for some popular bond funds:
Total Bond Market…………….+6.06 percent
Long-Term Treasuries……….+7.18 percent
Long-Term Corporates………+7.19 percent
Inflation-Protected…………….+6.80 percent
Had you asked 100 investment advisers at the end of the 1990s what returns they expected from stock during the ensuing 10 years, I suspect 90 or more would have responded with a number somewhere between +8 percent and +10 percent annually. How could they all have been so wrong?
Most would have relied on past data to support their expected return estimates. Ten years ago, the data on the modern era of equity markets would indeed have shown an average annual return in that range. But reliance on any given historical period of price movements in estimating returns from any asset class is an exercise in futility.
Consider gold. One can place some gold bullion in a safe and come back 10 years later. Whether one will have an investment return or not will depend solely on the supply and demand for gold at that time, which, in turn, will completely depend on many factors external to gold. How the expected returns from gold used in making your investment decision compare to your actual returns during a 10-year period will depend entirely on what historical period you selected.
Expected returns for bonds, in contrast, are pretty simple to understand: They’ll equal the interest rate on the bonds you purchase, since (by definition) a bond will return your principal at maturity. And the best estimate of interest rates going forward is the current one. So with 10-year treasuries today, you can expect to experience annual returns equal to the current interest rate of about 3.75 percent over the next 10 years.
Rob Arnott, developer of fundamental indexing, has looked at the past decade and concluded it wasn’t as lost for stocks as it might seem. The problem, according to Arnott, is that the design of most broad stock indexes guarantees you’ll commit too much of your money to overpriced stocks and too little to underpriced ones. The design to which Arnott refers is known as capitalization-weighting, or cap-weighting for short. The S&P 500 Index and most other commonly used broad stock market indexes are designed to capture, in one number, the price movements of hundreds or thousands of stocks. It matters how one averages out those numbers.
With cap-weighting, the movement of each stock has an impact on the index proportional to its capitalization, defined as the number of shares outstanding times the current price. Tech giant Microsoft, for example, currently has just under 9 billion shares of stock outstanding, giving it a market capitalization of $257 billion at a recent price of $29 a share. Micros Systems, which produces software and management systems for hotels and specialty retailers, has about 73 million shares outstanding, giving it a market capitalization of about $2.4 billion at a recent price of $33 a share. A 1 percent move in Microsoft will affect the S&P 500 Index more than 100 times as much as a 1 percent move in Micros.
As a result of cap-weighting, pegging your stock investments to a major index like the S&P 500 hitches you to a momentum-based investment strategy. The more popular a stock becomes, the more of it you hold, and vice-versa. According to Arnott, someone investing in a cap-weighted stock index at the end of the 1990s was buying stocks at a price earnings ratio of 44, far above the long-term average, distorted by a small number of popular tech stocks.
Arnott advocates fundamental indexing as a cure for cap-weighting. By this term, Arnott refers to tailoring your holdings of each stock in an index according to its place in the overall economy as measured by sales, book value or other fundamental factors. Arnott says a fundamentally indexed stock portfolio would have generated a 7 percent annual return for the lost decade. You can check out Arnott’s views at http://www.rallc.com/ideas/pdf/Fundamentals_201001.pdf.
Our proprietary Dividend Growth Portfolio uses a different approach to avoid the momentum-driven nature of cap-weighted index funds. In determining an expected return for any given stock, we use a bond-like analysis, adding the dividend yield to the expected rate of growth for the dividend. Audited returns for the lost decade for our portfolio show an average annual return of just under 6 percent, net of fees. While we haven’t been spared serious declines in both the 2000-2002 and 2007-2009 bear markets, those declines have been less severe and recovery has come more quickly.
As investments, stocks are risky. Holding less risky stocks—and watching what you pay for them—will go far in avoiding future lost decades.
