The use (and abuse) of short-term performance
The importance of using a variety of periods, risk adjustment, representative performance, and context
By John P. Hussman, Ph.D.
All rights reserved and actively enforced
As mutual fund managers, one of the most common questions we receive involves short-term returns. How should they be used when deciding which mutual funds to buy? At what point do flat or negative short-term returns become a sufficient reason to sell? These questions are really about information. Are there any good uses of short-term performance?
One of the key features of our investment approach is the view that market action does convey information. Specifically, market action carries information when it diverges in an important way from what would be expected, given a large amount of surrounding context and allowance for a reasonable amount of "random" fluctuation.
For example, suppose you believe that a certain stock is attractive. If the price of a stock is dropping sharply, but other stocks in the market and other stocks in the industry are performing well, that decline conveys information - somebody knows something you don't - and the stock should be approached defensively. In contrast, if the stock is declining in tandem with the market and other stocks in the industry, there is no negative information signal specific to that company, and the decline may be an opportunity to purchase an attractive stock at a discount.
As another example, suppose that a fund specializes in short sales. If the market advances sharply and the fund declines in value, the decline does not necessarily convey anything negative about that fund's management. However, if the market experiences a considerable and extended decline, and the short-selling fund fails to earn a positive return, the divergence between actual performance and reasonably expected performance is too large - it contains negative information that should be evaluated further. In this case, the performance is significantly out of context with what would be expected from the fund's strategy and prevailing market conditions, even allowing for a certain amount of randomness.
In short, context matters.
How should performance be used when buying or selling a fund?
First, raw short-term mutual fund performance is rarely informative. Statistically, funds with the highest one-year raw percentage returns have historically performed no better over the following year than funds picked at random. (In contrast, the poorest performers do have a slight statistical tendency to remain poor performers).
Mutual funds should certainly be held accountable for their performance. The issue is to clearly define what is a relevant measure of performance. In our view, the evaluation of performance should include:
No mutual fund, including the Hussman Funds, should be purchased solely on the basis of raw unadjusted performance, regardless of whether this performance is long-term or short-term. To do so ignores both risk and context. Most mutual funds are long-term investment vehicles. When you purchase a mutual fund, you are choosing an investment strategy. This choice should not be taken lightly.
You should invest in a fund only if you understand its investment approach, and then only if the expected return and risk from that approach are consistent with your financial objectives. Performance information can be helpful in evaluating the skill of the manager and the return/risk profile of the fund, but you should also read the Prospectus, as well as the investment manager's letter to shareholders in the annual reports. This additional information is available for every mutual fund, and is required by law. Use it wisely. It is difficult to properly interpret a fund's performance if you don't understand its investment strategy.
On the sell side, the following are some additional reasons for selling a mutual fund:
Occasional periods of negative performance generally do not provide sufficient reason to sell a fund. Poor risk-adjusted performance over a variety of time horizons is a stronger signal, particularly if that performance occurs during periods in which the fund's strategy should perform well. Again, the best inferences are based on a variety of time periods, adjustment for risk, representative performance, and context.
Don't mistake a bull market for genius
One of the most common mistakes investors make when evaluating mutual funds is to use samples that are not representative of the long-term return on stocks. For example, if you look at the period from a market trough to a market peak, the best performing funds will invariably be those that take a great deal of market risk ("beta") and invest in aggressive, often low-quality stocks. This sort of performance may have nothing to do with skill, as investors often discover during a subsequent market decline. In contrast, from a market peak to a market trough, the best performing funds will generally be the most defensive ones. Neither of these periods give you a good idea about the quality of the fund's management over the full market cycle.
As a rule, investors should always consider two periods when they evaluate mutual funds:
1) Performance from one market peak to the next, separated by at least a year (preferably 3-5). The best measure of long-term growth is to look from peak-to-peak (or trough-to-trough) over the full market cycle. Find the date of the most recent peak in the market. Then go back and find a market peak that occurred anywhere from one to five years ago. If you examine the returns on various mutual funds between those two points, as well as their volatility and the depth of their periodic losses, you'll learn a lot about the typical returns and risks that investors have experienced in each fund.
2) Performance over some period in which stocks earned 10-11% annualized returns. Again, the best way to measure the performance of a long-term investment approach is to select a period over which stocks delivered something close to their long-term rate of return. By examining periods like this, you'll get a much better idea about how the investment has performed during typical market conditions.
Performance information is useless without context
Consider the following hypothetical investment strategies and returns:
Over the 5-year market cycle, Strategy A earns an average return of 10% annualized. In contrast, Strategy B earns an average return of 17% annualized.
Despite the strong raw performance of Strategy B, there is not enough evidence to follow it without more information. Here's why. Suppose that Strategy A is a buy-and-hold approach on the S&P 500, and Strategy B is a large-cap growth approach committed to a fully invested position at all times. Given this context, the returns from Strategy B are so divergent from Strategy A that an investor should suspect that the portfolio is not sufficiently diversified, or is otherwise taking unusual risks. These risks may sharply reduce the statistical significance and reliability of returns from Strategy B.
In contrast, suppose that Strategy B is an approach that varies its exposure to market fluctuations depending on market conditions. Given this context, the year-to-year returns of Strategy B, despite occasional declines and lagging performance, do not convey negative information, because such a strategy would be expected to experience such periods from time to time. In the context of this strategy, investors would look for a reasonable long-term return, combined with the general avoidance of deep losses, as evidence that the strategy was being well-executed.
Chasing short-term returns can lead to long-term failure
Unfortunately, an investor who does not place short-term returns in context, and instead requires raw returns to be consistently positive and to outpace other alternatives, may destroy any chance at high long-term returns.
Starting in Strategy A and then switching to the best performer of the prior year leads to the following sequence of returns: -14%, -4%, +21%, +27%, +3%. The overall annualized return is 5.5%, far short of a committed approach to either strategy.
Starting in Strategy B and then switching leads to the following sequence: +19%, -4%, +21%, +27%, +3%. The investor earns a higher annualized return of 12.6%, but only because the investor was in Strategy B instead of A during the first year.
Equally important, an examination of historical year-over-year returns can substantially understate the amount of short-term volatility that an investment approach may experience. For example, while the two worst calendar-year returns for the S&P 500 in recent decades were a 22% loss in 2002 and a 26% loss in 1974, there have been many instances in which the pullback within some 12-month period (measured from peak-to-trough) has matched or far exceeded those declines. Needless to say, the risk of buying at peaks and selling at troughs is greatly increased when investors base their decisions heavily on short-term returns.
While the figures in the table are hypothetical, the lesson is not. Peter Lynch once remarked that despite the outstanding long-term performance of Fidelity Magellan under his management, shareholders as a group (on a dollar-weighted basis) only averaged a long-term return of about 7% annually. This is because investors systematically entered the fund after short-term advances, and liquidated after short-term declines.
Data from our own approach reinforce this lesson. Both in extensive historical tests, and in the actual performance of the Hussman Strategic Growth Fund (inception through December 31, 2002), a positive return in one month has been followed, on average, by another positive return. A negative return in one month has also been followed, on average, by a positive return. Most importantly, the strongest positive monthly returns, on average, occurred immediately after months in which returns were negative.
While there is no assurance that this pattern will continue in the future, and past performance does not ensure future returns, the lesson is obvious. Investors who regularly make investment decisions based on raw short-term returns will generally sabotage their long-term investment results, because short-term returns are unreliable (and often contrary) indicators of subsequent performance.
In our research, we have also found that the use of “stop-loss” rules based on short-term returns is generally inferior to the alternative, which is to sell lower-ranked investment holdings on short-term strength, and to buy higher-ranked investment holdings on short-term weakness. This is a strategy that we employ daily in the management of our own investment positions.
If investors toss a mutual fund into the “lower-ranked” group solely on the basis of weak short-term performance, the appropriate action would still be to sell that fund during some subsequent period of short-term strength. This is self-contradictory, because at that point, the basis of the sell decision would vanish. Raw short-term returns are not a useful basis on which to buy or sell a mutual fund. Short-term returns carry information only when they diverge from what would be expected given a fund's strategy, risk level, and prevailing market conditions, even allowing for a certain amount of randomness.
Again, key factors in evaluating fund performance include 1) a variety of time horizons, 2) adjustment for risk, 3) representative performance, and 4) context, which requires an understanding of the fund's objectives and overall market conditions.
Investing in the Hussman Funds
At the Hussman Funds, these views about performance translate into certain business practices. We don't advertise the Hussman Funds on the basis of short-term returns (even a year is fairly short-term from the standpoint of our approach). We impose a 1.5% redemption fee on shares held less than six months, specifically to discourage short-term investments. We refuse to charge sales loads, hidden “trailing fees,” or 12-b1 marketing fees. We don't want the Funds “sold” to our shareholders, particularly on the basis of short-term returns. We encourage shareholders to invest based on an understanding of our approach, and a good fit with their long-term investment objectives.
Accordingly, we provide our shareholders with frequent commentary and analysis aimed at increasing their understanding of our investment approach and the financial markets. By far, the best basis to purchase the Hussman Funds is an understanding of our investment strategy. Performance may help to reinforce that understanding, but should never substitute for it.
Our objective is long-term growth and total return, with an added emphasis on defending capital during unfavorable market conditions. The underlying strategy of the Funds is to accept those risks that are associated with some combination of favorable valuations and favorable market action, and to avoid, hedge, or diversify away those that are not. The principal sources of our returns (and risk) are security selection, the acceptance of aggressive investment positions during market climates characterized by both favorable valuations and market action, and the acceptance of defensive investment positions during market climates characterized by neither.
The objectives of long-term growth and total return can only be achieved through the willingness to take risk. The objective of defending capital during unfavorable market conditions can only be achieved through the occasional willingness to take investment positions that may not closely track the market. As a result, our investment objectives can be reasonably expected to result both in occasional periods of loss, and occasional periods during which our approach lags the major indices. Our hope is that our shareholders will understand our investment approach well enough to be patient during these periods.
The Hussman Funds are designed with the objective of substantially outperforming the major stock and bond market indices, with smaller losses, over the full market cycle (generally about 5 years in duration, combining both bull and bear markets). While there is no assurance that our objectives will be achieved, we do believe that they are reasonable and attainable. Our investment strategy is carefully designed, extensively tested, and we believe, well suited to meet these objectives.
|Home | The Funds | Open an Account | Account Access | Research & Insight | Site Map|
For more information about investing in the Hussman Funds, please call us at
513-587-3440 outside the United States
Site and site contents © copyright 2003 Hussman Funds. Brief quotations including attribution and a direct link to this site (www.hussmanfunds.com) are authorized. All other rights reserved and actively enforced. Extensive or unattributed reproduction of text or research findings are violations of copyright law.
Site design by 1WebsiteDesigners.