In the past year, the best way to play the three flagship stock indexes of Standard & Poor's was not to buy and hold them, as most finance professors and many advisers recommend. You were better off being the ultimate contrarian, purchasing stocks that impatient S&P index managers expelled from their lists for reasons other than mergers or corporate restructuring, and holding them for one to 11 months.

The difference between the returns of expelled versus added stocks in S&P's widely watched large-cap, mid-cap and small-cap indices from January through the first week of December was a stunning 45 percentage points, with expelled stocks gaining 42.3% and added stocks losing 2.4%.

These results did not accrue from just a couple of big outliers. Of the 48 stocks kicked out of the indexes this year for what S&P euphemistically calls "lack of representation" -- you and I would call it "we got sick of this stock because it's gone down so much" -- 28 subsequently went up from the date of expulsion through Dec. 7, and 20 fell. And you typically didn't have to wait long to capture the effect: Most gainers experienced the bulk of their move in one to two months.

Three typical examples: Paper products maker

Nashua

(NSH)

was booted from the S&P SmallCap 600 on Feb. 27 in favor of oil driller

Key Production

(KP)

. Nashua subsequently went up 53% over the next month, 89% over two months, and settled in with a 65% gain through Dec. 7. Meanwhile, Key fell 25% by Dec. 7.

Aluminum and lumber conglomerate

Maxxam

(MXM)

was booted from the S&P MidCap 400 on April 4 in favor of luxury-goods retailer

Coach

(COH)

. Maxxam subsequently went up 15% over the next month, 101% over two months, and settled in with a 42% gain through Dec. 7. Meanwhile, Coach gained 23% by Dec. 7.

Telecommunications provider

Global Crossing

(GX)

was booted from the

S&P 500

on Oct. 9 in favor of utility

TECO Energy

(TE)

. Global Crossing subsequently went up 176% over the next month and 239% over two months through Dec. 7. Meanwhile, TECO fell 1% by Dec. 7.

The notion that most companies humiliated by S&P would subsequently rise in value is counterintuitive, but it makes sense once you understand how much money is invested in funds that track these indexes, and how indexing works.

How S&P Selection Favors Growth

According to Elliott Shurgin, veteran member of S&P's index committee, about $1 trillion worldwide is invested in funds that directly track the results of the S&P 500 index alone. An additional $25 billion is in funds tracking the S&P 400. And $8 billion tracks the S&P 600. (The total market capitalization of each index is $10.4 trillion, $816 billion and $354 billion, respectively.)

The S&P committee has, in other words, come to manage what amounts to three of the largest funds in the world -- funds not just used in pension plans to allow investors to match the market, but which are also considered to be the market, and which are used as benchmarks for all other funds. Yet it has done so with remarkably little scrutiny of its stock-picking prowess or methods.

Few investors realize, for instance, that the process by which index stocks are added and subtracted virtually ensures that the index funds will favor growth or momentum stocks and shun value stocks. The reason is simple: Each of the three major indices is assembled to reflect stocks of all industrial sectors without regard to valuation in a range of market capitalization -- large, medium and small.

When a large-cap stock in the S&P 500 falls into such disfavor that it becomes a mid-cap or small-cap, the committee must either shift it to a smaller index or boot it clear out of its universe. Those stocks typically are replaced either by an S&P 400 constituent that has outgrown its own index or by an exciting large new company that's not yet in any index. Thus, in a very structural way, much-hated value stocks fall out of the indexes, and much-loved growth stocks are added.

Expanding the Anti-S&P Effect

It's been well documented that stocks the committee chooses to add to its big-cap index typically rise sharply in value from the day prior to the announcement to the close of the day the stocks join the index. But it's not well known that stocks subtracted from the index for reasons other than mergers or corporate spinoffs rise a lot more. And it's also not well known that the same effect occurs in the mid-cap and small-cap indexes.

The S&P kicked 19 mid-cap value stocks out of its 500 index in 2000 and replaced them with growth stocks at precisely the wrong time. From Jan. 1, 2000, to Feb. 12, 2001, stocks purged from the S&P 500 went on to gain an average of 44.5%, while stocks added to the list declined 15.5% on average. (The differential has persisted, with the deleted stocks of 2000 up about 26% through Nov. 29, versus a 35% decline in the '00 added stocks.)

In 2001, S&P seems to have cooled its ejection seat, jettisoning only eight stocks from its flagship index. But the gulf in returns has been greater this year. The eight stocks purged from the S&P 500 have risen an average of 70% from their last close in the index through Dec. 7, while added stocks have fallen an average of 1.9%. The most interesting two cases came in the fall when Global Crossing tripled soon after getting the heave-ho, and

Enron

(ENE)

doubled. Both were penny stocks at the time, but highly liquid and easily purchased.

Not much attention is paid to the S&P MidCap 400 index, but the same phenomenon transpired this year. The eight stocks expelled from the list have risen an average of 63% from their last close in the index through Dec. 7, while added stocks have fallen an average of 4.9%. Most often, the largest part of the move occurred in the first month to two months. Only two expellees failed to rise by at least 10% in the two-month period following their exit.

Action in the S&P SmallCap 600 was similar in the one- and two-month periods, although the effect faded substantially after that. The 32 stocks expelled from the list fell an average 6.3% from the close of their last day in the index until Nov. 28, vs. a decline of 0.3% for the added stocks. However, just one month after expulsion the stocks were up an average of 27.8%, and after two months they were up 24% on average.

Why does the phenomenon occur? It's largely because the extreme selling pressure by index funds on these hapless stocks -- which are already on the outs with investors -- leaves a vacuum. In most cases, only buyers are left.

Shurgin, who expressed surprise when he learned how well the ditched stocks had performed, admitted there was "a certain logic" to it. "If a company has lost so much market cap that it has to be removed from our index, then the actual removal could be seen as the classic, absolute last piece of horrible news," he said. "As long as the stock doesn't go to zero, you can see why it might rise from that selling climax."

The nice thing about this unusual effect is that you should be able to make money from it without having to predict anything. Just periodically check the Web page where S&P announces its index changes after hours, and note the date on which a stock will be removed from an index. Buy the stock at the close of its last day in the index, or soon thereafter. You should find that among the expellees, winners will continue to outpace losers.

At the time of publication, Jon Markman owned or controlled shares in none of the equities mentioned in this column.