For a moment, making a profit was less important than making a statement. Lots of Americans did something that cost them money, at least in the short run. Yet they felt good about it — a powerful example of how psychology affects the market.

Economists used to think investors were always coldly rational. Now they know that's wrong. And research is shedding light on how strangely markets can behave. In what follows, I'll explain several fascinating studies that turn conventional ideas about short selling upside down. I'll also explain related discoveries about stocks that are good buys — in particular, companies that give investors a clear view of their prospects.

First, the news about short selling — which is a bit like sex, the subject of more talk than action. Only a handful of all shares (1% or 2% of the total) are typically in short positions, which means investors have sold borrowed shares in the expectation that they can buy them later at a lower price. One popular view is that big short positions are bullish, on the theory that future buying, or "covering," by shorts will boost prices. The evidence is just the opposite: Companies with heavy negative bets by short sellers really do underperform.

So why is shorting so unfashionable? For starters, it's risky. Buy something and you're never out more than you paid. Sell a stock you don't own and there's no limit to potential losses. Also, stocks gradually rise over time. So shorts ought to be long-term losers. If you doubt that, name a short-sale equivalent to Warren Buffett or Peter Lynch. Hedge funds do lots of short selling, but it doesn't seem to make them stellar performers.

Negative bets also require a capital cushion — and patience. A regulation called the uptick rule allows short sales only when the previous trade has moved a stock's price higher. When you enter a short trade, your broker must borrow shares. It can take time to find that stock (which must be in someone's margin account), and you might pay a fee for the loan. If a stock you've shorted pays a dividend, you owe that amount to the real owner.

Seeing Clearly Now

To these highly transparent companies go the rewards — or so new research would suggest.

COMPANY	PRICE*	52-WK HI-LO	PEG RATIO**	ANALYST RATING***	MARKET VALUE ($MIL)
CACI International (CACI)	$50.56	$52-19	1.34	1.33	571
Coach ( COH)	23.00	42-17	0.64	1.44	1001
Medicis Pharm. ( MRX)	42.00	74-36	1.02	1.38	1250
O'Reilly Automotive ( ORLY)	24.99	35-14	1.07	1.40	1304
RehabCare Group ( RHB)	37.10	51-32	0.93	1.47	640
Respironics ( RESP)	30.44	36-16	1.11	1.44	922
Ruby Tuesday ( RI)	15.35	20-11	0.71	1.29	984
Varian Medical ( VAR)	59.32	77-43	1.47	1.43	1997
Median of 600 comparable companies ( N/A)	N/A	N/A	1.07	2.03	N/A

* Prices as of 9/21/01.
** Price/earnings ratio for the current fiscal year divided by estimated growth rate.
*** Consensus on a five-point scale, with 1=Strong Buy and 5=Strong Sell.
Data: Research Wizard 4.0, Zacks Investment Research

Academics didn't pay much attention to any of this until recently. But the tech-stock craze made researchers wonder why stock prices soared to such dizzying heights — and reluctance to sell short may provide a clue. In a 1977 paper, an outside-the-box academic named Edward Miller [Dr. Edward Miller of UNO !] made an observation that now seems obvious: When there are restraints on selling short, prices reflect the view of optimists. Pessimists will sit on the sidelines.

Carry this logic further and stocks will keep rising until even the most giddy optimist thinks they're expensive. Eventually, there are no more buyers, and prices tumble — a bubble, followed by a crash. That's extreme, I know. But researchers now wonder if markets might be less volatile if short selling were more common. They're also testing Miller's theory, with impressive results.

The stock market isn't a physics lab, so the inquiry tends to be indirect. But analyst estimates provide a neat research tool. Example: If everyone thinks Johnson & Johnson (JNJ) will earn $2.20 a share, the price should reflect that. But if there's disagreement about prospects for AT&T (T) (as there is), its shares are more likely to be bid up by optimists. Skeptics, because of psychological and regulatory constraints, are more likely to twiddle their thumbs than sell short.

When there are restraints on selling short, prices reflect the view of optimists. Short sellers keep markets honest.

Getting a fix on this is tricky. Several recent studies track the relationship between dispersion of analyst estimates (a statistical measure of disagreement) and subsequent price movement. For example, Anna Scherbina at Northwestern University looked at companies with two or more estimates between 1983 and 1999 — tens of thousands of observations. She divided stocks into portfolios based on dispersion.

Sure enough, companies on which analysts tended to agree beat stocks on which disagreement was highest (and optimism presumably ran unchecked) by about 10% annually. That's a powerful number, confirmed by other studies. The performance gap seems to be wider — maybe 18% annually — for small-cap stocks, where selling short can be particularly difficult.

Other researchers approached the question differently. Why do stocks perform better when analysts agree about earnings? James Ang at Florida State University and Stephen Ciccone at the University of New Hampshire believe the market rewards investors for owning "transparent" companies, which provide a clear picture of the future. It penalizes owners of "opaque" companies, where signals are mixed.

Look at it this way: When management thinks the future is bright, there are plenty of incentives to be clear and specific — which leads to unanimity among analysts. That unanimity may be one reason (if not the only reason) research indicates that shares in these companies are good buys. But when the conditions are cloudy, fog seeps into corporate communications. That may keep bad news from spreading. But it also causes a divergence of opinion, which, possibly helped by the absence of short selling, gives optimists the upper hand.

I see three lessons here for individuals. First, don't assume short sellers are bad guys. By keeping prices in line, they may save you money. Second, avoid controversial stocks. On average, they don't provide enough extra return to offset the extra risk. Finally, what to buy? Well, in this month's table I zeroed in on companies that rank near the top of my homemade Transparency Index.

Like the economists, I used earnings estimates. I looked at 600 nonfinancial companies with market values between $500 million and $2 billion that are covered by two or more analysts. I ranked this universe to find stocks with the least dispersion (greatest agreement) in earnings estimates for this year and next.

I also wanted situations where analysts got it right. So borrowing a tool from Ang and Ciccone, I compared last year's actual results with the final consensus — the closer the better. About 80 stocks ranked in the top 20% in each of my three yardsticks. I narrowed the field by focusing on companies with the best current analyst ratings.

These companies aren't stars based on my typical value benchmarks, but it's still easy to be impressed. They're growing faster and are more profitable than comparable firms. More important, prices (based on price/earnings-growth ratios) look modest. Most are in businesses that seem recession resistant.

CACI (CACI), a systems integrator, sells half its software and services to the Department of Defense, a customer with an expanding bankroll. O'Reilly Automotives's (ORLY) stores (6% net margins, 14% same-store growth) tend to do well when the economy slows down, and Ruby Tuesday's (RI) restaurants depend more on folks down the street than on global tourism. Both companies are adding new outlets rapidly. And Coach (COH), where management just announced a stock buyback, expects sales to grow by 14% in the current quarter.