Looking Over The Valley

In the wake of the stock market’s agonizing decline over the last three months, an update of the views expressed in our July 21, 2002 paper (See: Manic Depression In The Stock Market) seems to be in order. This is especially so since, after holding exceptionally well through the bear market, from its start in March 2000 through the end of June 2002, our portfolios have been caught in the whirlpool of the past three months.

On a total return basis (including the re-investment of dividends), the S&P 500 index had lost about 30% from the end of 1999 to mid-2002. In contrast, the Tocqueville Fund’s net asset value (adjusted for distributions) was essentially unchanged over the same period i. Individual accounts actually did somewhat better, thanks to holdings of smaller-capitalization stocks and international investments, which kept rising until a few months ago.

Since the end of June, however, our performance has actually been marginally worse than the S&P 500’s. While a more than 20% decline in just over three months still leaves us ahead of the index for the year-to-date by several percentage points, this is no consolation for, as we are prone to say, we do not strive for relative performance against any particular index but rather for absolute, long-term protection and appreciation of capital.

Yet, we have gone back over the performance of individual stocks over the last three months and the decline was literally across the board. There is little we could have done, while staying true to our philosophy and discipline (e.g. without “trading” the market), to isolate our portfolios from the recent downward spiral. On the more optimistic side of things, this has been a fairly typical phenomenon toward the end of extended bear markets. Investors sell in a panic, fund managers also must sell to meet redemption needs, and they all try to sell stocks on which they do not need to realize huge losses. Unfortunately, those often are the stocks that we own.

In spite of this, as we look at the investment environment today, we find that little has actually changed from the way we saw things shaping up in July:

· The U.S. economy does not look as bad as the media would make it. In fact, we see a number of reasons to expect a re-acceleration of both GDP growth and corporate profits in the months ahead.

· The global environment has shaped up broadly as expected, with Asia behaving somewhat better and Europe somewhat more disappointingly than we had anticipated. But even there, as well as in Japan, some planned relaxation of fiscal policies could bring improvement in the months ahead.

· After a 50% decline, now surpassing even the watershed bear market of 1973-1974, U.S. stocks seem to have reached a level from which a significant, cyclical bounce should take place. This holds true even if, as is our assumption, we have started on a long-term downtrend in Price/Earnings ratios.

The risks inherent in the current global environment have now been well publicized (deflationary spiral, financial failures, war, etc.) To some degree, this reflects an adoption by the broad public of the fears we had expressed as early as January 2000 (See our report: First The Silver Lining, Now The Cloud) Today, however, we believe that many of these risks have been incorporated in stock prices. What has received less attention is the potential rewards for investors from current levels.

We have no doubt that stock market returns over the next ten years will be much lower than the abnormally high ones reaped in the 1980s and 1990s. Over more than a century, the stock market has only returned an average of 9%-10% per annum. The abnormally high returns of the last twenty years all but guarantee that those of the next decade or two will be sub-par.

This is confirmed by a historical review of market returns and Price/Earnings ratios. [A stock’s price is meaningless in itself and only useful when compared to some measure of value of the underlying company. One such measure is the ratio of a stock’s price to its company’s earnings per share – the Price/earnings (P/E) ratio.] Various studies have documented that stock market returns over ten-year periods are inversely correlated to the level of P/E ratios at the beginning of each period. Since 1926, when a ten-year period started with a P/E ratio of 16x-18x (which is roughly where we are now, depending where one chooses to place the S&P 500’s current earnings power), stock market returns over the following ten years typically have averaged only 5%-7% per annum.

Source: Global Financial Data. www.globalfindata.com

As the chart above illustrates, the last secular bull market started in the early 1980s with P/E ratios of about 7x-8x and ended in 1999 with P/E ratios well in excess of 30x. This allowed stock prices to grow at a rate well in excess of corporate profits for the period as a whole. However, the rise in P/E ratios was accompanied and facilitated by secular declines in inflation and bond interest rates, which are quite likely to be reversed over the coming decade or two.

With P/E ratios now likely to be in a secular downtrend, average stock returns should fall below the growth of corporate profits over the coming decade or longer, so that stock market returns averaging 6%-7% per annum do seem a more realistic expectation than ones of 13% or 15%.

Does this means that we should give up on the stock market? We don’t think so.

At the recent level of just over 800 for the S&P 500, consensus earnings estimates of $48 for 2002 put the index’s P/E at 16.7x. Even if one makes adjustments to bring S&P 500 earnings in line with the corporate profits component of the National Income Accounts, which many consider to be more conservative (perhaps excessively so, since it is based on profits reported to the IRS), the S&P 500 earnings would still be about $44. On such earnings, the P/E ratio for 2002 would be only about 18x. That is still high by historical standards, but a far cry from well over 30x three years ago. For next year, estimates of $54 or more point to a P/E ratio of less than 15x.

Note that it is not unrealistic to anticipate a significant bounce in corporate profits in 2003-2004. In fact, most forecasting models of corporate profits are pointing higher.

True, wage costs are creeping up. But in periods of recovery for economic activity, production increases faster than payrolls, which holds back labor costs per unit produced -- and thus boosts profit margins. Also, when inventories are being liquidated (which has been the pattern for most of the past year), industrial production rises more slowly than the final demand for products, thus artificially but temporarily depressing capacity utilization. Production costs per unit (which allocate depreciation and overhead to units produced) are therefore temporarily inflated. Later, recoveries in industrial production and capacity utilization lower unit costs and boost profit margins. This is what we expect to happen in coming months.

Finally, there has been an excellent correlation between corporate profits and producer prices (which move earlier than consumer prices and should soon start to creep up, as have some materials prices). Low investment in new plant and the closure of inefficient capacity over the last several years, particularly in old economy companies, have resulted in an environment where global growth has the potential to bring promptly potential supply and demand into balance. When this happens, old economy companies particularly will regain a measure of pricing power that will also help to boost margins.

But, in the next couple of years, we believe that it will principally be P/E ratios that will move stock prices. As we mentioned earlier, P/E ratios are anything but low by historical standards – especially when compared to previous, important market bottoms, when they often were in single digits. However, P/E ratios have never moved in a straight line – either up or down. For example, during the 1960s, after starting above 20x, P/E ratios fluctuated between 15x and 18x for a decade and it was only after a push to nearly 20x in 1971-72 that they fell sharply down to settle in a range of 8x-12x. By that time, however, both inflation and interest rates were already surging.

We believe that P/E ratios could easily bounce back toward 20x over the next couple of years – not even approaching the euphoric levels of 1999. By that time, analysts would be looking at 2004 earnings of $59 or more on the S&P 500. Thus, without stretching the imagination, we could look at gains in the order of 40%-50% for the S&P 500 over the next 18-24 months.

Besides fundamental determinants such as inflation and interest rates, P/E ratios are heavily influenced by investor psychology since they incorporate qualitative judgments and reflect expectations of companies’ future growth in sales and earnings. As a group, however, investors are manic-depressive and periods of depression usually give way to periods of rising optimism, if not euphoria. Since 1926, there have been only two periods when stocks declined for three calendar years in a row. One was the Great Depression (1929-1932), and the other was the period before America’s entry into WWII (1939-1941). In both these cases, it could be argued that things looked a lot worse and uncertain than today. Yet, even before these economic or political crises were over, the bear markets ended and were followed by 50%-plus gains in the following two years.

In 1973-74, as economies slowly slid into what was to be a deep, worldwide recession, the S&P 500 fell a total of 45% over two years – only to bounce back 69% over the two following years. The stock market bounce started before any recovery in the economy became visible.

The purpose of citing these three watershed episodes is to illustrate that the psychological influences on the stock market can cause significant cycles in P/E ratios (and therefore in stock prices) – whether the long-term trend is up or down.

Our current view is that we are around the bottom of such a cycle within a long-term declining trend in Price/Earnings ratios. But the potential gain is worth considering.

October 14, 2002