Mention “competition” and “standard setting” in the same sentence and you’re likely to get a wide assortment of reactions. I offer a few thoughts on several layers of competition and how they can affect emerging standards.For some, setting standards implies reducing competition because a particular set of complementary technologies is chosen from among the potential candidates before any products or services ever get to market. Others (mostly antitrust policy types) tend to think instead of capturing competition in the technology selection stage so that licensing fees for any relevant intellectual property rights (IPR) are kept reasonably low. Still others conjure images of one standard fighting another for a given market, à la the recent battle between Blu-Ray and HD-DVD. And indeed, each of these viewpoints is legitimate in its own right, representing some of the many layers of “competition” relevant in any standard setting effort. Each facet of competition can have an important role to play in terms of consumer welfare and market efficiency. With this in mind, I offer a few thoughts on several layers of competition and how they can affect emerging standards.rnrnConsider first competition at the upstream level, among technology innovators. While it may be the case that many IPR holders “compete” with one another to have their technologies chosen for inclusion in some standard under development, it is also the case that this competition does not occur in an open market in the same way that, say, DVD player manufacturers compete with one another for sales at local electronics retail outlets. Many observers have expressed concerns that because the technology selection process occurs within relatively small working groups of standard setting organizations (SSOs), it is open to strategic abuse. For example, for SSOs with simple majority voting rules, members can “pack” voting sessions with additional “members” as a means of either pushing their pet technologies or blocking a rival’s technologies. As another example, company reps at an SSO may push their firm’s proprietary technologies for inclusion, but not inform their fellow SSO members about IPR relevant for the technologies. In antitrust circles this latter ploy is known as “patent ambush”, something the US Federal Trade Commission has already pursued as anticompetitive and an “unfair method of competition” – see e.g., the Dell, Unocal, and Rambus matters. The European Commission has taken up the cause as well and is currently investigating Rambus.rnrnBut the important question in terms of technology competition is not whether current SSO rules can be gamed – clearly they can. Instead, the relevant question is what process is the best of the available standard setting options and how can that option be perfected through appropriate oversight and organization rules? rnrnThe logical alternative to having company engineers meet and vote to select technologies for inclusion in a standard is having different variants of a standard develop in parallel, with each variant competing in the end market to decide the winner. That process would certainly be based on traditional market competition and may be less easily abused strategically. Of course, this is what happens with de facto standard selection, one of the forms of competition mentioned at the beginning with the example of competition over the high definition DVD standard. That said, there are valid reasons that de facto competition for a market isn’t how many industry standards evolve today. De facto standards may get to market faster, but they can involve a tremendous amount of duplication of effort. And they can be confusing for consumers. In the DVD example it appears that many consumers simply deferred their purchase of any hi-def DVD equipment until the battle was resolved –– which didn’t occur until a large retailer with enough market clout (Wal-Mart) made a choice about which standard it would carry in its stores. Selection by one large retailer hardly seems more democratic than selection by a group of engineers participating in an SSO effort.rnrnIt seems clear there will always be a role for both de facto standards competing for a market and cooperative standards working within a market. Rather than pick one approach over another, I believe that we’re better off figuring out ways to deal with the risks of strategic behavior that each method of standardization can present. rnrnWe should also bear in mind that cooperative standards, even though a select group of representatives choose the technologies to include, frequently face market competition from other cooperative standards. For example, this is the case in mobile telecom in many parts of the world (the U.S. and much of Asia for instance), where cooperatively designed standards compete with other cooperatively designed but incompatible standards – or at a minimum where different generations of the same standard compete against one another, such as 2G versus 3G.rnrnAnother important layer of competition referred to above is that among the firms implementing an interoperability standard. As already mentioned, with de facto standards, this competition takes the form of rival firms taking incompatible “standards” to market and letting the purchasers choose the one eventual winner. With cooperative standards, implementer competition can take a couple of different forms. If the standard defines interoperability, then firms in the downstream market can compete on differentiated versions of the standard, matching the necessary technical components to ensure compatibility, but trying nonetheless for branded products with differing features, degrees of quality, price points, etc. Cell phones are a good example here too: there are many different handsets with a variety of features and design shapes, but all choose the network on which they will run and the internal workings of the handset conform to that network’s specifications.rnrnWith the aim of gaining a first mover advantage, firms sometimes implement a standard before it has been finalized. In other words, downstream firms can bet on which draft will be published in order to get their products to market sooner. Just like the post-standard-publication branding and differentiated goods competition mentioned above, this kind of competition is not necessarily bad. Yes, it raises the potential for “forking”, where multiple and possibly incompatible versions of a standard enter the market. But it is important to remember that the vast majority of firms participating in standard setting efforts are for-profit entities. Unless there are economic returns to participating in the hugely time consuming and expensive process of cooperative standard setting, firms simply won’t do it. Moreover, to the extent the firms do make bets of this sort, one of the concerns raised above about technology selection occurring among a small group of firm reps is addressed because different strains of a cooperative standard may indeed end up competing in the marketplace.rnrnThe general point that I hope emerges from the musings above is that while competition is generally a good thing, when it comes to standard setting we need to remember that competition comes in many different guises and often requires compromises and trade offs. We shouldn’t blindly work to increase one level of competition within cooperative standard setting – say by increasing competition in the end marketplace – without recognizing that by so doing we may affect competition at other levels. With a consumer welfare benchmark in mind, in some instances we may want to work for the injection of more competition; in others we may not. And at any rate, we likely want to carefully consider the point at which competition can do the most good for the consumers of the standard that finally does get to market.