Politech mailing list archives
FC: More on new antitrust book sides with Microsoft, closing arguments
From: Declan McCullagh <declan () well com>
Date: Wed, 22 Sep 1999 11:40:35 -0400
I have been out of town in Pittsburgh and then busy at the Microsoft closing arguments. I was filing from the courtroom via my Palm VII yesterday with frequent updates to the article on wired.com's live site. It worked out pretty well, with near-perfect 9600 bps service through BellSouth's national pager network. Laptops are banned from the courtroom (and my Ricochet can't penetrate the thick walls anyway), but even the ever-so-persnickety guards who threaten to arrest you if you walk on the carpet (not kidding) didn't seem to care about me quietly scrawling Graffiti text. Highly recommended. My article: http://www.wired.com/news/news/politics/story/21860.html -Declan
Date: Tue, 21 Sep 1999 15:26:47 +0100 To: declan () well com From: Gus Hosein <i.hosein () lse ac uk> Subject: Re: FC: Review: New antitrust book sides with Microsoft Declan, FYI, The Economist also did a review of the issue of network externalities in its edition last week. It's a thinner version of your article, but equally interesting given the context. gus ECONOMICS FOCUS Lock and key According to some economists, today^Òs new technologies call for an entirely new economics: one that, unlike the orthodox sort, can cope with pervasive externalities and increasing returns. Are they right? TWO sorts of claim are made about the ^Ónew economy^Ô. The first says that technology is spurring growth so dramatically that old assumptions about productivity, inflation, profits and so on no longer hold. This is a bold position, but it poses no great challenge to orthodox theory: old-fashioned ^Óneoclassical^Ô economics can comprehend it very well. The second claim is different. It says that orthodox economics itself needs to be revised wholesale to take account of the way the new economy actually works. The ^Ónew theorists^Ô, let us call them, say that technology-driven market failures are now pervasive. The chief culprit is increasing returns. In a world where unit costs fall without limit as output rises, monopoly thrives: a bigger firm can always undercut a smaller one. Industries based on knowledge, the argument goes, are especially prone to increasing returns, and hence to monopoly. Think of software. The costs are largely fixed and upfront; once they have been incurred, production can be expanded without limit at very little cost. Increasing returns in the form of ^Ónetwork effects^Ô can affect consumption as well as production. A common feature of the new technologies, it is argued, is that their value to any user increases in proportion to the number of users. The result is that, once a product is established in the market, demand for similar products will collapse: consumers get ^Ólocked in^Ô. And if they get locked in to a bad product, you have another market failure to compound the first one. The classic example of the ^Óbad standard^Ô, or of ^Ópath dependence^Ô as this syndrome is called, is the QWERTY keyboard: the layout makes no sense, it is claimed, but by an accident of history it has established itself and there is no getting rid of it. Our Economics focus of April 3rd discussed ^ÓThe Fable of the Keys^Ô, a paper by Stan Liebowitz and Stephen Margolis, which showed that the QWERTY story was wrong, because the standard layout is not in fact demonstrably worse than the alternatives. This cast doubt on the whole new-theory movement. Now the authors have brought out a book, ^ÓWinners, Losers and Microsoft^Ô (published by the Independent Institute). It includes the keyboard paper but moves on from that, by way of a close look at the software industry, to a broader and deeper attack on the would-be heretics. By a long way, it is the best single thing to read on this tangle of issues. The book reviews the new theory carefully and in language accessible to the general reader, and then subjects it to a detailed empirical examination. At the end, very little of the fashionable critique of neoclassical economics is left
standing.
To begin with, the authors question the theoretical appeal of the path-dependence paradigm. Network effects are real, and it follows that lock-in is a possibility. But note that lock-in is inefficient (that is, it is a kind of market failure) only if the inferior product survives despite the fact that the benefits of switching would exceed the costs. If the inferior product survives because the costs of switching are high, that is as it should be: in that case it would be inefficient to switch. (Recall that the point of the bogus QWERTY story was that the benefits of switching would be great and the costs low: the market failure consisted in the difficulty of making a co-ordinated jump to the new layout.) Taking switching costs into account immediately narrows the extent of plausible market failures. The new theory needs to be qualified in another way, too. Where lock-in is a factor, it is wrong to suppose that consumers and producers will blunder on as if it were not. On both sides of the transaction, there is an incentive to find ways round the problem. On the demand side, groups of consumers can get together and co-ordinate their choices. On the supply side, producers can start by selling their superior new product at a loss: if it really is superior, the market will adopt it and move across. Or they can spend heavily on advertising. Or they can help newcomers to switch by promising compatibility, as when cable-television companies offer to convert old televisions to the new system. With these and other strategies, it becomes an empirical question whether inefficient lock-in is as common as is often supposed; it is certainly not self-evident. Turning to the actual evidence, the authors find no such cases at all. Again and again they show that good products win. The standard lock-in stories are examined and, like QWERTY before them, debunked. Betamax was not beaten by an ^Óinferior^Ô VHS video format: at the time, reviewers were divided over which system offered better quality, and VHS offered the unambiguous advantage of longer playing-time. The triumph of DOS over the Macintosh operating system is equally explicable. Macs cost more, and Apple had developed a reputation for changing operating systems so as to make earlier software redundant; until Windows 2000 (see article), Microsoft^Òs successive operating systems were consistently backwards-compatible (a feature that explains many of the problems that some Windows users complain of, but which is prized by many others). The book gives example after example of software that came and went, at one time dominating the market but then giving way to a better newcomer: throughout, reviewers^Ò ratings of the products explain market-share. Unsurprisingly, in view of all this, the authors take Microsoft^Òs side in the firm^Òs battle with the Justice Department. Their view of that endlessly complicated issue, set out in an appendix, is not so much a resounding acquittal as ^Ócase not proven^Ô^×but they would say this should suffice. Be that as it may, the case they make against the path-dependence paradigm as a way to see the world could hardly be stronger: it is a big idea that simply fails to stand up. ~~~~~~~~~~~~~~~~~~~~~~~~~~~ DH/DSS 2048/1024-bit key fingerprint: D8FA CF93 4C82 77C8 114C 33EA 5485 619E 3550 2083 key ID: 0x35502083 RSA PGP 2048-bit key fingerprint: D06E 15BA 3643 BD65 7CB4 AE14 CB2B E265 key ID: 0x6019F689 ~~~~~~~~~~~~~~~~~~~~~~~~~~~ Gus Hosein Tutorial Fellow Department of Information Systems The London School of Economics and Political Science Houghton Street, London WC2A 2AE ~~~~~~~~~~~~~~~~~~~~~~~~~~~
-------------------------------------------------------------------------- POLITECH -- the moderated mailing list of politics and technology To subscribe: send a message to majordomo () vorlon mit edu with this text: subscribe politech More information is at http://www.well.com/~declan/politech/ --------------------------------------------------------------------------
Current thread:
- FC: More on new antitrust book sides with Microsoft, closing arguments Declan McCullagh (Sep 22)