Path Dependence, Imperfect Markets, etc.

by on February 29, 2008 · 5 comments

I have to at least partly agree with Tom about the HD-DVD vs. Blu-Ray battle, and about libertarians’ tendency to sometimes over-sell the outcome of market processes. Libertarians have been known to go overboard in insisting that the outcome of market processes is superior to other potential outcomes in every conceivable respects, even when it’s clearly not so. To take the obvious example, as a Mac user, I’m not prepared to concede that simply because Microsoft has a larger share of the operating system market than Apple, that Windows must be a better operating system than Mac OS.

In particular, network effects and path dependence are real phenomena. For example, the x86 architecture really is a kind of Frankenstein’s monster that has limped along for decades despite, not because of, its underlying technical design. Yet it has become the undisputed champion of desktop computing because of network effects—there is now vastly more desktop software written for it than any other architecture, and so that’s what the market demands. From the perspective of an electrical engineer, this is kind of a pity. Newer architectures like the Alpha have a variety of advantages, and if we could somehow wave a magic wand and cause everyone to suddenly have an installed base of Alpha software instead of x86 software, it would almost certainly be an improvement.

On the other hand, I think non-libertarians (and geeks in particular) sometimes make the reciprocal error of overestimating the importance of their parochial interests when evaluating the outcome of market processes. So while I think it’s true that Mac OS X is a technically superior operating system than Windows Vista, I don’t think we can leap from that to the conclusion that the market process needs to be changed. I think that what happened, fundamentally, is that Microsoft is worse at engineering but better at other aspects of their business. For example, much of their success has come in the corporate market, largely due to a skilled sales force, decent tech support, and a relentless focus on adding the features demanded by their business customers. People don’t purchase an elegant technical architecture, they buy a bundle of software and services that meet a certain set of needs. There is no particular reason to think that the company with the most technically-elegant product will necessarily have the most compelling bundle of related products and services. It’s no more rational for geeks to buy a computer solely because it has a RISC chip inside than it is rational for a grandmother to buy a computer solely because she likes its color. People want things for a variety of diverse reasons, and the only way you can figure out which factors are the most important is to give consumers a choice and see what they pick.

On the HD-DVD vs. Blu-Ray battle specifically, I don’t think the battle had much of anything to do with the technical merits of the platforms, per se. As Tom notes, HD-DVD had a somewhat less onerous DRM system, which gave it an advantage in wooing Silicon Valley (and Redmond) but a disadvantage in wooing Hollywood. But I think it ultimately came down mostly to mundane business concerns: time to market, supply chain management, pricing, marketing savvy, strategic partnerships, etc. The PS3 seems to have been a major factor in tipping things toward Blu-Ray. Had Blu-Ray come to market 6 months later, or HD-DVD six months earlier, it’s easy to imagine that would have changed the outcome. Likewise, if HD-DVD had been able to cut prices more aggressively, or if Sony had been unwilling or unable to keep Blu-Ray reasonable close to HD-DVD’s prices, that could have been bad for them.

I don’t think we can say that the victory of technically inferior architectures in the marketplace has any straightforward implications for public policy. As Will has argued, people are far too quick to jump from a claim about the sub-optimality of some market outcome to sweeping claims about the desirability of government intervention to correct the alleged problem. To take the canonical example of the Microsoft antitrust case, there certainly wasn’t anything the government could (or should) have done in 1998 to correct the “market failure” of Windows domination. Congress was not about to implement a cap-and-trade program for Windows licenses, much as it would have warmed the hearts of the Slashdot crowd. Windows domination was likely to continue for the foreseeable future regardless of what policymakers did. Whether Microsoft originally achieved its market share via shady business practices, luck, or technical superiority was almost beside the point because the dominance of Windows was now an established fact. Rather, the important question was given the dominance of Windows, whether government-imposed constraints on Microsoft’s behavior would be beneficial to consumers in related markets.

So I don’t think libertarians have any particular reason to get invested in debates about the importance of path dependence. If it were demonstrated that Blu-Ray won the format war by bribing Hollywood studios to support it, that wouldn’t especially undercut the case for free markets in consumer electronics unless you could simultaneously demonstrate that some other arrangement (devices designed by a federal agency?) would do better. It doesn’t hurt if we can prove that the market produces outcomes that are perfect in every respect, but I don’t think It’s true, and I don’t think any major tenets of libertarian theory depends on it being true.

Previous post:

Next post: