TCS Daily

Give Me My Twisted Pair or Give Me Death!

By Andy Kessler - November 27, 2001 12:00 AM

In 1904, the United States took on the task of building the Panama Canal, after the French company that started the project gave up, having lost thousands of workers to malaria and yellow fever. Mosquitoes would suck infected blood from monkeys in the forest, fly to the dig zone, bite a worker, and pass on the infection. Can-do Americans contemplated their options. Get rid of the monkeys? Kill all the mosquitoes? Not likely. Finally, someone figured out that the mosquitoes could not fly more than a mile without eating. The forest was cut back a mile on each side of the canal. Disease declined drastically. Welcome to out-of-the-box thinking.

In the aftermath of the NASDAQ debacle, a lot of telecom investors and pundits are throwing in the towel, saying risk capital just doesn't work. The Internet backbone is over-stuffed with unused capacity; meanwhile the "last mile" to potentially paying customers' houses sits mainly unused. "Unleash" the phone companies, as some are saying? Fuggedaboudit-high-speed DSL at $50 month means that the real value of voice calls is $1 a month, 1/30th of what the telcos currently make. Broadband would-and some day will-put the phone companies on their death beds, so they delay. The regulated structure of the telecom business is a deformity mere capital can't cure. The only solution is to get rid of the structural cause of the disease.

I used to think that anything that touched a private residence needed to be regulated, to avoid duplication. One water pipe, one electric line, one sewer, one gas line, one TV cable. No need for competing companies, digging up your lawn or street. Same with airwaves-divvy up the spectrum, three TV stations and two cellular phone companies. This monopoly mandate is the status quo. Man, was I ever wrong.

Look at cellular. Today, there are six nationwide carriers, and four or five competing companies in any given area. But instead of four or five ugly towers, there's usually just one. The industry has organized itself horizontally, quite naturally. Companies like American Towers and Spectrasite Holdings own the towers, and lease them to the operators.

The Telecom Act of 1996 promised competition between phone companies and cable, between local and long distance players. It failed, merely creating new regulatory umbrellas for everyone to price under. Actually, the 1996 act didn't just fail-it created chaos, as Wall Street's spreadsheets screamed "huge returns" to anyone laying fiber under those huge price umbrellas.

The whole bubble could have been avoided with a modest knowledge of high-tech economics and a bit of free market thinking. In November 1997, I wrote a Forbes article about Worldcom's purchase of an amalgamation of local and long distance telephony and internet access, MFS/LDDS/UUNET. "Every guy with a Ditch-Witch and a right of way is laying fiber," I wrote. "The only way Worldcom can dry that supply up is by driving prices lower." I mean this is Economics 101-even people that slept through the lectures understood something this simple.

The following spring, I attended an investor dinner hosted by DLJ at the Four Seasons restaurant in Manhattan. (This was back when Wall Street was still making money-hand-over fist.) The guest speaker was none other than John Sidgmore, the squirrelly looking founder and CEO of UUNET and at that point vice chairman of Worldcom. I shook his hand at the bar, introduced myself, and smugly suggested that I knew exactly what Worldcom was going to do: cut prices to the marginal cost of laying new fiber, the whole Ditch-Witch thing. He looked at me like I was from another planet-Silicon Valley was pretty far away-and with a wave of his hand more or less told me to get lost.

At his talk, he used the sure-fire wisecrack that I have since heard a million times: "Bill Gates says bandwidth should be free. Well I say software should be free!" Applause, laughter.

So Worldcom and its other bloated telco friends milked the high margins allowed by regulated RBOCs, keeping prices high while technology-driven costs plunged 30 percent to 50 percent annually. But those same high margins also attracted aggressive new players fuelled by joyous Wall Street capital, Level 3, Qwest, 360 Networks, ICG-massive new supply, in other words, which ultimately broke the pricing curves, so they are now selling bandwidth below sunk costs.

Bridging the "last mile" (the actual distance can be a lot less) from all that over-stuffed fiber-optic backbone to people's homes was problematic, at least in 1996. A standard estimate of the cost was $2,500 per household; times 100 million homes equals a quarter of a trillion dollars-Capitol Hill kind of money. (Had the FCC ever heard of the stock market?) So instead of pushing for something new, the 1996 act mandated that the phone companies open up their existing copper wires to all takers. They did but at a pace best described as slothy.

Why? Because broadband-with its vanishing per bit costs-undermines their lucrative voice business. So they overcharged their guest-competitors for space and electricity. They sent lawyers into action. On a tour of one phone company central office, a three-foot pile of papers sat on a desk-requests for space and facilities. The tour guide winked and said they will get to that pile over the next few years. SBC has paid $48 million in fines since July 2000 for dragging its feet. Cheap revenge.

By 2000, the phone companies had enough-they raised wholesale prices for their facilities, and lowered retail prices, squeezing the debt-laden competition to death. With most players stiffed, they have raised retail prices, and why not?

So what is the fix? Two solutions are widely offered, and-like killing the mosquitoes or the monkeys-neither is palatable.

Wrong Answer #1: Unleash the RBOCs

One school of thought says unleash all the players. Complete deregulation, everyone is just free to compete. The problem here is obvious: very few new players will enter the fray because no one else but the incumbents has the facilities or the money to make even the smallest dent.

Worse, the incumbents, the Baby Bells, are monopolists. When you own a government-mandated monopoly, your programming models always output "do nothing" to maximize returns in response to innovation or change. With a lot of investment, they could indeed offer voice and data and audio and video over that same twisted copper wire, but they couldn't charge much more. At $50 per month, 640K DSL service implies a $1 per month cost for voice. The gap is so wide, in the RBOCs minds, it can never be closed.

Indeed, for the telcos, broadband investment increases costs without any revenue increase, so the spreadsheets scream "Why bother?" "Unleash" them, as the Tauzin-Dingell Internet Freedom and Broadband Deployment Act of 2001 suggests, and they will milk voice for another decade.

Wrong Answer #2: Structural Separation

The other school of thought has been pushed by the trendy Pennsylvania Public Utility Commission. In this scenario, you forcibly separate the telecom companies' wholesale and retail businesses. At first I was lured by the simplicity: you remove conflicts of interest and build in the efficiency of a horizontal business model. But it's flawed. The wholesale entity is still a monopoly, and therefore almost certainly required by law to price below replacement cost. Which means no new build-out of better lines or fiber to homes, by them or anyone else. Next.

Right answer: Out of the box, into your hands

So how do we cut back the forest, and make the last mile a workable proposition? How do we induce a build-out and innovation, with the promise of a return on investment for new capital?

Here's my suggestion. Give the last mile to me. Well not just me-I mean me and you and every homeowner. Declare that the homeowner owns the connection between a home and a designated termination point-the existing telco central office, a neighborhood gateway, a digital loop carrier. As in the cellular business, horizontal players can operate the termination points, others the phone poles, others the residential gateways. Companies can compete at the retail level for voice services. New players will emerge to upgrade last-mile connections, by enhancing the existing copper wires, running fiber, even going wireless. One enterprising Australian rigged up a laser pointer to transmit data a couple of miles at 10 megabits per second. That would look cool as the fog rolls in the Bay Area.

Only an asset grab from the phone companies will demonstrate the value of horizontal vs. vertical markets, and unleash innovation. And the famous quality-of-service question-when your house catches fire on a Sunday night in the middle of a blizzard, how do you dial 911? Use your cell phone.

Innovation typically comes from entrepreneurs, armed with capital seeking high returns. The worst form of capital comes from monopoly ratepayers, whose returns are guaranteed. Risky investment for monopolists is unnecessary and wasteful.

Is there any legal precedent? The Homestead Act of 1862, which gave away 160 acres of government land (and a regulator, er, mule) if you upgraded it. And even the FCC has what is known as Pioneers Preference, which gives out new spectrum at previously unusable frequencies to those that develop it first. And let's throw in reverse eminent domain, just for the hell of it.

One thing is for sure, we have paid for those dinky copper wires to our homes again and again-every five years is a good estimate. But is there any other justification? Sure, Metcalfe's law. Bob Metcalfe, the inventor of Ethernet data networking, famously noted that the value of a network grows as the exponential of the number of nodes on a network. Put another way, if there was only one phone, it would have no value; so by adding yours to the phone network, you increase its value. You add the value, not your phone company.

Instantly, new companies will surface that offer new services. Are you a heavy user and want a fiber optic line to your home? No problem-either put up the $2,500 yourself or let someone else do it in exchange for a five-year contract for data services. Cutting back the forest-or rather, turning it over to consumers-turns the whole industry upside down, and puts consumers, not regulators, in charge of upgrading the system.

The same idea can work for wireless. Despite six nationwide carriers with their own licensed spectrum, they all charge more or less the same rates for basic voice service and exorbitant rates for limited mobile Internet access. Yet unlicensed bandwidth can be shared by using packet technology. Many, many users share bandwidth simultaneously without stepping on each other, and new products like Wi-Fi are growing by consumer installations, away from carrier rollouts.

And cable? Absolutely. Local TV stations, yes, satellite, yes-just about any regulated communications system can be put back into the hands of users, creating naturally competitive market places.

The last mile of the telecosm may not seem as pressing as improving cruise missile guidance systems, but is there any doubt that he who owns the means of production of smart weaponry owns the world? High-tech fighters and missiles used to be the tail that wagged the dog; now they're a byproduct of the digital economy. To stay ahead requires an infrastructure for innovation-56K modems and even DSL don't cut it. Pushing the state of the world requires state of the art telecom.

Andy Kessler is a former hedge fund manager who is writing a book about predicting the past and studying the future

TCS Daily Archives