Of Stocks and Funds

omparing the performance of individual U.S. stocks and U.S. equity funds has become a classic annual comparison.

As shown in Figure 1, Morningstar reported on 4,952 U.S.-based companies (i.e. stocks) which were in existence for the entire year of 2004. The mean one-year return was 24.04%. The median one year return (recall that median is unaffected by a small number of extremely high returns) was 14.30%. The share-weighted one year return was 13.79%. This share-weighted figure is not reported by Morningstar. The share-weighted return (as calculated by the authors) assigns a proportionally greater weight to stocks that have more shares outstanding and less weight to thinly held stocks. It is only calculated for a one-year period.

It should be noted that all the data in this analysis are affected by survivorship bias inasmuch as only the performance of survivors at the end of the period is measured. Thus, as a general rule, the returns shown here are actually higher than experienced by those investors who owned stocks that disappeared during the time frames being measured.

The U.S. equity market in 2004, as measured by individual securities, was reasonably rewarding. Two-thirds of the 4,952 stocks had a positive return by year end 2004. The average return for the 3,317 stocks with a positive return was over 48%. On the other hand, there were 1,635 stocks that had a negative return in 2004. The average return of that group was -24.89%. Even during a relatively encouraging year for equities, there was a sizeable weak spot. The best performing stock in 2004 had a return of 2,260%, while the worst performing stock lost over 97%. There were 8 stocks (in the Morningstar Principia database) that lost in excess of 90% during 2004.

Over the 3 year period ending 12/31/04 the mean return for the 4,709 individual stocks which survived the entire 36 months was 15.66%. The median return over the 3 year period was 14.06%. There were 1,320 stocks out of 4,709 (or 28%) that had a negative average annual return over the three year period. The average return of that group was -18.97%. The average return of the 3,389 stocks that had a positive three year return was 29.14%. The highest 3 year annualized return was 350.55%, while one stock had an annualized loss of -82.85% between 2002 and 2004.

Over the 5 year period we observe that 33% of the 4,317 stocks that survived the entire period had a negative return. The average negative return was a discouraging -19.02%. The average 5 year annualized return for the 67% of stocks that had a positive return was nearly 21%. One stock had a five-year annualized return of 135.82%. On the other end of the spectrum, one poor soul had a 5 year annualized return of -82.11%.

Finally, over the 10 year period ending on December 31, 2004 the average return of the 2,781 stocks that survived the entire period was 10.79%. The median return was 11.81%. The highest 10 year annualized return was 68.83%. However, that level of annualized return is misleading. The standard deviation of return for that particular stock was enormous due to dramatic swings in return from year to year. For example, that same stock had a 3 year annualized return of -1.13%. Of the 2,781 stocks which survived the entire period, 16% had a negative 10 year annualized return.

Now, let's look at U.S. equity mutual funds (see Figure 2). There were 2,592 that survived the entire year of 2004. Only domestic equity funds with at least 12 months of performance as of 12/31/2004 were included in the analysis. Funds with more than 15% of their portfolios in cash, bonds, or non-U.S. stock were removed from the study. Redundant funds were also omitted (i.e. only "Distinct Portfolios" were selected) from the Morningstar Principia data.

The mean one-year return of U.S. equity funds in 2004 was 12.84%. The median return was 11.96%. The net-asset weighted one-year return (also calculated by the authors) was 12.46%. The net asset-weighted equity fund return is comparable to the share-weighted stock return calculation, where the one year return is weighted according to the proportion of each fund's net assets compared to the total net assets of the 2,592 funds.

For example, the largest U.S. equity fund included in this data set was Vanguard Index 500 with net assets of $84.2 billion. Total net assets for all 2,592 funds summed to $2.16 trillion. Vanguard 500's percentage of the total U.S. equity fund base was 3.893%. Thus, the 10.74% return of Vanguard Index 500 in 2004 was multiplied by .03893 to determine its asset-weighted return. This process was repeated for all 2,592 funds to determine the total asset-weighted return of U.S. equity funds in 2004. (The share-weighted return for stocks and the net-asset weighted return for funds are only calculated over a one year period due to the potential fluctuation in outstanding share totals and net assets over a period longer than one year.)

That the mean return of 12.84% and net asset-weighted return of 12.46% were so similar in 2004 indicates that large funds, on average, performed similarly to funds with smaller asset bases. Simply put, the big funds did about as well as all equity funds in general.

Only 49 funds (just under 2% of the group) had a negative one-year return. The average negative return of those lost souls was -5.03% while the average return of the 98% of funds that had a positive return was 13.2%. The best U.S. equity fund in 2004 generated a total return of over 57% while the worst performing fund lost nearly 21%.

One of the 49 funds that found a way to generate a negative return in 2004 deserves special attention. That fund is Frontier Equity. In 2003 it had an astonishingly poor return -44.23%. True to form, Frontier Equity turned in another terrible year in 2004 with a return of -6.90%.

Over the past three year period (as of 12/31/04) the performance of U.S. equity mutual funds was mediocre to poor. Eighteen percent of all U.S. equity funds had a negative 3 year annualized return. The mean return of the 2,332 funds that survived the three year period from 2002-2004 was 5.49% and the median return was 4.35%. Thus, 50% of all U.S. equity funds had a 3 year annualized return of 4.35% or lower as of December 31, 2004.

An anomaly is exposed in the 5 year time period: 48% of 1,849 equity funds had a negative five year average return whereas only 33% of the 4,317 individual stocks generated a negative 5 year return. This is a potent reminder of the redundancy that exists in the mutual fund marketplace. Very bluntly, too many funds invest in the same large-cap stocks. When many of the popular growth stocks began capsizing in 2000 and 2001 the meltdown in the vast majority of growth mutual funds was virtually assured -- precisely because of the huge amount of stock overlap in the growth fund sector. As a result, the median 5 year annualized return for 1,849 U.S. equity funds was 0.53%. Nevertheless, there are always heroes. One fund generated a 5 year return of 37.21% -- the enigmatic Bruce Fund.

When examined over the past 10 years, mutual fund performance generally returns to a respectable level. The mean return for the 944 U.S. equity funds which were in existence for at least the full ten year period was 11.21%, while the median return was 11.24%. Even more encouraging, only 2 funds (or 0.20% of the group) had a negative annualized return over the 10 year period compared to 16% among individual stocks. The highest 10 year annualized return was a stunning 26.35% by Bridgeway Ultra Small Company, which is currently closed. The worst performing fund, with a horrid 10 year annualized return of -27.08%, was (drum roll) Frontier Equity. Is it possible that this fund single-handedly disproves Darwin's theory of survival of the fittest?

The results of this analysis reinforce two important axioms: First, the maximum return of an individual stock will always be higher than the best performing equity mutual fund and the worst performing stock will always have a lower return than the worst performing mutual fund. The parameters of individual stock performance are much wider than equity fund performance.

Secondly, there will be periods when the portfolio effect (theoretically achieved within a mutual fund) can actually magnify losses. As already noted, the cause of this is redundancy in the holdings of many U.S. equity funds. This phenomenon was observed during the five year period. More mutual funds had negative returns than individual stocks. Furthermore, the mean stock return was 7.76% while the mean equity fund return was an anemic 1.42%. There is not always security in numbers (i.e. mutual fund portfolios), at least not when the numbers are market-cap weighted!

In defense of diversification, far fewer mutual funds had negative returns during the one, three, and 10 year time periods. It is interesting to note, however, that individual stocks had higher median return during the one, three, and 10 year time frames. Protection against loss can sometimes insulate against return. Diversification is a double-edged sword. These results reveal both edges.

We conclude with a brief discussion on a related topic, namely the use of benchmark returns to make sense of performance in the U.S. equity markets. Figure 3 includes three common benchmarks. As can be seen, at year-end 2002 the S&P 500 Index had a one-year return of -22.1%, in 2003 a return of 28.7%, and in 2004 a gain of 10.9%. Both the midcap (S&P Midcap 400) and small cap (Russell 2000) indexes demonstrate similar volatility in annual returns over the past 3 calendar years.

Using short-run performance data to manage long-term portfolios (as far too many 401(k) participants do) can lead to reactionary, non-strategic management with the ironic result of investors "churning" themselves. Moreover, if too much attention is paid to the components of an equity portfolio (e.g. the performance of the individual stocks or funds in the portfolio) the most important issue can be overlooked, namely the performance of the portfolio. The performance of a portfolio is more important than the performance of any of its components.

Rather than gauging the performance of the U.S. equity market by using one equity index (a component of the market) it is more realistic to combine several indexes (a market portfolio of indexes). An example of this is shown in Figure 4. The return data for each of the three indexes in Figure 3 was weighted by .3333 (and then summed) to calculate the performance of a "portfolio" of indexes over the various time frames. These results better represent the performance of portfolios which have been diversified by market cap - a strategy that investors are increasingly sensitive to. Perhaps more importantly, presenting U.S. equity returns in a portfolio context, rather than as isolated components (i.e. individual indexes), may motivate a more holistic approach to investing. Investors may begin to focus more on the performance of their portfolio as a whole, rather than obsessing on the short-run, "under-performance" (and that's a relative term) of individual components within the portfolio.

A helpful analytical technique to minimize this type of reactionary micro-management is to make use of rolling returns in an attempt to smooth potentially volatile annual performance data. Having said that, notice that the 3 and 5 year rolling returns for each index in Figure 3 are still very susceptible to large gyrations from year to year. For instance, the S&P 500 had a 3 year rolling return of -14.6% as of 12/31/02, -4.1% by 12/31/03, and 3.6% at year-end 2004. The Russell 2000 Index had erratic rolling 3 year returns of -7.5%, 6.3% and 11.5%.

Only when using 10 year rolling averages do we see a marked reduction in return volatility from period to period in Figure 3. The S&P 500 rolling 10 year returns are 9.3% as of year-end 2002, 11.1% as of year-end 2003, and 12.1% as of year-end 2004. The midcap index demonstrates a bit more volatility in the 10 year rolling returns, with the small cap index demonstrating the greatest amount of variance in rolling 10 year returns (as expected). In each case, the equal-weighted index returns reported in Figure 4 have a smoothing (and potentially soothing) effect on our perception of equity market performance.

Equity investors absolutely must understand that downside risks are very real inside of 5 years, or even longer. Generally, however, over periods of at least 10 years the excesses on both the upside and downside are absorbed into the mean. Keeping clients on-board during rough periods is a function of educating them that such periods will occur. More important, however, is centering attention on the long-term, central tendency of diversified equity returns rather than equity returns that are heavily influenced by large-cap stocks (which is the case for most U.S. equity indexes and many U.S. equity mutual funds).

If not put into a long-term context, volatility of return can easily be used by clients as a rationale for gutting their portfolios. If investors will not make long-term commitments they will repeatedly experience periods of self-inflicted mediocrity. Educating clients about the stability of capitalization-diversified U.S. equity returns over the long-term is the antidote to the sensational news of the day that causes them to abandon ship when only a slight course correction may have been needed.

Figure 1. Individual U.S. Stocks

Figure 2. U.S. Equity Mutual Funds

Figure 3. The Rolling Terrain of Return

Figure 4. A Portfolio of Benchmarks

____________________________________________________________________________________
Craig L. Israelsen, Ph.D. is an associate professor at Brigham Young University. He teaches family finance in the Department of Home and Family Living. His research interests include mutual fund analysis. He writes monthly for Financial Planning magazine.

____________________________________________________________________________________
Amber Clement is an undergraduate at Brigham Young University studying Family Finance.