On Wednesday, Judge Thomas Penfield Jackson invited government attorneys to draft an even more sweeping proposal for breaking up Microsoft. But a split -- whether into two Baby Bills, three or more -- isn't the way to increase competition. Judge Jackson's zeal seems destined to hurt consumers, the very constituency antitrust law seeks to protect.

Last month, Judge Jackson found Microsoft liable for monopolizing the operating-system market by tying its Internet Explorer browser to Windows and by attempting to monopolize the browser market. The Justice Department recommended splitting Microsoft into two pieces: an operating-system company focused on Windows and an applications company selling other software.

In Wednesday's hearing, Judge Jackson demanded a new proposal that considers a three-way breakup. A vertical three-way breakup would place Explorer in a stand-alone company, separate from a Windows company and an applications company. A horizontal three-way breakup would split Microsoft into three identical pieces, giving all intellectual property to each of the three. A third, hybrid proposal would combine elements of both of these, resulting in five new companies -- an applications company, an Internet company and three competing operating systems companies. The Justice Department is supposed to submit its revised plan today.

Yet there are signs that even the government is worried about going too far. When asked by Judge Jackson why it hadn't considered a three-way breakup, the government's lead attorney, David Boies, replied that such a remedy would have "time-consuming and disruptive effects" on Microsoft and on the industry as a whole.

That's exactly right. And despite Judge Jackson's best efforts, any breakup is unlikely to increase competition. Justice's original two-way breakup plan was premised on the hope that an autonomous applications company would create a new operating system to compete with Windows. But more than 70,000 applications run Windows, creating what the government calls an "applications barrier to entry" in the operating-system market. However capable the new applications company, it still wouldn't be able to single-handedly create a successful rival operating system. Separately, even with a new applications company's support, Microsoft's biggest operating-system competitor, Linux, is unlikely to become a serious desktop threat to Windows.

What a breakup is likely to do, however, is raise prices. Microsoft's main "anti-competitive" actions -- giving away Explorer or selling Windows at low prices, in order to sell other Microsoft products -- have only helped consumers. After a breakup, none of the new divisions of Microsoft will have any reason to subsidize each other, but will be forced to sustain independent development costs. The breakup is therefore laying the groundwork for the Baby Bills to exercise their market power and increase prices.

This problem becomes even worse with a three-way vertical breakup. A browser company would have to charge for its product -- something Microsoft currently doesn't do -- making Internet access more expensive.

If a vertical breakup is a bad idea, a horizontal division would be even worse. Under this scenario, each Baby Bill would begin producing its own "improved" version of Windows. It is unlikely these systems would be compatible with each other. Some software vendors might write programs that worked with all versions; others might not, greatly reducing the range of consumer software. Such incompatibilities would also hurt shareholders, since the combined value of the Baby Bills probably would be smaller than that of the original company. The situation would also be a nightmare for corporate technology departments, which would struggle for years with the new operating systems until one becomes dominant.

Some breakup supporters point to the successful 1981 dismantling of AT&T as reason for splitting Microsoft. It's true that, since the AT&T breakup, long-distance prices have fallen, but the AT&T and Microsoft cases are very different.

The various parts of the old Ma Bell empire were managed separately, so it wasn't hard to separate them. There are no such natural divisions within Microsoft. AT&T also had an abundance of managers to help cope with the breakup. Microsoft employs only about 20 senior executives; wrenching them apart could cause huge managerial problems. Finally, AT&T was a regulated utility and the companies emerging from the breakup were guaranteed to stay interconnected. In contrast, the Microsoft breakup is likely to lead to incompatibilities and further loss of efficiency.

The government hasn't yet demonstrated that other "remedies" or punishments wouldn't work. Why is it not enough to impose "conduct" restrictions on Microsoft, for example limiting its ability to sell different products as a bundle? Such restraints may involve harder implementation work for the government, but they will not be as damaging as dismantlement.

A Microsoft breakup is likely to cripple the company and create huge losses for both consumers and the economy. In his desire to enforce antitrust law, Judge Jackson needs to remember just who it is he's trying to help.