the time. Silver, then fetching as high as $49.50 a troy ounce, was $6.24 in late 1997, while gold was down from $875 an ounce to around $290. In the same period, the Dow Jones Industrial Average has risen from 850 to 7900.

A similar overreaction occurred in the art market in the 1980s. A prosperous economy and a good stock market sent art on a wild joyride. Many contemporary paintings (art created after the Second World War) went up as much as 20- and 30-fold, while the stately blue chips (Impressionist and Modem) of this market appreciated a more demure ten times. With the major increase in wealth in Japan, the Middle East, the U.S., and elsewhere, thousands of new investors tumed to art. There was a shortage of good paintings relative to demand, said the experts. After all, you could never create another Impressionist painting. The "shortage of good stock" thesis, as we just saw, was one of the most widely held investor beliefs in the years prior to 1929.

The auction houses never had it better. Bidding wars took paintings up to many times expert appraisal values. The atmosphere became carnival at staid Sothebys and Christies, where dignified auctioneers with Oxford accents presided in tails. The thronged galleries applauded each sale that went overthe appraisal price, which in itself had soared outside of the range of reality. In some auctions 70% of the paintings sold above the estimates. Van Goghs Portrait of Dr. Cachet was auctioned off for $82,500,000, while Renoirs Au Moulin de la Galette went for $78,100,000.

But as always, overreaction corrects. By late 1990, a recession, a falling stock market, and serious financial problems in Japan, where buying had been exceptionally heavy, particularly for Impressionist works, killed art prices. Impressionists dropped 50%, while the Yuppie aficionados who bought contemporary paintings for six figures from the recent "hot contemporary artists" now saw their prized conversation pieces drop to virtually nothing.

How quickly the tide changes. By 1990 and 1991 the atmosphere in the auction houses was sheer gloom. Painting after painting "passed" (was not sold), even though the appraised prices had been reduced substantially. I was able to buy a good Picasso in 1991 for one-third of what it had sold for only two years earlier, during a forced liquidation by a Swedish bank. Although the overall art market stabihzed by mid-1997, most prices are still well below their highs. The pattem, whether it is overenthusiasm or overpessimism, is ever recurring.

We have galloped briefly through some of the major overreactions of recent years. Dozens of others have similar radar plots. As a contrarian you might ask, "besides the contrarian strategies outlined, are there any

Junk Bonds: A Money-Making Overreaction

The most consistent evidence supporting another workable overreaction sti-ategy is buying high-yield bonds, as their proponents would call them-junk as they are known in the ti-ade. Junk bonds are issues rated bb or lower by the credit agencies.

I can imagine some snickers, because the common perception is that these bonds are a passel of bankruptcy candidates steering toward the corporate scrap heap. But hang on. Studies stretching from 1900 to the 1980s show the performance of this group to be nothing short of remarkable.

The Hickman report, the grandfather of studies in this area, analyzed the returns on public and private corporate debt between 1900 and 1945." The result: lower-grade bonds yielded 50% more on average than their higher-grade 1 8, after adjusting for defaults (the failure to make timely payment of interest or principal).

More interesting, perhaps, was the default rate-only 1.7% annually through a span that included the Great Depression. Too, defaults were not substantially larger in the lower- than the higher-quality ratings.

T. R. Atkinson, updating the Hickman study from 1945 to 1965, discovered that the default rate in that period was even lower, averaging only one in a thousand issues annually. A later study extending the results to the end of 1981 found the annual default rate to be between 1.5 and 1.9 issues per thousand with a yield premium of 4% or more above higher-grade bonds. In other words, if you held a diversified portfolio of junk bonds, the added income was 20 times more than the average amount lost in a default. Not a bad risk/reward ratio in anyones book.

Perhaps even more surprising was that default was not the nightmare everyone thought. Quite the contrary, the greatest gains were made on the bonds purchased at or near the date of default.

Although at first glance the findings are puzzling, they mesh perfecdy with the reasons that contrarian stocks consistendy outperform the market. Investors shun both high-income bonds and the out-of-favor stocks because their prospects seem less clear than for favored issues. To buy these risky critters they demand a large increase in retum. After all, who wants a Dell Computer or a Wool worth bond when they can have the unquestioned safety of a Warner Communications or a Microsoft?

Other types of predictable overreactions that can malce me a winner?" Fortunately, the answer is a definite yes.

Even worse is a bond that defauhs. "If it does," the consensus goes, "forget price-just sell." With large quantities suddenly dropping on the market and few buyers, prices become cheap, cheap. Benjamin Graham wrote about the pattern of institutions dumping such bonds more than 50 years ago, and he consistently made good money buying their mistakes. Not much has changed since then.

But if you are thinking of playing against knee-jerk thinking, you should be familiar with some of the background. The studies that junk bonds provided outstanding rate of return were carried out primarily before the 1980s. Lower rated credits could be more easily analyzed to assess their potential. Indeed Mike Milken in his earlier days made enormous money for his clients, and to a lesser extent for himself, carrying out precisely this analysis. In "artistic" terms Id refer to this period as "early Milken."

Later, after the junk bond market grew exponentially and investors clamored for higher-yielding issues, Drexel Burnham and scores of other supposedly respectable firms filled the need with bonds that promised high yields, but with such poor credit quality, that large numbers stumbled or died near the starting gate. I would call the period, which ran through the last halfofthe 1980s, "late Milken." This crop of junk bonds had little in common with its earlier counterparts, and resulted in large losses for many buyers, as well as bringing down a number of major S&Ls who chased the impossibly high yields. Quality has improved since then, but you have to know what youre doing.

First, diversify widely to protect yourself from the odd bad egg that can creep into your portfolio. Fortunately, large numbers of such bonds exist. (According to Securities Data Company, $178.45 billion worth of junk bonds were issued during the five years ending 1996.)

Second, these bonds arent for everyone. You have to understand the ins and outs of financial statement analysis, have detailed knowledge of the markets, and have nerves of cast iron. You should also own a dozen or more for adequate diversification.

An easier way for most investors to play this opportunity is to buy a junk bond mutual fund with a good 5- and 10-year record. Table 11-1 offers a number of such funds to consider.

If you can keep your own knees from jerking, you will find a very rewarding overreaction.