To all who believe that a heavy short position is, on its own, a reason to be bullish about a stock......another opinion.
And to Andy Chen who believes that an increasing short position LOWERS the float of that stock.......another opinion.
From Barrons today:
<<Short Shrift
What one firm gives to the idea of heavy shorting as a bullish
sign as Wall Street used to tell it, a way to make easy money was to buy a heavily shorted stock. The reasoning was simple. Short-sellers eventually have to cover their positions by purchasing shares, so the short interest is just a measure of future buying pressure that will soon drive up the price. This argument has proved seductive for much of the bull market, in which many short-sellers have been battered and some even have been driven out of business. But in truth,short-sellers
often are right, and their thumbs-down vote against a company can amount to a signal worth heeding.
In fact, after reviewing New York Stock Exchange short-interest data from 1977 through 1990, Professors Paul Asquith of MIT's Sloan School of Management and Lisa K. Muelbrook of Harvard Business School drew a simple conclusion about what to do with a heavily shorted stock: Sell it.
Wrote the scholars: "If an investor already owns a stock that develops high sustained short interest, the clear and strong advice is to sell the stock immediately." Not tomorrow, but now. This tip was based on the simple finding that such stocks seriously underperform the market, even tending to decline in price. Their bottom-line conclusion: Short interest "does indeed convey negative information"
the market often ignores.
So maybe the short money is relatively smart money. And in this regard one should scrap the typical image of the short-seller as the curmudgeon who hates the world and spreads malicious rumors. Of last month's $136 billion worth of short interest, perhaps $3 billion came from full-time shorts. The rest was dispersed broadly among hedge funds and mutual funds that often look on the long side as well.
Enter Mike Long and Nat Guild, co-partners of a Charlotte, N.C.-based firm called Short Alert. If short interest does signal price weakness, they reasoned, then maybe they could provide a service by massaging the numbers issued every month by the New York and American Stock Exchanges and Nasdaq. And if Wall Street persisted in its delusion that heavy short interest is a bullish indicator, their analysis would have a special edge: The market wouldn't be discounting the information right away.
So Long and Guild repaired to their computer for a few months and emerged with a proprietary model keyed off short-sale data. One product of the model is a hit list they call the "Most Dangerous Stocks", so named because they counsel against holding any of them in your portfolio - except, of course, as short sales.
To sell short, an investor borrows shares owned by some other investor. His broker gets the shares from a margin account, from an institutional investor, or from another broker-dealer. (Some stocks are so heavily shorted that they're hard to borrow.)
Regardless of how the shares are obtained, they're effectively pulled out of someone else's holdings and then sold to a new buyer. Thus, 100 shares do the work of 200, and the short-seller artificially expands the float for as long as his position is maintained. Once he closes out his trade, the float is effectively reduced. He engages in what's called ``short-covering,'' a maneuver that involves buying an
equivalent number of shares and returning them to the investor from whom they were originally borrowed. He profits if the price of the initial sale exceeded that of the subsequent purchase.
Short-interest data provide a snapshot of all the as-yet-uncovered positions. But, as Long is the first to admit, the numbers aren't completely reliable, particularly when they show unusually large month-to-month fluctuations. "Incorrect data are not going to have the impact that prices, exchange-required filings, earnings reports and even volume would have," he admits. "So the time spent collecting it and making sure it's right is not exactly on the same level as the Manhattan Project." He and Guild expend a lot of effort on cleaning up the figures, even to the point of calling a company directly when a number looks particularly questionable.
Phone calls and related research also help them ignore companies whose short interest doesn't necessarily reflect a negative market view. For instance, it's common for traders to arbitrage between a short position in a company's stock and a long position in its convertible bonds. (On the other hand, Short Alert still believes Boston Chicken belongs on the "most dangerous" list, even though a lot of its shorts are convertible bond arbitrageurs.)
But Long asserts that one kind of technically induced shorting shouldn't be ignored in his analysis: short sales that result from trading put options. When the public buys puts, market-makers frequently take the other side - and then hedge their exposure by shorting the underlying stock. A potential distortion of the numbers? No, says Long, because puts are just another way of betting that a stock will decline.
The table shows the 15 stocks Short Alert now considers to be the most dangerous to own. How were they selected? While Long and Guild won't disclose their proprietary methodology, they offer clues. To begin with, in line with the system used by Asquith and Muelbrook, Short Alert calculates "percentages of supply." This is represented in two ways: short interest as a percentage of shares outstanding and then of the float - the number of shares actually available for trading by the public. Sometimes, the figures differ sharply.
For instance, shorts on shares outstanding for Yahoo! come to 8.5%, but on a float basis, the figure is a hefty 74.9%.
Short Alert also calculates the net new shorting that took place over a month, which is the difference between this month's short interest and the previous one's. This figure is expressed in dollars and taken as a percentage of market capitalization. As Long points out, the percentages of supply ratios are historical - you don't know
exactly when the positions were put on - while the net new shorting ratio is current.
Then the list is taken through 15 possible ``red flag'' tests, mostly involving a company's fundamentals. For instance, Cannondale gets a red flag because its sales growth has slowed. Long is circumspect about his track record, but he is willing to show how his most dangerous picks have done in relation to the S&P 500. Not surprisingly, since he started doing his analysis back in November 1995, the stocks chosen have underperformed the S&P by a wide margin. And what's even more impressive is that they have almost consistently declined in price, despite the bull market's surging tide. For instance, the 25 stocks chosen as most dangerous in mid-January 1996 fell almost 20%, on average, by the end of December.
So when it comes to viewing heavy short interest as a signal to purchase shares - buyer beware.>>
PS: CUBE did not make the "most dangerous" list. rj
|
