December 12, 1998

Re:	Federal-State Joint Board on Universal Service; Forward-Looking Mechanism for high Cost Support for Non-Rural LECs ; (CC Docket Nos. 96-45, 97-160) .

Today, the Common Carrier Bureau releases an Order denying GTE's Emergency Motion for Disclosure of Data and Information to Permit Public Review and Extension of Time. I object to the denial of GTE's reasonable request that this agency provide all information necessary to allow outside parties to fully evaluate the model platform that we have already selected. Without such information, it is impossible for GTE to determine whether or not the model that the Commission has selected is operating consistent with the findings and conclusions contained in the Commission's Order.

I have repeatedly objected to the Commission's adoption of any federal model. I agree with GTE's conclusion that "the Model platform is inherently ill-suited to estimate accurately the costs of an efficient local exchange carrier to provide universal service."⁽¹⁾ In addition, however, I have repeatedly objected to the adoption of a "black box" distribution scheme that is not transparent to all concerned parties. Such a process will lead to complete control of universal service distribution at a federal agency with little or no recourse to outside parties wanting to challenge those decisions.

It is inequitable to refuse to make available the complete federal model platform that this agency has already adopted. While I appreciate that the components may be available on the website, that is not a sufficient substitute for making the model fully available to all parties who are interested. Requiring parties to reconstruct the models from its various components is not just burdensome, it may also result in unanticipated variations. Why should be it be GTE's responsibility to re-invent the Commission's "public" model? I fail to see why the Commission has failed to make at least the platform -- if not the default geocode date and a standard set of inputs that can be run on all of the models -- available to the public. Moreover, why can't we provide some sample runs of how this model will work on a state by state basis with the common inputs that the Commission used to evaluate the two submitted models last spring? Indeed, I would like the same information myself!

In addition, GTE asserts that "documentation has not been provided that fully explains the engineering assumptions and standards that are the basis of the FCC model."⁽²⁾ As an economist, I have spent much of my professional life building and evaluating economic models. Without such information, I agree that it would be difficult if not impossible to determine whether the model complies with engineering design standards. Similarly, the failure to explain how the modules interface would make it impossible to assess whether the assumptions and algorithms are consistent from module to module.⁽³⁾ In light of these concerns, why has the Bureau failed to make the entire model platform fully available? Indeed, I believe that we must make it available.

As I have indicated previously, the regulation of markets through models is inherently inequitable. If no one outside the agency fully understands the model, then there is no easy way to appeal the model results. Few people understand how any model works; fewer still understand complex models; no one really understand models that produce no results. Models entail a certain degree of opaqueness in contrast to the transparency of markets and market outcomes, or even in contrast to simpler rules based on accounting information or even simpler information.

This opaqueness leads to inherent uncertainty and instability. The results of a model at one point in time can be arbitrarily helpful or harmful to one individual. The key word is "arbitrarily." Models can easily be changed, and often are. An apparent benefit one day can become a liability tomorrow. The net effect is unpredictability and uncertainty. No economic cost model can meet the "specific and predictable" standard required of the federal universal service support mechanisms by the Act.⁽⁴⁾

Regulation based on accounting rules or even simpler rules may provide -- at any moment in time -- a less accurate portrayal of a competitive market than an economic cost model potentially could. Simpler rules such as accounting rules, however, are far more transparent to the world, are not easily corrupted, are easily appealed, and provide a greater degree of market certainty and stability.

It is not surprising, and even predictable, that within months of adopting a model we already have parties complaining that they do not have sufficient access to information to evaluate the results and the methodology selected by the Commission. I continue to think that such a result is unfair. We should provide the model platform that we have selected and other necessary information to the world and let it be evaluated openly.