A Market that Probably Shouldn’t Exist

by on June 1, 2006 · 8 comments

Soon computer viruses, spyware, and other malware may be a thing of the past.

If, that is, you pony up an extra $50 per year to Microsoft. This week, the software giant (obligatory cliche) announced a Windows Live OneCare, which promises to provide subscribers with “round-the-clock protection and maintenance–virus scanning, firewalls, tune ups, file backups, the whole nine yards.” This comes after years of demands from customers that the company shore up its operating systems and focus more attention on security.

Then again, regular major security flaws haven’t done much to erode Microsoft’s market share, and so perhaps there is some justification for making security an add-on service rather than a part of the operating system itself.

But one factor in the decision to sell Windows Live OneCare live, according to several industry analyst, is that Microsoft is aiming to avoid the attention of the antitrust squad. According to at least one commentator, this restrained competition is laudable:

It’s fair to argue that something like OneCare ought to be built into Windows. But think about what would happen next: The competition would probably find its air supply cut off, as people flee from the cost and complexity of adding third-party replacements for something built into Windows. And recent history has shown that Microsoft tends to slack off if it doesn’t feel a competitive threat–witness how Windows Media Player and Internet Explorer stagnated until iTunes and Firefox got Microsoft’s attention.

With OneCare, Microsoft is trying to clean up its own mess while preserving competition. That’s especially important in the security-software market, where many of the companies that have dominated it so far seem to have adopted Microsoft’s worst habits of sloth.

Remember that the ostensible purpose of antitrust law is to protect consumers. When companies become monopolists, the argument goes, they can raise prices and extract undue sums from their customers, while innovation, and the consumer benefits that come with it, stagnates.

But here, it’s consumers that are getting the short end of the stick. For fear of the antitrust boys (and perhaps because the market will bear it), Microsoft has essentially unbundled what was traditionally a part of an operating system–strong security to protect users’ files, networks, and operating environment–and is now selling it as a separate service for $50 per year. (Among UNIX-derived operating systems, for example, including advanced security auditing tools is the norm.) If antitrust concerns are behind this decision, then antitrust is costing security-conscious consumers, by my calculations, $50 per head per year.

That’s protection? It seems more like a protection scheme. One wonders whether Microsoft isn’t exactly upset that it can hint about antitrust concerns as its reason for unbundling security so as to avoid consumer ire.

Even if not in the U.S., Microsoft surely would have run into antitrust problems in the EU, where antitrust authorities are much keener to meddle, especially when a U.S. firm is involved. Recall that the company’s last settlement with the EU gave birth to “Windows N,” a version of the operating system without a bundled media player. (Not surprisingly, “N” has not exactly been a hit with consumers, leading to the joke that the ‘N’ stands for ‘no one.’)

Inadvertently, this time, the antitrust enforcers may have just created yet another version of the popular operating system: Windows NS. ‘NS’ stands for ‘no security,’ and it’s likely the version that you’re using right now.

Better than restrained competition–the sort of term that makes any economist nervous–is plain competition, unburdened by the threat of arbitrary government meddling. Without the antitrust hatchet hanging overhead, Microsoft’s OneCare might well just be a part of the operating system, correcting (perhaps; hopefully?) a series of flaws in the design of its products that have frustrated users for years.

  • enigma_foundry

    This has nothing to do with anti-trust issues, and everything to do with Microsoft’s attempt to move from a shrink wrap business model to a subcription and a service business model.

    As I have noted in my comment to “A Different Type of Technological Protection” article, the shrink wrap model is on decline and the ascendency of service oriented model will be on the ascendency as FOSS continues its take off.

    The line that this subcription based service is an anti-trust issue is pure MS propaganda.

    Incidently, many have criticized FOSS for this move (see IP Central Posts, which do not allow comments) somehow implying that the software must be low quality to move to a service oriented model. Well, that is not the case at all–in the case of SuSE Linux for example.

    Now here is an example of a shrink wrap company doing what those at IP Central have accussed FOSS of doing–moving to subcription based sales to cover up their deficient software….

  • http://yglesias.tpmcafe.com Matthew Yglesias

    There’s no reason to think the $50 fee for the service reflects an actual $50 increase in costs to consumers. If these features were bundled into the operating system and offered consumers $50 in additional value on top of what’s already there, then Windows itself would just cost $50 more.

  • http://enigmafoundry.wordpress.com eee_eff

    This has nothing to do with anti-trust issues, and everything to do with Microsoft’s attempt to move from a shrink wrap business model to a subcription and a service business model.

    As I have noted in my comment to “A Different Type of Technological Protection” article, the shrink wrap model is on decline and the ascendency of service oriented model will be on the ascendency as FOSS continues its take off.

    The line that this subcription based service is an anti-trust issue is pure MS propaganda.

    Incidently, many have criticized FOSS for this move (see IP Central Posts, which do not allow comments) somehow implying that the software must be low quality to move to a service oriented model. Well, that is not the case at all–in the case of SuSE Linux for example.

    Now here is an example of a shrink wrap company doing what those at IP Central have accussed FOSS of doing–moving to subcription based sales to cover up their deficient software….

  • http://policy.heritageblogs.org Andrew Grossman

    Matt, you are right in that the actual cost increase could be less than $50, relative to a bundled scenario, but I would guess that it wouldn’t be much less than that. Here’s why:

    1. MS hasn’t had great success pushing customers into annual payments for the OS, especially non-corporate buyers. If a user buys an OS once, say, every 4 years, the cost increase would have to be the present value of 4 annual payments to equal out the cost. I don’t think the market for OS’s is necessarily that elastic–remember, MS is competing primarily against older versions of Windows.

    2. Consumers don’t buy Windows, OEMs do and then install it on computers that are then sold to consumers. MS could not possibly squeeze the full $50 (or the present value of n annual payments, as above) out of OEMs, who buy Windows far below its retail price. Bundling lets customers use OEMs to get a lower price.

    3. Most customers don’t value security all that much. Balanced against the security conscious, who would happily pay $50 / year, the optimal price to sell to this broader market, as part of a bundle, would have to be less than the present value of n years of $50 payments. If customers who value security little far outnumber those who put a great value on it, then the optimal price would be far less. (optimal==profit maximizing)

    4. There are collective action issues w/r/t viruses and other malware. Total consumer welfare is likely to be greater if more people use tools that block viruses and the like–networks would be faster, there would be less spam from ‘zombie’ computers, etc. No individual consumer is willing to pay anything for this potentially great benefit, but it would still lower the net cost to consumers of the new software/service.

  • http://yglesias.tpmcafe.com Matthew Yglesias

    There’s no reason to think the $50 fee for the service reflects an actual $50 increase in costs to consumers. If these features were bundled into the operating system and offered consumers $50 in additional value on top of what’s already there, then Windows itself would just cost $50 more.

  • http://policy.heritageblogs.org Andrew Grossman

    Matt, you are right in that the actual cost increase could be less than $50, relative to a bundled scenario, but I would guess that it wouldn’t be much less than that. Here’s why:

    1. MS hasn’t had great success pushing customers into annual payments for the OS, especially non-corporate buyers. If a user buys an OS once, say, every 4 years, the cost increase would have to be the present value of 4 annual payments to equal out the cost. I don’t think the market for OS’s is necessarily that elastic–remember, MS is competing primarily against older versions of Windows.

    2. Consumers don’t buy Windows, OEMs do and then install it on computers that are then sold to consumers. MS could not possibly squeeze the full $50 (or the present value of n annual payments, as above) out of OEMs, who buy Windows far below its retail price. Bundling lets customers use OEMs to get a lower price.

    3. Most customers don’t value security all that much. Balanced against the security conscious, who would happily pay $50 / year, the optimal price to sell to this broader market, as part of a bundle, would have to be less than the present value of n years of $50 payments. If customers who value security little far outnumber those who put a great value on it, then the optimal price would be far less. (optimal==profit maximizing)

    4. There are collective action issues w/r/t viruses and other malware. Total consumer welfare is likely to be greater if more people use tools that block viruses and the like–networks would be faster, there would be less spam from ‘zombie’ computers, etc. No individual consumer is willing to pay anything for this potentially great benefit, but it would still lower the net cost to consumers of the new software/service.

  • Steve R.

    I looked at Microsoft’s “Service Overview” and I would conclude that this product would provide no extra value to the consumer. Other than “automating” certain actions, the “features” of Windows Live OneCare are already part of the Windows operating environment. Basically, Microsoft is repackaging what it has already sold to you and selling it to you again. As with enigma_foundry, this no more than attempt to gouge the customer and has little to do with antitrust issues.

  • http://www2.blogger.com/profile/14380731108416527657 Steve R.

    I looked at Microsoft’s “Service Overview” and I would conclude that this product would provide no extra value to the consumer. Other than “automating” certain actions, the “features” of Windows Live OneCare are already part of the Windows operating environment. Basically, Microsoft is repackaging what it has already sold to you and selling it to you again. As with enigma_foundry, this no more than attempt to gouge the customer and has little to do with antitrust issues.

Previous post:

Next post: