The Likely Range of Market Returns in the Coming Decade

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

Just a note - the semi-annual report of the Hussman Funds is now available online. The report includes a letter to shareholders, performance graphs, portfolio composition as of December 31, 2004, and other information. As always, shareholders are encouraged to spend some time reading our reports and Prospectuses. The report is currently in press, and will be mailed to shareholders in the coming weeks.

"For as long as I can remember, veteran businessmen and investors - I among them - have been warning about the dangers of irrational stock speculation and hammering away at the theme that stock certificates are deeds of ownership and not betting slips. The professional investor has no choice but to sit by quietly while the mob has its day, until the enthusiasm or panic of the speculators and non-professionals has been spent. He is not impatient, nor is he even in a very great hurry, for he is an investor, not a gambler or a speculator. There are no safeguards that can protect the emotional investor from himself."

- J. Paul Getty

One of the important elements in the success of great investors like Warren Buffett is the ability to maintain what J.K. Galbraith once called a "durable sense of doom" during periods of market overvaluation. This is not easy, because investors often confuse temporary stock returns with permanent ones. They accept excessive risk in overvalued markets because they fear "leaving money on the table."

Historically speaking, advances that emerge from low valuations have typically been "permanent" in the sense that they are not erased by later market declines. In contrast, market returns from price/peak-earnings ratios over 18-19 have regularly been given back, often painfully. Though I certainly wouldn't advise it as a strategy, investors would have historically outperformed the S&P 500 with much less risk than a buy-and-hold simply by selling stocks when the S&P reached 19 times earnings and staying in T-bills until the P/E reverted to 15, even if it took years to do so. In effect, market gains (over and above T-bill yields) from P/E multiples over 18-19 have historically been purely temporary.

Investors are often warned that missing a modest number of the strongest days or weeks in the market would have resulted in dismal long-term performance. Aside from the absurdly low probability of missing those specific periods while holding stocks in all the others, what's not recognized is that those strong periods don't occur randomly. If you look at historical market data, over two-thirds of the best 30 weeks, for example, have occurred in periods when market valuations were below their historical medians.

Importantly, the converse is not true. The majority of bad weeks don't concentrate in overvalued periods. Instead, they concentrate in periods when the quality of market action has already deteriorated on the basis of yield trends, price action and so forth.

Notice the implication: periods of favorable valuation and favorable market action are loaded with "top" weeks, and very few "worst" ones. Periods of unfavorable valuation and favorable market action lack many "top" weeks, but they aren't typically periods when the market falls apart. Periods of unfavorable valuation and unfavorable market action are the worst of both worlds, lacking many "top" weeks but being loaded with "worst" weeks. Finally, periods of favorable valuation and unfavorable market action provide acceptable average returns, but are roller-coasters, including a significant share of both "top" weeks and "worst" weeks.

"Characteristic" performance

In any event, it is clear that taking a defensive investment position during periods of overvaluations has never harmed long-term returns. I emphasized issues like valuations and peak-to-peak returns in the semi-annual report, because the period of unusual overvaluation since 2000 might otherwise leave shareholders with an inaccurate understanding of "characteristic" performance for the Strategic Growth Fund. The Fund achieved an annualized total return of 16.07% annually from its inception on July 24, 2000 through December 31 2004, versus a -2.65 annual total return for the S&P 500 during that period. Though these overall returns have been consistent with my expectations, the distribution of returns between falling markets and rising markets differs from what I would expect under normal valuation conditions.

Specifically, from the Strategic Growth Fund's inception in July 2000 through February 2003, the Fund achieved an annualized total return of 16.67%. During the subsequent market advance from March 2003 through December 2004, the Fund achieved an annualized total return of 15.21%. Because of the unusual profile of valuations over the past few years, the Fund's returns were higher during the 2000-2003 bear market than I would expect during typical bear markets. In a typical bear market, valuations decline to historical norms or below, which would lead us to gradually increase our investment exposure during the late part of the decline. In contrast, the 2000-2003 decline never took the market to sufficiently low valuations to justify a value-driven investment exposure, so our returns were somewhat higher by virtue of the fact that we remained fully hedged throughout the selloff. (The Fund has also benefited from stock selection performance in excess of the S&P 500 and Russell 2000 indices since inception).

In contrast, Fund returns during the advance that began in 2003 have been as intended, given the level of valuations at which the advance began, but have been lower than I would expect during typical bull markets. A typical bull market begins at valuations that justify a fully invested, and often somewhat leveraged investment position. In contrast, the recent "bull market" (probably better viewed as an upward correction in an ongoing secular bear market) started at valuations too rich to justify an aggressive investment position. So while overall returns have reasonably matched my expectations, I expect that the distribution of performance across rising and falling markets will probably be different in a normally valued market than what we've experienced during the recent period of rich valuations. The Fund can be expected to be more defensive in richly valued markets than otherwise, and the past few years reflect that reality.

As always, past performance does not ensure future returns, market behavior may differ from historical market behavior that has emerged under similar conditions, and there is no assurance that the Hussman Funds will achieve their investment objectives. An investor's shares, when redeemed, may be worth more or less than their original cost. Investors should consider the investment objectives, risks, and charges and expenses of the Funds carefully before investing. For this and other information, please obtain a Prospectus and read it carefully. Please see The Funds page for Prospectuses, reports, and further performance information.

Valuations from a historical perspective

Our analysis of valuation considers not only earnings, but free cash flows, dividends, book values, revenues, profit margins, interest rates, inflation, risk premiums and other factors. Still, as a practical matter, the price/peak-earnings approach has proven the most popular and understandable one in explaining my view of market valuations (the market appears even more overvalued on the basis of other measures, so the conclusions below follow without loss of generality).

Let's look at the data. First, note that measured from peak-to-peak, S&P 500 earnings really have grown no faster than 6% annually over time. This is something to keep in mind when you hear that "earnings are growing at 18% annually," etc. Those figures represent trough-to-peak recoveries from depressed levels, not sustainable earnings trends that are appropriate for valuing stocks over the long-term.

Next, let's look at P/E multiples.

I frequently present calculations of probable long-term stock returns by making a few assumptions about long-term peak-to-peak growth in earnings and various future P/E multiples. I generally base these P/E ratios on peak-earnings. The price/peak-earnings multiple is the ratio of the S&P 500 to the highest level of earnings attained to date, even if current earnings on the index have declined below that peak. I constructed this in order to filter out the uninformative spike in the P/E ratio that occurs when earnings plunge during recessions. The price/peak earnings ratio is equal to the raw P/E when earnings are at a new high, as they are today, and is otherwise lower than the raw P/E. This makes it a conservative measure of market valuation, so very high levels properly merit concern.

Based on newly released quarterly earnings figures, the S&P 500's price/peak earnings ratio is nearly 20. Consider that level from a long-term perspective. Except for the reckless and unsustainable valuations at the 2000 peak, the current multiple has been observed only at the 1929, 1972 and 1987 market extremes and at the late 1960's market extremes. While the market plunges beginning in 1929, 1972 and 1987 are well-known, those mid-1960's valuations deserve some context as well. The Dow first approached 1000 in late 1965. It reached a durable low in 1982, seventeen years later, at 777 - a level it first achieved in January 1964. Investors forget these things if they ignore the data long enough.

Pulling all of this together, we can look at the range of 10-year returns that could have been projected at any point in time, based on various assumptions about the ending P/E ratio.

Using price/peak earnings multiples and assuming peak-to-peak earnings growth of 6% annually, the projected annual total-return on the S&P over T years is:

Long term total return = (1+g)(future PE / current PE)^(1/T) - 1

+ dividend yield(current PE / future PE + 1) / 2

The first term is just the annualized capital gain, while the second term reasonably approximates the average dividend yield over the holding period.

For example, start from the current price/peak earnings multiple of 20 and a 1.7% dividend yield. A still-elevated future P/E of 18, achieved 5 years from now, with earnings growth of 6%, implies a total return on the S&P 500 of approximately (1.06)(18/20)^(1/5)-1+.017(20/18+1)/2 = 5.58% annually over the coming 5 years.

In our last chart, we'll use four terminal P/E ratios. The first is the very optimistic assumption that in the decade following each starting point, the price/peak earnings multiple will move to a level of 20 (the same level seen in 1929 and other major extremes). The "normal cases" are future price/peak earnings multiples of 14 (the historical average) and 11 (the historical median). Finally, we calculate 10-year projected returns based on the pessimistic assumption that the future price/peak earnings multiple will be just 7 (matching major secular lows like 1982).

In the chart below, the thin lines are various 10-year total return projections, given the market's starting valuation at each date. The heavy blue line is the actual, realized 10-year return on the S&P 500 (note that it ends 10 years ago because that's the last point for which data exists).

Stare at this chart. Notice first that the range of outcomes contains the actual record of stock returns, and that outliers are relatively short-lived. For example, during the 10 years beginning in 1964 (when the price/peak earnings ratio approached 20 for the first time since 1929), the S&P 500 achieved a total return of close to zero, including dividends. That was a bit worse than even the estimate based on a terminal P/E of 7, because the brutal 1974 bottom formed a sharp but temporary "V." In contrast, in the 10 years beginning in 1990 (when the price/peak-earnings ratio was close to 11), the S&P 500 achieved a total return of fully 20% annually. This substantially exceeded the 10-year return of about 14% which would have been achieved had the 2000 bull market peak been held to a P/E of 20 (the market's actual price/peak-earnings ratio moved over 32 during the bubble). Of course, this outlier was also corrected, as the S&P 500 remains below its level of 6 years ago.

At the market's actual 2000 peak, valuations were so high that even a future price/peak earnings ratio of 20 could have been expected to result in a nearly zero annualized returns over the following 10 years. Not surprisingly, in the 5 years since the 2000 market peak, the S&P 500 has actually produced a total return of about -2% annually. The likelihood is that the coming 5 years will not be substantially better.

Presently, the likely range of S&P 500 annual total returns for the coming decade is in the 2-3% range based on average and median scenarios, with outside possibilities as low as -3% in the very bearish case and still less than 8% in the very bullish case.

Valuations matter. But again, investors forget these things if they ignore the data long enough.

Market Climate

As of last week, the Market Climate for stocks remained characterized by unusually unfavorable valuations and modestly favorable but generally deteriorating market action. We still don't have enough evidence to warrant a fully defensive stance against market risk, though there is an increasingly tenuous quality to market action which I am watching closely. For now, about 70% of the stock portfolio of the Strategic Growth Fund is hedged against the impact of market fluctuations, with the remaining 30% hedged with put options only. While I consider this stance to be modestly constructive, I am clearly concerned about both valuations and the increasingly tenuous quality of market action.

In bonds, the Market Climate remained characterized by unfavorable valuations and modestly unfavorable market action. The recent increase in nominal yields has not been sufficient to improve the relatively poor tradeoff between expected return and duration risk. While I can envision some circumstances in which we would be willing to accept a higher duration investment position even at present yields, those circumstances would require a general deterioration in credit conditions. That sort of deterioration would help to pull down monetary velocity (that is, individuals would be likely to demand more cold, hard cash rather than bank deposits or securities), and that in turn would short-circuit inflation pressures that are becoming evident here. For now, the Strategic Total Return Fund continues to carry a limited duration of about 2 years (meaning that a 100 basis point move in interest rates would be expected to impact the Fund by about 2% on the basis of bond price fluctuations), mostly in Treasury Inflation Protected Securities. The Market Climate in precious metals remains generally favorable on our measures as well.

---

The foregoing comments represent the general investment analysis and economic views of the Advisor, and are provided solely for the purpose of information, instruction and discourse.

Prospectuses for the Hussman Strategic Growth Fund, the Hussman Strategic Total Return Fund, the Hussman Strategic International Fund, and the Hussman Strategic Dividend Value Fund, as well as Fund reports and other information, are available by clicking "The Funds" menu button from any page of this website.

Estimates of prospective return and risk for equities, bonds, and other financial markets are forward-looking statements based the analysis and reasonable beliefs of Hussman Strategic Advisors. They are not a guarantee of future performance, and are not indicative of the prospective returns of any of the Hussman Funds. Actual returns may differ substantially from the estimates provided. Estimates of prospective long-term returns for the S&P 500 reflect our standard valuation methodology, focusing on the relationship between current market prices and earnings, dividends and other fundamentals, adjusted for variability over the economic cycle (see for example Investment, Speculation, Valuation, and Tinker Bell, The Likely Range of Market Returns in the Coming Decade and Valuing the S&P 500 Using Forward Operating Earnings ).