The FCC to Unbundle Fiber?

FCC_New_LogoChairman Wheeler at the FCC announced last week that he would be bringing two proposals to the FCC meeting on November 21 associated with the IP Transition. The first involves some rules that will insure that 911 continues to function on an IP network and there ought to be no controversy with that. But his second idea is going to be very controversial, which is to give competitors access to RBOC fiber networks in the same manner that they have access today to the copper network.

The Chairman says that he doesn’t want customers, particularly business voice customers, to lose competitive options – and he believes that the unbundled network elements that are in place for copper today have brought competition to that market.

Let me step back and look at this idea at the big picture level. What the Chairman is proposing is a form of arbitrage. In general, telecom arbitrage comes when regulators force an artificial price on a product or service. In this specific case, the arbitrage would come from having the FCC or state commissions define the price, terms and conditions for a competitor to gain access to a fiber network. Arbitration is not necessarily good or bad, but if the price is set too low then there is an larger demand for the product than ought to be expected.

The industry does not have a very good history over the last two decades of dealing with arbitrage and the last mile network. There have been three times when FCC-administered arbitrage turned out bad for a lot of the industry and the public. First was unbundled network elements on copper that the Chairman is now acknowledging – the primary one being the unbundled T1. This was incredibly popular in the late 90’s and dozens of huge CLECs were funded to compete in this business. I had an office then in a business high-rise near the DC beltway and I remember a dozen different CLECs knocked on my door trying to sell a bundled T1/data connection.

After that came UNE-P. This was a creation that was a virtual unbundling of the network. With UNE-P a competitor didn’t have to collocate to get access to the RBOC copper. Instead they just bought all of the UNE elements and reconstructed a network. Finally, there was resale which forced the RBOCs to give set discounts on many retail products. Both UNE-P and resale were mostly used to compete for residential customers and some giant companies grew in the space. I remember Talk America, for example, that had well over a million residential customers on resale.

But for the most part all of the companies that leaped into these arbitrage situations failed. I remember well over a dozen UNE CLECs that went public with a few dozen more hoping to do so. Heck, the telecom industry was so juiced in the late 90s that there were even several telecom consulting firms that worked for the large CLECs who tried to go public. But in the end, the arbitrage opportunity became less attractive, as always happens, and all of these companies crashed and burned. The same thing happened with UNE-P and resale and all of the companies that tried to make a business using these arbitrage opportunities ultimately failed.

Arbitrage is rarely permanent and this makes it almost impossible to build a business plan to take long-term advantage of an arbitrage opportunity. The main reason for this is that the RBOCs are really good at resisting and challenging arbitrage. They file law suits and lobby and within 5-7 years after the start of an arbitrage situation they largely get it killed, or at least weakened to the point of being useless for a competitor.

Now we are looking at a new arbitrage opportunity of allowing competitors to get access to fiber networks. I have dozens of questions about how this might work, because it’s not as obvious on a physical basis how one unbundles a fiber network in the same way that has been done for copper. With copper, in essence, the copper line from a customer is physically redirected to a CLEC. But that is not going to easily work for a fiber connection.

How big this opportunity might be depends upon how the FCC implements it. For example, if they only allow fiber interconnection in places where there had once been a copper UNE connection then this is going to be very limited in scope. But it’s hard to see how they can stop there. After all, CLECs that compete using RBOC copper have always been allowed to grow, and if a competitor can’t ever add a new customer then this form of competition will be nearly worthless.

But if all of the fiber in the RBOC network comes available to competitors, then we are looking at the possibility of a whole new major push of competition. Competitors have largely been kept from the RBOC fiber network and this opens up huge market possibilities.

My advice to my clients is going to be to be cautious about leaping into this kind of opportunity. History has shown us that AT&T and Verizon will be working to kill this kind of arbitrage from the minute that it’s proposed – and so it’s likely that this will only remain lucrative for a few years before those companies squeeze the ability to use unbundled fiber.

Don’t get me wrong. As a consultant this opens up all sorts of new work for me. But having lived through the last arbitrage trials in the industry, my alarm bells are already going off and I am going to be advising caution. If the FCC tilts the arbitrage opportunity enough in favor of the competitor then there is going to be money to be made, but I will be reminding everybody that whatever the FCC giveth they can also someday take away.

3 thoughts on “The FCC to Unbundle Fiber?

  1. This huge for rural America. It opens the middle mile to competition. In Colorado, for example, CenturyLink fiber runs to the central offices in almost all (save for San Juan county) county seats. A T1 can be $1,000/month in those communities (resellers at $400/month). Many of our 178 school districts pay $100+/month/Mbps for bandwidth. Open those fiber networks (144 strands carrying up to 1 terabit per second) to arbitrage and we should see a crash in the cost/Mbps/month in bandwidth.

    Memo to CenturyLink: Kindly do not whine about “investment protection”. That fiber was paid for by CO taxpayers to the tune of almost $40 million in the early 2000’s followed by $140 million a year in federal and state subsidies.

    Like

    • We’ll have to wait for the details to se if that’s how it will work. The FCC tried implementing a dark fiber UNE a number of years ago where competitors could gain access to long lengths of dark fibers between central offices. But after a few years the RBOCs where able to get those rules changed such that the only places that would allow competition were essentially those places where nobody wanted to compete. Any town that had three of more registered competitors (whether they competed or not) made that town off limits to the dark fiber UNE. And as you can imagine, almost none of this is sold to competitors although its still on the books. That kind of dark fiber could be used to bring terabit speeds to an area, just as you suggest.

      We’ll also have to see if they unbundle the middle mile, such as you are suggesting, or the last mile, or only selected parts of the last mile. The FCC announcement wasn’t detailed and I’m not even sure we’ll know much more about it on the 21st. But you can bet that the RBOCs already know all of the details.

      Like

      • Thanks Dennis!

        “Never waste a crisis”: propaganda opportunity – “net neutrality through unbundled fiber! If Comcast had to worry about losing tens of millions of subs to a competitor they probably would think real long and hard about throttling my Netflix.

        Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s