It Ain't Gonna Happen That Way

John P. Hussman, Ph.D.
All rights reserved and actively enforced.

I'm pleased to note that as of last Friday, the Strategic Growth Fund has appreciated in excess of 100% since its inception on July 24, 2000, including reinvested distributions, versus an annualized loss of just over 2% in the S&P 500 index. Since inception, the deepest pullback in Fund value has been less than 7%.

Past performance does not ensure future results, and there is no assurance that the Fund will achieve its investment objectives. Please see additional disclosures at the bottom of this page. Annualized total returns for the Strategic Growth Fund as of 5/31/05: 1-year 3.83%, 3-year 9.25%, since inception 15.13%.

---

The Buddha said "with our thoughts we create our world." Our self-talk - the phrases we quietly and often unconsciously say to ourselves, and the questions we ask - contain the concepts, filters, blinders, and occasionally, the crystal clear windows through which we observe and interpret reality. If we approach the world with expectations to be met, rather than an openness to accept (or at least begin with) what is there already, those expectations become the source of our own misery. If we form tight, detailed concepts of what we need to be happy, those concepts become the root of our unhappiness, because they shut out all of the opportunities to find happiness already around us.

There are two main questions I constantly ask myself as I manage the Funds from day-to-day. They're from chess: "what is the opportunity?" and "what is threatened?" Those bits of self-talk are my way of constantly getting back to the present moment, rather than getting caught up in hope about the future or distress about the past.

Another bit of self-talk comes from a sign that a veteran trader once had on his wall. It said, simply, "It ain't gonna happen that way." It's a constant reminder to allow for uncertainty, and not to form detailed "scenarios" about future market movements.

That one takes a while to learn. There's a constant tendency to want to plan out the market's future course, draw trendlines, plan on a strong second-half, etc. Just like our detailed concepts about what we need to be happy can often make us miserable when something seems missing, our scenarios about the specific future path of the market can drive us absolutely out of our minds if the market moves the "wrong" way.

Write this down. There is no "wrong" direction for the market. The market absolutely doesn't care about the scenarios investors have carefully planned for it. It ain't gonna happen that way.

It's important to understand that I'm talking about specific scenarios. There's nothing wrong with saying, for example, that stocks are priced to deliver unsatisfactory long-term returns. Or saying that, given some set of conditions, the market has historically performed well on average. When you phrase your investment views in terms of average outcomes and ranges of possible error, you're doing what you should be doing, which is carefully analyzing expected returns, and also allowing for uncertainty. In contrast, when you phrase your investment views in terms of what the market is going to do, in this particular instance, between now and some reasonably close future point in time, you've formed a scenario. The problem is simple - over short horizons (generally anything under a couple of years), the potential forecast error absolutely overwhelms any specific forecast you can make. It ain't gonna happen that way.

Again, it's fine to set an investment position on what you would expect on average, allowing for a large degree of uncertainty over short horizons. It's another (absolutely insane) thing to establish a substantial investment position that requires the market to do something specific over a limited horizon.

As I've noted before, even in the most favorable and unfavorable Market Climates we identify, the average expected return over a short horizon of say, a week, a month, or even a quarter is overwhelmed by the "standard deviation" of returns. However, as the number of instances increases, the average outcome starts looking much closer to the expected value. So it's enough to align our investment positions with the average return/risk profile we observe at every point in time, provided that we avoid positions that rely on a particular outcome. The key is to allow for uncertainty in every specific instance. That's what scenario-builders constantly fail to do.

Longer horizons, less uncertainty

Consider this principle as it applies to long-term returns. I've noted in prior updates that the probable range of returns on stocks over the coming decade is likely to be fairly unsatisfactory in comparison with typical returns that investors have earned (see for example the February 22, 2005 comment).

We know that the average price/peak earnings multiple on stocks has been about 14 in historical data. When earnings have been at a record, as they are now, the average multiple has been closer to 12, even if you restrict the data to periods of low interest rates and inflation. The historical median multiple is 11. The current value is 20.

We can go back over the data and calculate actual annualized returns over various horizons, say 2, 3, 4, 5, 7, 10, and 15 years, and compare those to the returns we could have expected , assuming that the market's P/E multiple moved from its initial level, whatever it was, to a terminal level of say, 15. We can also calculate the range of error (standard deviation) in these forecasts. Using the current multiple of 20 times peak earnings for the S&P 500, the resulting analysis is presented in the chart below.

There are three bars for each horizon. The purple bar is the projected annual return assuming a terminal price/peak earnings multiple of 15. The red bar is that return minus one standard deviation. It turns out that this is approximately the return that would be achieved with a terminal price/peak earnings ratio close to 11, which is the historical median. The green bar is the expected return plus one standard deviation, which in most cases corresponds to a terminal price/peak earnings ratio close to 20, which is the multiple we saw at the 1929, 1972, 1987 peaks, and currently. Of course, some observations have gone beyond one standard deviation from what an investor might have expected, which include the 1974 and 1982 troughs (about 7 times peak earnings), and of course, the 2000 peak (over 30 times peak earnings).

Notice two features of this chart. First, the return profile is negative for near-term horizons - there's a distinct negative tone to potential returns over the coming few years. As you get further into the future, cumulative earnings growth can be expected to keep long-term returns in positive territory even if P/E multiples decline to more typical levels.

It's very difficult to envision long-term total returns (from current valuations) of even 10% for any horizon in the foreseeable future. However, it's critical to remember that the likelihood of unsatisfactory returns from current valuations doesn't mean that stocks have to be a disappointing investment indefinitely: we could very well have a few poor initial years and then normal or excellent returns thereafter. Starting from lower initial valuations (which could be achieved over some shorter horizon), 10%+ long-term returns on stocks would become more plausible and even likely.

Second, note that the range of annual returns gets much narrower for longer-term horizons than for shorter-term ones. For 2-year returns, there is about a 30% range of likely annualized returns for the S&P 500. For 15-year returns, however, the range of probable outcomes is only about 4% wide. This is why I frequently note that valuation is an excellent indicator of potential long-term returns, but is relatively useless (at least by itself) as a shorter-term indicator.

In calculating the above returns, I chose a terminal P/E value of 15 not because it provides the best historical fit (which is provided by a terminal value closer to 12), but because I view it as plausibly optimistic while still being consistent with historical outcomes. Alternatively, you can get similar figures by assuming a lower terminal price/earnings ratio but earnings growth in excess of historical norms. In any event, the figures certainly are not based on pessimistic assumptions.

As for specific figures, probable 5-year total returns for the S&P 500 range between -2.1% and 6.4% annually, centered around 2.2%. That's an interesting figure, because the market's returns over the past 5 years have been about -2.2%. If the market's returns remain dull over the coming 5 years, the 10-year return as measured from the 2000 bubble peak would be about zero. That's precisely what one should have expected following the market's valuation at the time, and I absolutely view low single-digit total returns for the S&P 500 as plausible in the coming 5-year period.

On 10-year returns, the likely range is 2.8% to 7.4%, centered around 5.1%, which I again view as reasonable (if one uses a terminal P/E of 12 times peak-earnings, which is tighter on a historical basis, the center is slightly lower, about 3.1%). On 30-year total returns (not pictured), the likely range is between 6.4% and 7.8%, centered around 7.1% (about 6.6% if you use a terminal multiple of 12 instead of 15 - the terminal P/E doesn't make a great deal of difference at that horizon). That's a pretty tight range for 30-year returns, but again, it's based on actual historical standard deviations. The projected range of returns can also be derived algebraically, assuming 6% peak-to-peak growth in S&P 500 earnings (see again the February 22, 2005 comment), and future price/earnings multiples ranging from 11 to 20. Notice that if you hold the market's P/E multiple constant at 20 indefinitely, and earnings grow at 6% annually from peak to peak, then prices also grow at 6%. Add in the 1.8% dividend yield (which wouldn't change so long as dividends, earnings and prices grow in tandem), and there's the high-end projection of 7.8%.

Investment always involves uncertainty. The longer the horizon, the narrower the range of that uncertainty becomes. Detailed scenarios about near-term market direction are generally not useful. So for investors hoping for a strong second half starting in mid-July, or for a breakdown from a head-and-shoulders pattern to an October bottom, or for a tight trading range between 1150 and 1200 on the S&P 500, you may be right. But then, you're just as likely to be right playing roulette in Vegas. There's one outcome that's more probable than any of those scenarios: it ain't gonna happen that way.

Market Climate

As of last week, the Market Climate for stocks remained characterized by unusually unfavorable valuations and modestly unfavorable market action. On early strength last week, I closed out our moderate exposure to market fluctuations by selling the S&P 500 against a portion of our stock portfolio in the Strategic Growth Fund. It should be clear that I alter our investment positions frequently enough that these comments should not be used for trading purposes - particularly since I generally don't comment on changes until after the Fund's positions are established.

Presently, suffice it to say that I have no opinion at all on near term market direction. The Fund is substantially hedged based on prevailing valuations and market action, but we can't rule out the possibility that investors will adopt a more speculative attitude toward risk-taking.

In bonds, the Market Climate remains characterized by unfavorable valuations and fairly neutral market action. The Fed can be expected to raise Fed Funds another quarter of a percent on Thursday. Frankly, I think the Fed is irrelevant, but to the extent that investors believe it is powerful, we can expect some volatility on Thursday. Beats me which direction. Precious metals remain in a favorable Market Climate in my view, holding us to a roughly 20% position in precious metals shares in the Strategic Total Return Fund.