Monday, August 3, 2009

Telecom's Third Rail

There's a thoroughly uncomfortable question nobody in the telecom business wants to touch. Most people have heard the analogy that communications networks are fundamental underpinnings to future economic success, much as roads, railways or airline routes might be.

Most people are at least casually familiar with the plight of the newspaper business, which seemingly has yet to find a way to remain financially viable in a market that offers many substitutes.

Now there are the first projections that the advertising business, which except for normal economic fluctuations has grown for 30 years or more, might now face a future where brand spending actually decreases over time, in a structural, not secular shift.

So the uncomfortable question is whether the telecom business is in fact becoming a sort of infrastructure, like roads. And what I mean by that is that the business model for roads is largely indirect. True, there are some toll roads, but most roads, though an input to other revenue-generating enterprises, do not make money: they lose it.

So the "roads" or infrastructure analogy should be troublesome in the extreme. It at least implies an industry that has the profit sucked out of it, that is foundational and important, but not profitable in the way it has been.

At some point, should that continue, the industry as we now know it could not continue to exist. There would be a need for big pipes to virtually all locations. We need roads. But all the other economic activity would then be created by industries that support people driving cars.

This is something here more than the fear of being reduced to "dumb pipe" providers. Many businesses operate as low-margin "commodity" affairs, especially when they have large scale.

The newspaper analogy is more unnerving. That analogy suggests that communication networks could become something more akin to "roads," where there actually is no viable business model, and the infrastructure is a societal "cost" borne by taxpayers.

Perhaps you think governments globally have enough extra headroom to increase taxes to do such a thing. If there is no viable business model, that will be your only choice.

Being an optimist, I suspect the analogy is imperfect. I think executives will show enough creativity to avoid the worst case, and that regulators will be prudent.

What seems less debatable is the risk that if enough profit is drained out of the telecom business, even a robust "pipes" business might be tough to sustain.

That's an insight regulators in many countries have learned they must grapple with. Let us hope sane heads will prevail

2 comments:

Patrick Murphy said...

Interesting comment Gary. I would like to see you tease out the anlaogy in a longer piece.
A primary market difference between newspapers and telcos is the regulatory environment. We all recognize for better or worse the major incumbent telcos do have a preferred regulatory status across the globe.

That seems to be the primary obstacle to getting cannibalized by a new competitor.

I would like to read more from you on the subject.

Gary Kim said...

Agreed. But it is precisely the regulators I now worry about. The bit pipe business is evergreen and essential, no doubt.

What I don't yet know is whether U.S. and other regulators will attempt to commoditize transport and access, which is trending that way in any case, without recognizing that if the pipe business gets commoditized too much, investment capital will dry up, and so will some amount of innovation.

Much hinges on whether access and transport are natural monopolies or not. But it's complicated. Parts of the business clearly are not "natural monopolies" (wireless, video, voice, other services).

Fiber-to-home might or might not be. Wired networks in rural areas likely are very close.

Natural monopolies are not known for innovation, though they tend to be operated in ways that emphasize availability (power, gas, water, sewers, roads, rail lines).

Regulators are looking more closely at separated models where access is a monopoly, but services are not.

In most U.S. markets there are now two wired access networks, with multiple wireless networks, so the natural monopoly argument seems somewhat weak.

Observers will argue about whether oligopolies are much different from monopolies, but the point is that a true natural monopoly sometimes must be heavily regulated.

The danger is that if too much profit gets sucked out of the business, we might get stuck with a highly-regulated access business with low rates of return and all the other consequences that flow from that.

Some will argue that's fine. An access utility might be fine if all the apps are providers by third parties with access to the pipe. But that might also mean some "pay for delivered bits" financial model.

If one assumes all "services" become "apps," then app providers will have to pay for use of the pipe. It is terribly complex in terms of all the shifts in business models.

Many Winners and Losers from Generative AI

Perhaps there is no contradiction between low historical total factor annual productivity gains and high expected generative artificial inte...