Sell These Stocks

Here's a little tale of two giant companies going opposite ways yet bound for the same place: Mediocreville. Both are trying desperately to avoid that end, one by boundlessly buying new businesses and the other by ripping itself apart. These are choices that companies face all the time, by the way, and they are critical to stock-market performance. One company is AT&T;, which last week announced its third breakup in 17 years. The other is GE, which unveiled its umpteenth and largest acquisition--$45 billion for defense contractor Honeywell.

GE mediocre? The corporate icon headed by Jack Welch since 1981 has thrived by being focused. Be No. 1 or No. 2 in a business. Period. Welch has amassed operations in a dozen industries ranging from financial services to aerospace to media. With Welch steering, GE has increased its market value from $13 billion to $518 billion, becoming the most valuable company in the world.

Shareholders continue to be amply rewarded. GE is up modestly this year, while the Dow is down 8%. Call it the Welch premium. GE's remarkable ability to keep its stock rising means investors are willing to pay more for it. Currently, they pay $43 for every $1 of annual GE earnings. At Honeywell, investors before the announced takeover were paying a mere $21 for $1 of earnings.

Welch simply is without peer, and that's exactly why GE's stock is headed for ho-hum. He's retiring. With the Honeywell deal, Welch agreed to stick around eight months longer, until the end of next year. After that, though, he's teeing off at Augusta.

The new GE CEO will be well trained and whip smart. Be he won't be Welch. Few CEOs have had any luck running conglomerates in the past decade, and no one else running an industrial behemoth like GE will get the Welch premium. It will erode, and the stock will lose some magic. On top of that, Welch's successor faces the daunting challenge of converting Honeywell's slower-growing businesses into the kind that expand 20% a year, as GE does.

The issues at AT&T; are vastly different. CEO C. Michael Armstrong tried his hand at Welch-style growth, spending more than $100 billion on acquisitions in the past three years as he attempted to build a one-stop solution for TV, cable, the Internet, wireless and wired local and long-distance phone service. It hasn't worked, and now he's busting the company into four parts.

Breaking up has worked in the past. Ma Bell's dizzying array of spin-offs, including the regional Bell companies in 1984 and Lucent and others in the '90s, have mostly done well by being good businesses to start with and then being set free to raise capital, allocate resources independently and enter new markets as they liked.

The current dismantling is about failure. AT&T;'s business services and long-distance operations are suffering, reflected in the stock's 64% plunge since March. And two of the parts, initially anyway, won't enjoy the freedom so vital to effective spin-offs. The cable and consumer businesses will be tracking stocks. Their earnings will be reported separately, but managers must still compete for resources within AT&T.;

Big stocks. Big decisions. Neither company is likely to hurt you badly. GE will keep growing without Welch, and AT&T; is getting so cheap that it will attract bargain hunters. But as icons go, both will disappoint.

E-mail Dan at kadlec@time.com . See him Tues. on CNNfn at 12:20 p.m. E.T.