Monday, May 12, 2008

Gold Timer's sentiment index is bullish

Barron's Online
Wednesday, May 7, 2008
0

HULBERT ON MARKETS

When Bad News Is Good News for Gold

By MARK HULBERT

I HAVE SOME GOOD NEWS for beleaguered gold investors: The editors of gold timing newsletters finally have thrown in the towel and given up hope that the bull market in gold will soon resume.

If you have a hard time understanding why that is good news, you're not familiar with contrarian analysis. According to contrarians, the market rarely accommodates the majority, especially at major market turning points.

That means that the rallies that have the most staying power tend to be those of which the majority is skeptical, while declines thrive on the hope that the decline will be only temporary. To put it in terms of phrases that most of you probably have heard before: Bull markets like to climb a wall of worry, while bear markets like to descend a slope of hope.

This psychological perspective helps investors to differentiate between a decline that is a mere correction from one that is the beginning of a major bear market.

From this perspective, things as recently as mid-April were not looking good for the gold market. During the second and third weeks of April, for example, a period in which gold bullion dropped some $25 per ounce, the editor of the average gold timing newsletter actually became markedly more bullish. This reaction far more closely fits the template of what precedes more serious declines than mere bull market corrections, and, as a result, contrarians concluded that a bottom was not yet at hand. The price of an ounce of gold quickly dropped another $50.

Today, however, the editors of gold timing newsletters are beginning to throw in the towel. This has dramatically changed the sentiment picture, to the point that contrarians are now willing to entertain the notion that a sustainable rally can now begin.

Consider the latest readings of the Hulbert Gold Newsletter Sentiment Index (HGNSI), which reflects the average recommended gold market exposure among a subset of short-term gold timing newsletters tracked by the Hulbert Financial Digest. As of the close of trading on Tuesday, May 6, the HGNSI stood at minus 10.7%. This negative level means that the editor of the average gold timing newsletter is net short the market, advising that 10.7% of the typical recommended gold portfolio is invested in a bet that the gold market will decline.

Since the beginning of 1985, some 23 years ago, the HGNSI has been this low or lower only about one-tenth of the time.

But that's not the only reason that contrarians are encouraged: The HGNSI's decline in recent weeks has been precipitous. In mid-April, for example, the HGNSI stood at plus 25.0%. So in only about three weeks' time, the HGNSI has declined by nearly 36 percentage points. This quick a drop suggests that many gold timers have thrown in the towel, which is a bullish sign according to contrarians.

The Hulbert Financial Digest has rigorously analyzed the HGNSI back to the 1980s, studying the correlations that exist between high and low sentiment levels and how gold bullion has performed over subsequent weeks and months. These correlations are statistically significant at the 95% confidence level that statisticians often use to assess whether patterns are genuine.

To illustrate, consider first the 10% of weeks since 1985 in which the HGNSI was as low as it is currently, or lower. (About 120 individual weeks are included in this decile.) Over the 30 days following each of these instances, gold bullion produced an average annualized return of 14.1%.
That's a lot better than the 4.4% average annualized produced by gold over the entire period since the beginning of 1985.

Now consider how gold bullion performed in the wake of sentiment readings at the opposite end of the spectrum -- when the typical timer was quite exuberant, in other words. On average following the 10% of weeks since 1985 in which the HGNSI was highest, gold bullion produced an annualized loss of 1.4%. That's markedly worse than average.

These results definitely point to a higher gold price over the next month. But note carefully that there is no guarantee: Statistical significance does not equal a guarantee. So one most definitely should not throw caution to the winds. Nevertheless, unlike the situation that prevailed as recently as mid-April, the odds are now looking good.



Mark Hulbert is founder of The Hulbert Financial Digest. He is a senior columnist for MarketWatch.

No comments: