Wednesday, May 7, 2014

Would Verizon Recover its Cost of Capital if it Built Out FiOS Across the Rest of its Footprint?

Verizon has capped its FiOS deployments to about 19 million homes passed, enough network to reach about 70 percent of locations served by Verizon’s fixed network. Obviously, that means 30 percent of the network never will be upgraded for FiOS.

That is certain to raise hackles in some quarters. But a reasonable person might conclude that the additional investment, at current costs, revenues and adoption rates, might not recover its costs of capital.

Other companies, without Verizon's legacy cost structure, might have an easier time of it. But few would likely be able to compete against both Verizon, a cable competitor, the satellite video providers and any other ISPs that decided to enter specfic markets.

Is that a gamble? Certainly. It means Verizon will continue to lose market share to cable TV and possibly other ISPs over time, as Verizon is unable to offer equivalent speeds as its key competitors, and also is unable to compete fully in the video subscription business, at least using its fixed network.

But Verizon is betting that capital invested elsewhere, in mobile assets and services, and possibly at some point in acquisitions, will produce a higher financial return than building FiOS to reach another 20 percent or so of its installed base of fixed network customers.

The cost and revenue implications of competitive markets are the reason for the caution. Any extension of FiOS would lead to stranded assets that could represent as much as half of the access network investment.

The reason is simply that FiOS is unlikely to attain long-term penetration rates in excess of much more than 40 percent, either for Internet access or video services, where it operates.

FiOS Internet penetration was 39.5 percent at the end of fourth-quarter 2013, meaning that Verizon was able to sell a high speed connection to about four homes out of 10 it passes.

FiOS video penetration was 35 percent. In other words, Verizon also could sell a subscription video service to 3.5 out of every 10 homes it passes. For the most part, FiOS customers buy two or three services, with triple-play packages seemingly most popular.

The FiOS network passed 18.6 million premises by year-end 2013. Perhaps 68 percent of FiOS customers buy a triple-play service. Most of the rest likely buy a two-product bundle of Internet access and video.

That might imply FiOS overall penetration of about 50 percent (assuming 90 percent of FiOS customers buy a dual-play or triple-play service) while about 10 percent of households only buy a single service.

For the sake of argument, assume Verizon gets a long-term, sustainable penetration rate of 50 percent. On any new FiOS builds, that implies, at a network cost of $750 per home, a network cost of $1,500 per customer, plus about $600 to install a drop.

In a typical 100-home neighborhood, that suggests network investment of about $75,000 and drop install costs of about $30,000, for a combined per-customer cost of about $105,000.

Assume that 70 percent of the FiOS customer homes generate about $150 a month in revenue, some 20 percent generate $100 a month and about 10 percent generate $50 a month worth of revenue.

That works out to about $63,000 in annual triple-play revenue; $12,000 in dual-play revenue and about $3,000 in annual single-play revenue, for total gross revenue of $78,000.

Assume operating cost of about $46,800 (assuming gross margin is about 40 percent). That would suggest net revenues (before dividends) of about $31,200.

Assuming half of net revenues has to be reserved for dividend payments, That might imply just $15,600 in profits from that 100-home neighborhood. Even if Verizon had no interest payments, it might take nearly seven years to reach breakeven.

Over a 10-year period, that further implies profits of about $4,680 or perhaps four percent annually. The issue is whether that actually covers Verizon’s cost of capital. If actual 10-year profits are anywhere close to this simple analysis, it isn’t clear the investment makes sense.

No comments:

"Tokens" are the New "FLOPS," "MIPS" or "Gbps"

Modern computing has some virtually-universal reference metrics. For Gemini 1.5 and other large language models, tokens are a basic measure...