Fast Net News, DSL Prime, DOCSIS Report

The government can either spend tens of billions to duplicate the existing fiber or use the "special access" rules very narrowly only where there is proven high costs. Forget arguing the rights or wrongs of this; it's impossible to find tens of billions in the budget after the stimulus is used up. Anything less would have little impact. Massachusetts has asked for $110M for an area there are only 3,000 unserved homes. Verizon has fiber in place nearly throughout the Massachusetts territory, so it's well served. Verizon wants a monopoly-like price, which is why the state government wants to build an alternative.

Building a duplicate fiber network, for better or worse, is realistically not a possible choice going forward. Either the broadband plan gives up or moves to special access, regulations for when there aren't enough competitive suppliers. The Bush administration, hating government rules, cut back on special access as much as possible. It's mostly used for access to city office buildings and commercial service, where the money and attention focuses. But a senior FCC staffer tells me the law clearly applies, if three Commissioners decide it's the right thing to do. There's an open proceeding with a complete record that can move almost immediately. Jonathan Adelstein, now at RUS, tells me he voted for similar as an FCC Commissioner.

The "unserved" are only 3-6% of U.S. homes, almost all rural. Presidant Obama has made serving them a national priority, with support from both Republicans and Democrats in Congress. There are another 3-6% in rural areas that can only get inferior service. That's only a very small part of the U.S. market and the telco income for backhaul. The FCC is deciding on possibly ten billion dollars or more in the urban special access proceeding. The rural part is a fraction of that, especially if it is narrowly tailored only to the places where the $100 or $200 prices provide strong evidence of market failure.

Hundreds of applicants thought the stimulus should pay for them to build unneeded fiber and sent up a roar of "gimme, gimme." The biggest carriers boycotted the first round, but now they are negotiating a way into round two. The noise overwhelmed the limited Washington attention span; I can't recall a single  until the FCC workshops, everyone was thinking of what to do with federal billions and the "special access" rules were just not on the radar screen.

There is now no choice, since the money for subsidies isn't available.

1) Apply this very narrowly, only where there is very limited competition and very high backhaul rates. Bringing down the most extreme prices should be the first goal. The total dollars are significant compared to $500M to $2B in high cost USF, for example, but much less than the urban commercial access battle. That simplifies the politics.

2) Many of the rural carriers simply do not have this problem. They are close to an urban market or a fiber point of presence, are part of an efficient state coop, or simply businessmen who found a good deal. I know several New England companies that worked together on a fiber connection to the Boston market. I believe their bandwidth costs are under $25/megabit. At least one-third of the rural carriers do not have a major problem with bandwidth costs. There's no data I can find. The number with high bandwidth costs is somewhere between 30% and 70%.

3) Only use special access when a high price clearly indicates a problem. I'd set a trigger (?50/megabit) below which it wouldn't be considered. I'd move rapidly on anything above $100. In all cases, special conditions - like a requirement to run new fiber - would be a reason to deny special access. A obviously competitive market, say 4 choices, would also rule out special access.

IANAL, but it would seem a good way to codify that is to create a rebuttable presumption of market failure if the local price is more than ?three or four times the national average. Unless new construction is required, the actual carrying cost of bandwidth is modest. In most of the country, the fiber has been in place for 10 or 20 years, mostly depreciated, and has plenty of capacity. So the natural price in most rural areas would be only modestly above the urban price.  I can think of many possible exceptions, which is why the system should allow for a showing of truly high costs. That would be appropriate, for example, if fiber needed to be run to a previously unserved area.

The result: government intervenes first and possibly only in the most egregious cases. If there were an understood guideline, I bet many would cut a deal near that guideline to avoid government involvement. Set a procedure that works quickly but allows the parties to reach a compromise, save the cost of lawyers, and may allow the government not to get involved at all. If the guidelines or early rulings come to about $40/megabit, for example, many companies will settle for something like that to avoid uncertainty and lawyer costs. Bringing a $200 price down to $55 will make an enormous difference for the price/availability of broadband, but leaves a generous profit for the backhaul carrier. This is about getting rid of ridiculous prices, not drawing blood. I leave it to the lawyers to think this out.

4) Be very generous in the special access pricing. It's usually on a system called TELRIC, which includes a profit. The difference between a market or TELRIC price and the actual price for some of these rural carriers is massive. That means a generous formula, perhaps TELRIC plus 40%, solves much of the problem while leaving a large (but not monopoly) profit for the telco with fiber.

5) A huge fraction of the unserved - estimates range from 20% to 50% - can get cable TV but not cable data, even if the cost of upgrading the cable system would be recovered quickly. ($100-500/home) Most of the D.C. focus has not been on this group, but they are by far the least expensive to reach. For many of the small rural cablecos, the excessive middle mile charges are the reason they don't aren't offering broadband. If I'm correct, this would reach many of the "unserved" - maybe 1 in 5 - without requiring a direct government subsidy at all.

6) Think like a businessman running an ISP rather than a lawyer and the choices become much clearer. The economics of the ISP business imply a $30-60 price is a problem, a $100 price can be killing, and a $200 price requires massively higher broadband prices, inferior service, or some subsidy somewhere.  In practice, many of the rural carriers charge slightly more than the urban $30 rate for a megabit of service and wildly more, perhaps $60, for the same speed as $30 buys you in most of the country. Others absorb something on the bandwidth to keep the voice customer, especially where cable competition is strong.

The takeaway is that if you help the most likely gouged ($150 and above) you'll have a big impact. If you get something moving that gets many of the $75-150 down to $30-50, that will also have important results. Of course you'd rather have prices that match the urban areas, but going after the extremes first make politics and legal issues easier.