JDSU--When to sell your shares tomorrow:
Stolen from the "Wall Street Stand"
From: Sharck Post:
Well its about time...
In midtown Manhattan there meets in secret a small group of men known as the Standard & Poor's Index Committee. As far as I know, no member of this committee has managed money professionally. They're nothing but a group of well-meaning bureaucrats serving a life sentence in a division of McGraw-Hill, with no possibility of parole. Yet just a word from this committee can mobilize a trillion dollars worldwide to buy any stock they name without the slightest question -- any stock they choose to name as the next member of the S&P 500 Index.
The latest chosen one was Broadcom (BRCM:Nasdaq - news), which was trading at 147 1/2 on June 19, when S&P made the announcement. By June 30, when it was actually added to the S&P 500, it had run up 48.4% to 218 15/16, while the S&P and Nasdaq both fell during the same period.
Lucky Broadcom. And lucky traders. No one's handing out free lunches on Wall Street, but this one's close: When stocks are added to the S&P 500, they almost always move sharply higher as the date of actual addition to the index approaches. It has been happening for more than a dozen years, and it's all because of the seemingly crazy way indexers invest and trade.
Here's why it works: Indexers have no choice but to do with real money exactly what the index does with pretend money, and at exactly the same time. That applies to Vanguard, Barclays Global Investors (where I used to run portfolio management and trading), State Street, and all the other money managers and pension-plan sponsors who run index funds. They're all good little soldiers: when the order is given, they carry it out without question.
Let's say the S&P Index Committee orders that JDS Uniphase (JDSU:Nasdaq - news) is joining the index at the close on July 31. That means that a trillion dollars worth of index funds are buying JDS Uniphase at the close on July 31, no questions asked -- and price is no object.
How much JDS Uniphase will the good little soldiers have to buy? That, too, is just a matter of following S&P's orders. With a market cap of about $60 billion, JDS Uniphase represents about 0.47% of the total capitalization of the S&P 500. So any S&P 500 index fund will have to allocate 0.47% of its value to JDS Uniphase. For a $109 billion S&P fund such as Vanguard's, that means buying $512 million in JDS Uniphase -- at $100, that's 5.1 million shares, or about a quarter of a day's normal trading volume.
That's just one index fund. No one really knows the value of all the S&P index funds in the world. But it could easily be $1 trillion -- even that astronomical number is well under 10% of the total value of the S&P 500. If $1 trillion is the right number, that means that there will be $4.7 billion, or 47 million shares of Uniphase to buy -- about 4 days' normal trading volume.
Here's how the good little soldiers will carry out their orders.
First, they'll figure out how many shares of JDS Uniphase are already held in their "extended market" index funds -- funds that track the Wilshire 4500 Index, designed to hold any stock that the S&P 500 doesn't hold. Any stock that becomes a member of the S&P 500 is, by definition, ejected from any "extended market" fund. That means it can be crossed off exchange between the funds at the closing price on July 31. Based on my experience, I'd estimate that this method satisfies about one quarter of the total demand for stocks added to the S&P 500 index.
That leaves some 35 million shares that will have to be bought in the open market. The good little soldiers don't care -- really can't care -- what price they pay, as long as it's market-on-close on the date of the addition -- exactly the same price that will be used by S&P when JDS Uniphase enters the index, however high that price may be! That means that other traders in the market can buy JDS Uniphase with nearly complete confidence that there will be someone to sell it to no matter how high the price goes.
Let's say Broker X, one of a several bulge-bracket firms that handle all the trades for the good little soldiers, has a ticket to buy 20 million shares of JDS Uniphase, market-on-close. About an hour before the close Broker X will start buying for his firm's proprietary account. His goal is to buy 20 million shares in a single hour, more than the average volume for an entire day. It's called "ramping the close." The idea is to make sure that the closing price -- where Broker X will sell his shares to the good little soldier -- is higher than the average price he paid in the hour before the close.
It almost always works. But it's not as easy as I make it sound. For one thing, the good little soldiers may be crazy, but they're not stupid -- they know that their buy order is a one-way ticket to free money for Broker X, so they won't give it to him for nothing. The typical deal is for the broker and the index fund to split the profits 50/50. Even then it is such a good deal for the broker, he usually commits to eating any possible losses just in case it doesn't work out according to the playbook.
Another risk is that other traders will try to horn in on the action. Just as Broker X knows that his index-fund client will be buying market-on-close, other traders guess that Broker X will be buying -- so they get in a little earlier, say two hours before the close. Then when Broker X starts to buy in the last hour, the other traders start to sell. Which makes it tougher for Broker X to "ramp the close," so he starts earlier. Or maybe he gets scared and waits until the last 15 minutes, and jams all 20 million shares through then. It's like a souped-up version of musical chairs, where the game is not only to get into a chair first when the music stops, but out of the chair first, too, when the music starts again.
The game doesn't end at the close. It starts up again the next morning, when all of a sudden no one wants to buy JDS Uniphase quite so urgently. It's not surprising to see a stock added to the S&P get run up right at the close, and then collapse the very next morning.
Depending on your tastes, there are dozens of strategies for playing the S&P game. The easiest is just to coattail it with Broker X at the close -- no guarantees, but it usually works. If you're more the cerebral type, you can try to outguess the 'crats on the S&P Committee: Which company will be the next lucky winner?
That's the toughest game of all. The S&P Committee prizes its independence above all else, and doesn't like to be outguessed. That's why my example, JDS Uniphase, is such a longshot: It has been rumored as a lady-in-waiting for so long that S&P keeps not picking it just to make the rumor-mongers look bad.
I am long JDSU now as just like I called BRCM and you should be too.
Trade smart,
Sharck |
