|< Day Day Up >|| |
Douglass C. North (1990, 84) identifies only two sources of institutional change: changes in relative prices and changes in tastes. To that we can add a third: the creation of new resources in technical systems. When North attempts to reduce technological change to 'changes in relative prices,' he assumes that technology only enables us to do the same things more cheaply, e.g., to make transportation faster, weapons more deadly, information less expensive. That perspective overlooks technology's ability to create new collective action problems by throwing into the economy new resource spaces that must be allocated, regulated, and traded.
The preceding narrative about the Internet root makes it clear that resource creation is a significant and disruptive source of institutional change. The close historical parallel with radio spectrum management bolsters the significance of the argument.
The most striking feature of the ICANN regime is its perpetuation of scarcity at the top level of the name space. According to Paul Vixie, the keeper of the BIND software that implements DNS on almost all the world's name servers, millions of new top-level domains are technically feasible. The root zone is just a zone file, after all, and if the DNS protocol can support the .com, .net, .uk, or .de zones with millions of registrations in them, there is no reason to believe that it could not also support a root zone with millions of unique names. Beneath each top-level domain, of course, there can be tens of millions of second-level names, and below that, millions more third-level names (think of the number of members and user names under aol.com). The name space created by the DNS protocol is practically inexhaustible.
There is, however, a tremendous disjunction between the capabilities of the technical system and the behavior of the new institution. The new regime has been able to authorize only seven new top-level names over the better part of a four-year period. The abundance of the technology is stringently limited by rules and procedures imposed by a central authority. This has occurred despite numerous demonstrations of consumer demand for new names at the top level and the existence of many businesses eager to supply them.
What accounts for the artificial scarcity? It is the product of a vicious cycle that, while lamentable from a public policy perspective, neatly corroborates some of the recent, more pessimistic theories about institutions and institutional change. North's (1990) foray into the development of a new theory of institutional change was motivated largely by an attempt to explain how societies could settle upon and retain institutional forms that were inefficient and even destructive. To solve that riddle, North called attention to the 'symbiotic relationship' that exists between institutions- the rules constraining human action-and the organizations and human perceptions that evolve as a result of the incentives provided by the rules. He suggested that a society can get locked into a dysfunctional institutional framework when the rules reward people and organizations for acting in ways that perpetuate its inefficiencies. [1 ]This mutual reinforcement or 'positive feedback' can explain why 'natural selection' and competition don't eliminate inefficient institutions.
In the case of domain names, a Northian vicious cycle is clearly identifiable. We can spell out the stages as follows:
As the Internet was commercialized, the availability of only one global commercial top-level domain gave second-level domain names under .com a special value as an economic resource.
With common pool conditions in force, that special value stimulated hundreds of thousands of speculative, defensive, and abusive registrations.
The speculative and abusive registrations in turn provoked those with preexisting rights in names (mostly, but not exclusively, trademark holders) to lobby politically against further expansion of the name space. Other vested interests, such as incumbent registries and speculators with large holdings in the .com space, also benefited from closing off new entry.
Failure to expand the name space further enhanced the value and power of names in .com. This fueled more speculation, more hardening of the attitudes of trademark owners, more politicization of the DNS, and continued the cycle.
Although the restrictive attitude of the trademark holders was in many ways irrational, in that it enhanced the speculative value of names, it is probably unrealistic to expect a brand manager or trademark lawyer to understand the long-term benefit of standing by idly while their names are left unprotected in thousands of new spaces. (Indeed, it is precisely these kinds of 'mental constructs' that North tries to invoke in his theory of institutional change.)
Thus it should come as no surprise that almost all domain name disputes involve .com names. We need to be clear about what exactly is being disputed in these cases. Domain name disputes are rarely about actual trademark infringement as that term is normally understood in the law. [2 ]They are not really about the public use as an Internet address of words or marks similar to a trademark per se, because, as noted in chapter 11, trademarked character strings can appear anywhere in a Web site's URL: at the third or fourth level, after the slash, in other top-level domains, and on the Web site itself. In reality, the vast majority of domain name disputes are motivated by conflict over the right to get a favorable position in the second-level of the .com space. Conflict occurs because the second level of .com is reputedly the one most likely to be found, guessed, typed in, or remembered by users. What is disputed, therefore, is the ability of a name to attract traffic and attention. In that regard, not all domains are created equal: .com is the premium real estate. The problem of its dominance is exacerbated by high switching costs. Any business or organization that establishes an identity and a presence under .com is not going to want to change it.
The obvious solution to this is (or was) to blow away .com's special status by authorizing thousands of new names in the top-level space and encouraging users to distribute their registrations over a much broader array of TLDs. Most of the economic basis for name speculation and cybersquatting would be instantly eliminated by such a policy. It would also address the market dominance of Verisign/Network Solutions and the dangerous centralization of control over the Internet in a single entity's hands. Competing registries would offer customers a far more favorable set of contractual conditions, at least as long as the registration market continued to grow (FTC 1998).
That solution was apparent-and eminently possible-in 1996, when Jon Postel proposed to create 300 new TLDs. There were (and still are) hundreds of entrepreneurs willing to operate new registries, and thousands of ideas for new TLD strings. Yet the confusion and contested authority surrounding the root in 1996 prevented action and gave way to a period of intense politicization that allowed the vicious cycle to kick in. The cycle prevented the addition of any new open TLD for nearly seven years after the problems began.
The failure of draft-postel was a critical branching point in the evolution of Internet governance. When the new institutional regime finally was able to act, political consensus limited it to policies that perpetuated the cycle instead of eliminating it. The cycle was perpetuated in the following ways:
The value and dominance of .com was enhanced by opening it up to multiple, competing registrars. The lower price and more active marketing of .com names increased speculation.
The Internet governance process could not get large numbers of new TLDs past the trademark interests and other groups with a vested interest in a restricted name space; on the other hand, it could not avoid creating some new TLDs because of the demands of potential entrants. So it opened up a small number of new TLDs, two of which (.biz and .info) are put forward as alternatives or substitutes for .com.
With that restrictive policy in place, trademark holders will rush to register their existing .com names in the new generic TLDs to preempt any competition with their .com holdings. The mad desire to protect existing names also magnifies the opportunity for speculation.
In other words, a restricted name space reinforces the land rush mentality and potential for abuse that created the conflicts to begin with. And by reinforcing the problems, it rationalizes the continued existence of a restrictive regime that regulates the conflicts via collective action. [3 ]If there is artificial scarcity, there will be a land rush; if there is a land rush, there will be speculators, disputes, and preemptive registrations; those problems in turn create a political demand for rules and protections governing entry into the name space: 'sunrise' policies, exclusions, gradual and slow expansion, and so on. Stability is the new regime's leitmotif, and in practice, stability means all change is guilty until proven innocent.
This is not a deliberate result. No one wanted it to happen this way. That is precisely what makes it interesting from the standpoint of institutional economics. A form of 'positive feedback' led to the formation and entrenchment of an inefficient regime, just as North described. It is the product of social processes locked into a dysfunctional pattern by a kind of recursive political logic that no one knows how to break out of.
[1 ]'The path of institutional change is shaped by (1) the lock-in that comes from the symbiotic relationship between institutions and the organizations that have evolved as a consequence of the incentive structure provided by those institutions, and (2) the feedback process by which human beings perceive and react to changes in the opportunity set' (North 1990, 7). I consider this aspect of North's theory to be suggestive and interesting but not very well articulated and badly in need of further development and testing.
[2 ]The vast majority of cybersquatting cases involve domain name registrations that are not being used, and hence cannot confuse or deceive customers.
[3 ]For more than a year after the new domains were authorized, ICANN was still presenting them to the public as a 'proof of concept,' an experiment, a step into unknown territory. This despite the fact that over 100 country code TLDs had been added to the root since 1994 and that there is nothing unknown or experimental about the process of adding a new name to a name server's zone files (it happens thousands of times a day in the .com zone).
|< Day Day Up >|| |