Bundling As A Rationale For Regulation? I Should Think Not

I am really glad that folks around here have picked up this bundling conversation. See, for example, Stephen Bainbridge’s observations, who argues that bundling is anticompetitive. I disagree, as long as the consumer has an opportunity to say “no” and use something else. And of course under the usual dynamics of R&D and competition, some people will be happy with IE, others will not, and as long as there is diversity of preferences, a monopoly browser is unlikely to exist in reality.

Not surprisingly, Brad DeLong feels he’s been harmed by Microsoft too, and argues accordingly. The comments on his post are particularly interesting, especially the one from Phil Hallam-Baker, who was on the original web team at CERN.

Alex Tabarrok has a great post from this morning on the topic, focusing on Brad’s argument that he’s been harmed. Alex’s argument does three important things: it focuses us on the fact that when goods have complementarities and we don’t allow bundling, if both markets develop into monopolies then you have the double monopoly or double marginalization problem, which is a textbook recipe for deadweight loss and a decline in consumer well-being. He also brings in the fact that vigorous competition for browser market share is a form of R&D-based franchise bidding, in which the competitors know they are competing for a large and stable market share. Alex’s argument also reminds us of the importance and value of contestability, which a lot of the pro-regulation voices on this topic discount, incorrectly I believe.

Arnold Kling also has a post on the subject, correctly noting that bundling is all around us, always has been, and has value to consumers. He concludes:

Regulators could argue that bundling by Microsoft or the cable companies needs to be regulated because those companies have monopoly power. But I would rather see product specifications and pricing set in a market, however imperfect, than set by a government bureaucrat. At best you are exchanging one uncompetitive decision process for another.

I do not see any way to preserve the free market system if you decide that bundling provides a rationale for regulation.

Consumers making choices decide on the optimal degree of bundling. Some of them are large consumers that move the market share and the bundling more toward single provider (like the federal government’s procurement decisions on its computers). But we can still say no, and we have options.

Phil Hallam-Baker is right, that Microsoft did get ahead in the browser part of the value chain by building a better browser. That is no longer necessarily the case. Note IE’s absence of tabbed browsing, for example, which is a great organizational tool when you have to have lots of pages open.

In my case, the only time I use IE is if a page is not loading properly in Mozilla. Uusally it involves Java script problems. No biggie.

BTW, do the Movable Type folks know that the little url/bold/italic/underline macros only show up in IE? I hard code everything on this blog b/c the cute little buttons at the top right only show up when I use IE. Not worth it. But if the MT folks want to improve MT, having the macros work with more browsers would be good.

10 thoughts on “Bundling As A Rationale For Regulation? I Should Think Not

  1. The problem is not that the consumer is directly inconvenienced by being given an inferior browser for free, when he can still download a superior browser for free if he wants.

    Why are these browsers free? Browsers are given away for free because the developers of browsers that people use are at an advantage in selling server software, since the job of the server software is to communicate with the browser software.

    The point of regulating bundling is not to prevent the monopolist extending the monopoly from operating systems to browsers — that is just adding features. It is to prevent the monopolist extending the monopoly from operating systems to web servers, via the browser.

    The situation with media players that the European Commission has just ruled on is equivalent. Again, the players are given away free by all competitors in order to support the market for their servers.

    It still remains to be argued that the regulation is actually useful, but the argument is about control of communication protocols, not adding features to desktop operating systems.

  2. Another problem with MS bundling apps: the network effect of apps that have proprietary protocols is another barrier to entry into the OS market. Migrating to Linux would be easier if it weren’t for the fact that the entire business world communicates in MS Word and MS Excel. Apple won’t go near the x86 OS market for fear of having MS yank their offering of MS Office.

    If a MS audio codec becomes the standard for audio streaming, Linux (and possibly OSX) is further marginalized.

  3. I hard code everything on this blog b/c the cute little buttons at the top right only show up when I use IE.

    I never knew the cute little buttons existed until I saw an ex IE user complaining.

    Nifty tools like w.bloggar (which I use) or SharpMT provide much more then the little buttons.

  4. Huh? MSIE has _extrmely_ little to do with IIS _or_ ASP.

    MSIE is not standards compliant, and well, now that Longhorn is in the works, will never be. THAT IS THE PROBLEM WITH IT.

    Exactly what kind of proprietary ‘server’ product was Netscape after? Safari? Mozilla? And please, don’t insult our collective intelligence and say Apache.

    The message in the posting is dead-on. If Netscape would have won, they’d have monopolized things nearly as much as MS. Still, with all the various browsers out there, I’m not sure _anyone_ can really speak of a ‘monopoly’.

  5. Netscape formerly developed Netscape Enterprise Server, alias iPlanet (in conjunction with Sun, IIRC).

  6. > If a MS audio codec becomes the standard
    > for audio streaming, Linux (and possibly OSX)
    > is further marginalized.

    Only because it is a “proprietary standard,” which can only exist because Micro$haft was granted a legal monopoly on it that is enforced at government gunpoint, i.e., a “patent.”

    De-facto standards only represent a barrier to entry when competitors are artificially _prevented_ from market entry by the totally artificial fiction called “intellectual property,” which is fact is nothing but a government-granted and government-enforced monopoly. If there were no such prior restraint, the existence of a de facto standard is actually an _aid_ to entry, since it increases one’s pool of potential customers —- the additional effort required to reverse-engineer the standard notwithstanding.

    The Framers of the U.S. Constitution at least had the wisdom to specify that a patent monopoly should only be granted for a _LIMITED_ amount of time, and that the patented knowledge should become public domain after the person who disclosed the patented item had had a _REASONABLE_ opportunity to _ATTEMPT_ to earn a _REASONABLE_ profit from it, _IF_ they could. The Federalist Papers make it quite clear that the Framers =NEVER= intended that corporations be allowed to own patents, nor that the patent monopoly become effectively _UNLIMITED_, being extended essentially _in perpetuum_ (or until the patent monopoly item ceases to any longer be profitable) as the APPALLINGLY stupid court rulings on patent law during the previous century have perverted the originally _LIMITED_ patent monopoly into! (If anything, given how rapidly technology is now progressing, and how rapidly new items are becoming technologically obsolete, the period of patent monopoly should have been _DECREASED_ by the courts, not increased!)

  7. NOTE: The cost of “reverse engineering” (stealing) a de facto standard is far lower than the cost of developing the capability which becomes the de facto standard. That is why there are patents.

    Cross-licensing agreements between competitors are very common in industries in which multiple R&D participants have created technologies of value to customers and each other. However, it is difficult to cross-license if you have nothing of value to bring to the party.

    Research engineers are more expensive to retain and support than “reverse” engineers. Also, I am unaware that any “reverse” engineer ever CREATED anything other than a working copy.

  8. The amount of “research engineering” needed to create a file format is going to vary by application; I can see plenty of research in creating media player files (trading off sound/image quality, size, etc.), but I don’t see that in word processor files. (Particularly not in the case of .doc files, which are apparently a crufty mixture of text and binary goop. Ironically, the reverse-engineering of the format by OpenOffice has supposedly been so successful that OO can recover many corrupt documents which will crash MS Office.)

    Browsers are a bit of a different story, because the “de facto standard” for HTML is the corpus of HTML documents on the web, many of which were produced by hand (or by tools so bad as to make no difference). Microsoft enjoys a significant advantage here because of its market dominance; if a new release of IE changes its parsing behavior (within reason), authors will assume they’re doing something wrong (IE is the de facto standard, after all) and fix it. If another browser doesn’t reverse-engineer IE’s parsing to the letter, it’s far more likely that an author will blame it on the browser rather than fixing his markup. (I speak from bitter experience.)

  9. Here are some (perhaps) counterintuitive results from Stan Liebowitz and Steve Margolis, that seem to support our hostess’s point:


    “In other words, Microsoft was not acting like a textbook monopolist (raising prices) in the market where it clearly has a structural monopoly. If anything, the prices in the market where it was dominant were lower than in the markets where it was competing. After Microsoft had come to dominate the PC market, it might have been expected to raise prices, but it lowered them dramatically. We can not attribute this result to some idiosyncratic difference between PC and Macintosh markets since Microsoft equalized the prices in the two markets after gaining dominance in both. What might be going on, then?

    “One answer, that appears consistent with all our findings, is that Microsoft worries about competitors even when it has a very large market share. Such concern about potential entrants might explain why Microsoft has not lost any markets it has gained. The decline of Lotus might have been due to an erroneous lack of such concern.”

  10. m.bloggar only works on my Windows box, drat it. What about when I’m working on my stylish, fabulous little Macintosh laptop?

Comments are closed.