More Sensible Sex Discussion

Cathy Young's article ("Sex & Sensibility," March) was the most sensible discussion of gender differences I've run into in a long time.

One addition: You can't use gender stereotypes (or their absence) to predict what fields women will enter when the glass ceiling is lifted. You have to look at where they were before. Hence, bookkeepers became accountants (myself among them). Nurses went to medical school. Legal secretaries went to law school. Engineering secretaries didn't become engineers because, by and large, there was no clear and obvious progression from one to the other. The progression from mechanic to technician to engineer was and still is a largely male career path from bottom to top.

The next rush was women entering professions traditional to the men in their families: Policemen's daughters became cops and soldiers' daughters joined the armed services. Again, these were things they were familiar with, if only (this time) at second hand.

Patricia (Pat) Mathews

Microsoft's Misdeeds

I greatly enjoyed James V. DeLong's piece on antitrust ("The New Trustbusters," March). I'm afraid he's right about the rebirth of antitrust activism–and this may be just the beginning. I knew I should have taken all the light bulbs with me when I left the Federal Trade Commission in 1989.

However, Mr. DeLong's visceral opposition to antitrust enforcement has prevented him from properly analyzing the Microsoft case. He says the government is trying to prove that Microsoft is a meanie and that the government hasn't successfully proved that Microsoft is a monopoly. Wrong on both counts.

Microsoft so obviously has a monopoly that denying it doesn't pass the laugh test –even in Washington. More than 90 percent of PCs are shipped with Microsoft Windows installed; Microsoft's prices have risen faster than hardware components; and entry into the PC operating system market is clearly all but impossible. If that doesn't indicate the existence of a monopoly, then there are no monopolies.

Second, the government has been trying to prove not that Microsoft is a meanie (Microsoft has made that case), but that the meanie engaged in restrictive practices which produced no economic efficiencies. According to press accounts of the trial, the government seems to have had considerable success in making that case.

Antitrust law says there are certain restrictive practices which, when engaged in by monopolists, are illegal. Even "public choice" theorists like Mr. DeLong who realize that the history of antitrust enforcement is dark indeed should be able to understand that a good case comes along every now and then.

Daniel Oliver
Chairman (1986-89) Federal Trade Commission
Washington, DC

"The New Trustbusters" didn't address what I consider to be Microsoft's most insidious practice: shrink-wrap licensing. Microsoft may not be the only company that uses this method, but its licenses are among the most restrictive.

It is impossible to buy a laptop without paying for Windows. Even VA Research, a vendor that specializes in Linux, includes Window disks and manuals when you buy a "naked" laptop (no operating system) from them. This is because all the laptop manufacturers have a per-processor licensing arrangement with Microsoft. Microsoft gets paid even if Windows is not installed or used. (Licensing applies to desktops and servers as well, although you can avoid it by building your own.)

According to Microsoft's End User Licensing Agreement (EULA), you cannot sell, rent, or lend the software because it is licensed to that particular machine. Also, you are not allowed to use the software if you do not agree with the license. So non-Window users are in the position of paying for Windows even if they do not use it.

For now, there seems to be a legal out for non-Windows users. A clause in the EULA states: "If you do not agree to the terms of this EULA, PC Manufacturer and Microsoft are unwilling to license the software product to you. In such event, you may not use or copy the software product, and you should promptly contact PC Manufacturer for instructions on return of the unused product(s) for a refund."

It's simple enough: Return Windows for a refund. However, it is not so simple in practice. Microsoft's position is that the licensing agreement is between you and the manufacturer and you have to contact the manufacturer for the refund. The manufacturers' position is that their licensing agreement with Microsoft prevents them from giving a refund. There is some validity to this argument because the manufacturers' agreement with Microsoft prevents them from disclosing the per-processor price. A refund given specifically for Windows is a violation of the contract.

Microsoft and the laptop manufacturers are rushing to close the refund loophole even though only five people have gotten a refund (between $5 and $70) to date. The Australian version of Windows '98 removed the reference to a refund. The manufacturers have also tightened their definition of what constitutes acceptance. Compaq has stated that you have accepted the license if you turn on the computer. Dell is taking the position that you have to return the entire machine if you will not consent with the EULA.

Another problem is that you are not allowed to see the EULA until you have already purchased the laptop. You cannot find a copy of the EULA on Microsoft's Web site. If you call Microsoft for a copy of the EULA, you are told that you have to get it from the manufacturer.

Buyers are stuck in a Catch-22. You can't read the EULA until you pay for the software. Once you have paid for the software, no refunds.

I realize that these disputes fall more into the category of contract law than antitrust law. But they demonstrate that Microsoft does have monopoly power on the PC market and they use this power to force every purchaser to pay for Windows even if they do not use it.

Open source software, Linux in particular, has started to make inroads against Microsoft's monopoly. Apache is used on 52 percent of Web servers while Microsoft's IIS is only used on 26 percent. Several software manufacturers have announced that they will port some of their server software over to Linux. More companies are replacing their NT servers with Linux, and a growing number of users are becoming more vocal about their opposition to the "Microsoft Tax."

The current battle between Microsoft and open source makes an interesting test of the statement, "The only monopolies that can exist are the government enforced monopolies." Will Microsoft be able to use their dominance of the desktop OS to force the Internet companies to switch from the current open standards to Microsoft's proprietary APIs? Will Microsoft's policy of "embrace, extend, and extinguish" work against open source? Will Microsoft be able to leverage their near monopoly control of the desktop market to gain control of the server market? Only time will tell.

Garnet Harris

James V. DeLong replies: Since Daniel Oliver is one of the people lauded in my article for the salutary antitrust reforms of the '80s, I hesitate to dispute him. But I will do so anyway.

A continuing problem in antitrust law is the slipperiness of the terms. Oliver says that Microsoft has a monopoly, but this really does not tell you much. First, the "90 percent" he cites is a dubious number; see the thoughtful piece by Alan Reynolds of the Hudson Institute in The Wall Street Journal on April 9, 1999. Second, market share is not the important part of the issue. The real question is market power, which is the power to extract high prices, and whether such power is enduring over time. I doubt it. If Microsoft tried to raise prices more than minimally, I think it would rapidly find itself under siege. For more on this debate, check out the testimony of economist Franklin Fisher on the Department of Justice Web site (www. and the white paper "Why Does Microsoft Charge So Little for Windows?" on Microsoft's site.

Similarly, I disagree with the comment that entry is "all but impossible." Microsoft is saddled with a legacy called DOS, which most nerds regard as clunky. In developing Windows NT and Windows 2000, Microsoft bit the bullet and started over, writing on blank paper. The market was open to any company that preferred to innovate rather than whine, to leapfrog Windows, build from scratch without the burden of DOS clunkiness, and then use this newness as a marketing pitch to the I-hate-Microsoft crowd and to the hard-eyed corporate IT managers who are well aware of the advantage of multiple sources. One problem of aggressive but vague antitrust doctrines is that they seduce business people into trying to get the government to do their work for them.

As to the business practices, I have not read the whole trial transcript, but I am skeptical. They look to me like hard bargaining over the terms of cooperation and partial integration among firms that are partly allied and partly antagonistic–"coopetition" is a current buzzword. I doubt that Microsoft is in the business of shafting developers for the simple reason that this is not where its self-interest lies. By being fair to developers, Microsoft can get the benefits of intense work by creative people. Reputation over time really counts in this area, and Microsoft's victory over OS/2, which was by no means foreordained, may well be due to its superiority over IBM in dealing with the developer community.

On the other hand, I do not doubt the existence of individual cases of over-reaching. But such cases are for contract lawyers, not antitrust enforcers. Antitrust by evil anecdote is bad policy. I was once involved in a case of alleged price fixing in which a couple of huge corporations were assessed more than a billion dollars in damages. The evidence? Three conversations among low-level salesmen at trade association meetings during which prices were mentioned. This is the kind of thing that gives antitrust a deservedly bad rep.

Mr. Harris raises a different set of issues, which are, as he notes, primarily questions of contract and intellectual property rather than antitrust. However, the software companies have a devilish problem of combating the rampant piracy that permeates the industry, and I am a little puzzled over what Mr. Harris wants.

Does he want to negotiate a license that lets him use his software on more than one machine? It is the nature of the world that large users can negotiate for special treatment, but individuals cannot because the transaction costs are too high.

In any event, the computer hardware/software package–the total functionality–is dirt cheap and dropping every year. The EULA system may be annoying, but it is part of the whole system that has showered computer users with incomparable blessings. The whole controversy reminds me of the old joke about the man who jumped into a raging torrent, pulled out a drowning child, and triumphantly returned him to his mother. She glared at the rescuer and said: "Where's his hat?"