CAF Phase II Cost Model Presented
Industry and agency officials gathered at the FCC Thursday for a detailed explanation of the proposed Connect America Fund Phase II cost model, presented by CostQuest Associates. The goal of the CostQuest broadband analysis tool (CQBAT) is to identify the high-cost portions of broadband buildout throughout the country, defined by the ABC Coalition as anything over $80. A map presented at the meeting showed most of the high-cost areas exist in the western half of the country, and some loops can even cost in excess of a million dollars. This can happen when a dedicated plant is required to serve a single customer, said CostQuest President James Stegeman.
Sign up for a free preview to unlock the rest of this article
Communications Daily is required reading for senior executives at top telecom corporations, law firms, lobbying organizations, associations and government agencies (including the FCC). Join them today!
Several policy decisions must be taken into account when developing a proper USF model, Stegeman said. The model must determine what technologies will be used, whether a network is fiber-to-the-premises or fiber-to-a-DSLAM, what the take rate is of customers who purchase the broadband service, and what geographic unit of consideration to use. Cost modeling issues include when to use forward-looking vs. embedded costs, and the source of input data. Designers must also make assumptions regarding how the plant mix is determined, whether capital expenditure cost should be adjusted to capture purchasing power, and what lives of assets should be used.
The model takes pains to use the best data to determine which areas have broadband coverage and which don’t, using Census information coupled with alternative sources of data, Stegeman said. Based on the price and coverage, CQBAT uses data from Warren Communications News as an alternative source of cable competition data, with “tweaks” to try to make everything as accurate as possible, Stegeman said. “It’s really indeterminate what is the best public source for data coverage, because they are all different and -- if you're a third-party provider here -- none of them are really right,” Stegeman said. “They're best guesses of what the cable coverage is."
A “fun tidbit” is that “this really doesn’t matter in the end,” Stegeman said. “When you set a benchmark at $80, it typically already counts for where cable is. Cable is not going to build where it costs more than $80 per customer -- they can’t make money. So by the use of a benchmark, you naturally filter out competitive areas."
Speakers also discussed Alaska Communications’ proposed model, intended to take into account the different costs of serving remote and insular areas such as Alaska. “ACS believes that Alaska-specific costs have not been captured, and that current modeling underestimates support for ACS LECs,” said an ACS presentation (http://xrl.us/bnpnjc). The telco said its model takes into account costs it said were not captured by the CQBAT model, such as increased middle mile transport, undersea fiber optic cable connections and the above-average costs for site maintenance.