Local loop unbundling for cable

by on February 23, 2014 at 11:48 am in Economics, Law | Permalink

Felix Salmon endorses local loop unbundling for cable, so does Kevin Drum.

My earlier analysis simply was assuming that we will not make this policy shift and then asking how worried we should be about the resulting semi-monopoly power in that market.  If you would like to see the pro-case, here is a UK study (pdf) showing unbundling improves quality.  Here is French evidence for higher penetration, often through quality rather than just price effects.  Here is Tom Hazlett on related issues (pdf) and Vernon Smith is a long-time proponent of related ideas.

I don’t, however, agree with Felix’s presumption that all we need do is refine the current infrastructure, or his claim that there are no other effective forms of competition at current margins.  Penetration rates could be a few percentage points higher, and that is an economic cost from the status quo, but in Felix and some of the other commentators I am seeing a black and white version of a monopoly story that simply does not correspond to the facts.  Furthermore the current monopoly power of cable means that infrastructure will be laid down more quickly next time around, and moving to local loop unbundling would weaken this incentive by confiscating some of the rents from the infrastructure investments of the cable companies.  I probably would make this trade-off, but that further blunts any estimate of the net costs from the U.S. status quo.

Note also that Netflix has turned out to be worth a lot of money as a company, a reality which those who pushed the “cable as extreme monopoly” view denied could happen, out of a belief the cable companies would simply confiscate any Netflix rents.

And here is Peter Huber on how deregulation — yes the dreaded “D word” — can improve cable competitiveness.

Z February 23, 2014 at 12:26 pm

A couple of things:

1) The bottleneck is in the deals made between cable of the content makers. That incestuous relationship is the root of the problem. The cable monopoly operates as a go between much in the same way the insurer stands between the patient and the doctor. if you force a la carte pricing on the middle man, he is going to force it upstream.

2) Netflix is profitable largely because they can transfer most of their costs to the ISP. If I can transfer 80% of my costs to the ISP, my company would be worth a lot more too. Now that the ISP can start sending Hulu and Netflix a bill for services, those business models will take a beating.

ummm February 23, 2014 at 1:05 pm

Would not be surprised if google buys out netflix. Would work really well with Google’s fiber internet.

ac February 23, 2014 at 1:19 pm

Which costs are Netflix pushing to the ISP? They are paying for bandwidth to send their data out to the internet; the costs to the ‘receiver’ ISP are ALREADY paid for by the subscribers. If the ISP starts sending bills to the content providers, why exactly are consumers paying anything to their internet provider?

Z February 23, 2014 at 2:05 pm

Netflix causes bandwidth consumption and collects $8 a month for it. The ISP is forced to spend money to upgrade their network because of this and pass the costs onto millions of people who are not customers of Netflix. It is just another example of socializing risk and privatizing profit.

av February 23, 2014 at 2:21 pm

In other words the ISP advertises X mbps for $Y/month and then when their customers attempt to use the advertised service it turns out the costs are too high. The response isn’t to demand money from the third party, it’s to charge people appropriately for what they use i.e. tiering and throttling. Otherwise the costs of people who torrent Bluray rips all day are socialized to Netflix customers.

Z February 23, 2014 at 3:00 pm

They are not demanding money from a third party. They are charging Netflix for access over and above what everyone else gets. The deadbeat culture of the Internet is finally coming to an end.

andrew' February 23, 2014 at 3:14 pm

Idk, Netflix might be why I don’t ditch cable yet. I think the cable comp is the deadbeat in many ways. Not least of which they can’t figure out how to rationalize a quality product. How did we make sports (ESPN) expensive, BTW?

ac February 23, 2014 at 3:33 pm

How is Netflix getting access over and above what anyone else gets? As far as I can tell, the cable companies are doing their best to make sure they get LESS than what others get, in regards to data transmission latencies and rate. This is akin to asking a successful store to contribute to the costs of a toll road.

If I am paying for X mbps, why does it matter if it comes from Netflix, downloading data for work, or what? That is what I am paying for.

Emil February 23, 2014 at 4:17 pm

ac: you are paying for peak bandwidth not average. It’s like saying that you are allowed to go 70 miles an hour on a highway which doesn’t mean that the highway could support everyone going 70 miles per hour at the same time all the time.

Jan February 23, 2014 at 5:51 pm

This is not how the cable companies market their internet service. I am paying to stream at X mb/s whenever I want to from wherever I want, so long as it’s not illegal. The cable companies will soon get much quicker about shutting down high-bandwidth services and extracting additional payments. This will hurt innovation and probably prevent some really cool stuff from getting to consumers because the developers can’t pay Comcast’s toll.

Emil February 24, 2014 at 3:40 am

Jan: actually if you go onto the comcast website they never say that they guarantee you bandwidth. What they do say, extremely upfront and clearly, is that you get UP TO a certain bandwidth… Ergo your perception is wrong

Jan February 24, 2014 at 6:27 am

You’re talking about fine print here. They don’t market it as “you can get X mb/s, but not at certain times of day and not if we disapprove of the completely legal content you’re perusing.”

Emil February 24, 2014 at 9:55 am

Err no, I’m not talking about the fine print. I’m talking about the most basic description of the offer. Up front on the comcast webpage. See e.g. here: http://www.comcast.com/internet-service.html. It very very clearly say Up to 50Mbps.

When you are wrong, you are wrong.

john personna February 23, 2014 at 5:15 pm

Netflix does not “cause consumption.” I do, as a customer, when I point my paid-for bandwidth at Netflix. I could point it somewhere else. That is my choice as bandwidth licensee.

Mark Thorson February 23, 2014 at 2:10 pm

Because they don’t have any choice.

Doug February 23, 2014 at 4:59 pm

There’s no reason Netflix has to consume such high bandwidth. The ISPs could forward deploy Netflix edge servers in their own networks. Think of the House of Cards premiere, essentially ~20GB of video probably used 1PB+ of backbone bandwidth. That’s the height of inefficiency. It would make sense to host the media inside the ISPs, and the most accessed media at a very local level. The ISPs could easily eat this cost and still be way ahead in terms of cost.

The only reason they don’t is because most ISPs are also content providers themselves. Going along with this plan would essentially be admitting to their investors that their business model is dead.

john personna February 23, 2014 at 5:19 pm

I kind of get what you are saying, but we play this bandwidth game from top to bottom. We all buy a nominal speed pipe. If we use it at rated speed too often, we are considered hogs. Whatever, but I don’t think double-charging, sender and receiver, is the solution.

Give the consumer a clean contract for a rated speed and if necessary a monthly cap and leave it at that. Don’t double-bill.

Dan Weber February 23, 2014 at 9:09 pm

Monthly cap? That’s the crudest way to handle the bandwidth hogs.

The IETF has researched and debated dozens of ways of dealing with congestion, so it’s kind of silly that the ISPs only know how to apply the hammer of a monthly cap.

john personna February 24, 2014 at 9:33 am

Fine. Whatever. My point is that as a subscriber I should know what I’ve bought. After that whether I point at Netflix, Hulu, or YouTube is up to me.

Dan Weber February 24, 2014 at 9:37 am

Not all bytes are equal. My VoIP session is relatively low-bandwidth but requires low-latency, while scp’ing my daily backups requires a lot of bandwidth but is very insensitive to temporary latency issues.

But, “fine, whatever.”

john personna February 24, 2014 at 11:40 am

Don’t you get what I’m saying with “I should know what I’ve bought?” When you say “Not all bytes are equal,” what are you saying? Something that should be in my service contract?

Jan February 23, 2014 at 12:36 pm

“Furthermore the current monopoly power of cable means that infrastructure will be laid down more quickly next time around…” Not sure this makes sense.

Anon. February 23, 2014 at 4:11 pm

I don’t understand that sentence either. The monopoly status is what caused the terrible state of the infrastructure in the first place: no competition means no incentive to spend money on infrastructure.

Doug February 23, 2014 at 4:51 pm

I believe what TC implied is that monopoly pricing makes the returns to infrastructure higher than under perfect competition. So next-gen infra will be laid sooner because companies won’t wait for tech costs to fall to the competitive equilibrium level. They’ll be incentized to invest earlier than that to capture monopoly rents over the lifetime of the infra.

This is the same idea behind granting patents to drug developers. Giving the early investors monopoly power can make sense in a market with high fixed costs and low marginal costs. For example pharmaceuticals or telecom.

Jan February 23, 2014 at 5:40 pm

But why would they want to improve infrastructure? The profit-raising incentive to do that is absent without the threat of competition.

Komori February 24, 2014 at 10:43 am

Demonstrably false, unfortunately. It took a company with Google’s level of money and clout to start building out a new infrastructure. Living in Austin, I’ve taken a rather darkly amused note of AT&T’s response. They’re going to roll out a (not very competitive) high-speed option, but will not go one house beyond where Google’s roll out will be.

fwiw February 23, 2014 at 12:36 pm

Bad day to say that cable companies won’t extract Netflix rents…

In today’s WSJ: “Netflix Agrees to Pay Comcast to End Traffic Jam”


Z February 23, 2014 at 2:08 pm

Netflix has made it clear that they cannot make a profit if they have to pay ISP’s for their bandwidth usage. Maybe that is just bluster, but there’s no way they can provide the bandwidth and content to user for $8 a month and make money. They will have to raise fees. Some people claim they will need to triple those fees, which makes Netflix far less attractive to users. If I were a betting man, I’d wager Netflix ends up the property of an ISP within five years.

john personna February 23, 2014 at 5:14 pm

Z, my cable company says I pay for 20+ Mbps right now. Why should Netflix pay again?

William Rinehart February 26, 2014 at 6:26 pm

Netflix already has been connecting with Comcast to serve the end user through an intermediary, and has been paying money for this. What they did with the deal is cut out the intermediary and sign an agreement for service, an SLA. In the long run, Netflix’s costs will probably decline. Every company that want’s users to have fast access to their products do these kinds of deals. Apple, Yahoo, and Google are just a few of the names. And Netflix is about to sign these deals with Verizon and AT&T.

Phill February 23, 2014 at 12:45 pm

> Note also that Netflix has turned out to be worth a lot of money as a company, a reality which those who pushed the “cable as extreme monopoly” view denied could happen, out of a belief the cable companies would simply confiscate any Netflix rents.

Well… until very recently they simply lacked the regulatory agency to do that. Which they are now free to do.

>Furthermore the current monopoly power of cable means that infrastructure will be laid down more quickly next time around, and moving to local loop unbundling would weaken this incentive by confiscating some of the rents from the infrastructure investments of the cable companies.

The time frame expressed here strikes me as odd. Why is this desirable? Isn’t this just basically a variant of techno utopianism – accept monopoly rents now to incentivize the even awesome technology just around the corner? What of the great stagnation? We already have perfectly adequate infrastructure.

To quote Salmon,

>What’s more, all of that effort is redundant and duplicative: we already have perfectly adequate pipes running into our homes, capable of delivering enough broadband for nearly everybody’s purposes. Creating a massive parallel national network of new pipes (or pCells, or whatever) is, frankly, a waste of money. The economics of wholesale bandwidth are little-understood, but they’re also incredibly effective, and have created a system whereby the amount of bandwidth in the US is more than enough to meet the needs of all its inhabitants. What’s more, as demand increases, the supply of bandwidth quite naturally increases to meet it. What we don’t need is anybody spending hundreds of billions of dollars to build out a brand-new nationwide broadband network.

At the end of the day, we’re arguing over how big of a piece of the tech industry money pie the infrastructure providers should have assigned to them and given the current state of affairs – not that I’m a fan of Silicon Valley oligarchy – I’d rather have it taxed than simply going to cable monopolies.

mavery February 23, 2014 at 1:10 pm

“Note also that Netflix has turned out to be worth a lot of money as a company, a reality which those who pushed the “cable as extreme monopoly” view denied could happen, out of a belief the cable companies would simply confiscate any Netflix rents.”

Recent net neutrality rulings could make it possible for cable companies to begin to confiscate these rents to a much greater degree than was possible previously.

eugenew February 23, 2014 at 1:11 pm

Shafer/Huber nail the basic situation:

(“…blame it on the government… cable TV wasn’t a natural monopoly but a government-made one.”

Monopolies can not endure in a free market, but flourish long term in our progressive American ‘regulated’ economy.
[P.S.] There is no such thing as a “natural monopoly”.

Sebastian H February 23, 2014 at 2:19 pm

“Furthermore the current monopoly power of cable means that infrastructure will be laid down more quickly next time around, and moving to local loop unbundling would weaken this incentive by confiscating some of the rents from the infrastructure investments of the cable companies.”

I’m intuitively attracted to this argument, but don’t we already have facts to suggest it isn’t working that way? The cable companies already have fallen well behind local loop examples in providing infrastructure on the last two rounds of upgrades. Why would the next round be different?

Also when you talk about “penetration rates could be a few percentage points higher” you are ignoring speed. Penetration could be a few percentage points higher, giving access to some. But speed would be much higher almost everyone. That speed difference is increasingly noticeable even for regular users.

Emil February 23, 2014 at 2:25 pm

Pity that it’s just not technically possible to unbundle cable networks (they are a shared media and not point-to-point (in the access) as copper networks (where unbundling is done) are. You can do other types of wholesale access but it has to be after the CMTS (which controls the cable modems in peoples homes) which means that the owner of the cables controls the quality of service. Anybody who claims that this is just an excuse of the cable companis is profoundly ignorant of the technology of telecoms and cable access networks.

Ps1: imposing another type of wholesale access may still be a good idea but it will not be as efficient as local loop unbundling in stimulating competition

Ps2: the cable companies need to continue to invest a lot in active equipment and in bringing fibre further down in their networks (replacing coax cable) to keep up with bandwidth demand. It’s simply not true to claim that the current networks are capable of providing all the needed bandwidth in the future

Ps3. I work in telecoms in Europe (both with telecoms providers and cable companies)

andrew' February 23, 2014 at 3:02 pm

Co-ax? No tgs?

Ray Lopez February 23, 2014 at 3:06 pm

I agree with Tyler Cowen, the original TC not this post, that says unbundling does not matter, and indeed bundling is analogous to the Aesop’s fable involving The Farmer’s Sons. Read this:

The Farmer and His Sons
A FATHER, being on the point of death, wished to be sure that his sons would give the same attention to his farm as he himself had given it. He called them to his bedside and said, “My sons, there is a great gold treasure hid in one of my vineyards.” The sons,after his death, took their spades and mattocks and carefully dug over every portion of their land. They found no gold treasure, but the vines repaid their labor by an extraordinary and superabundant crop

Decoder key (decoder as in telecom):

Dying farmer = LEC (Local Exchange Carrier, or your Cable Company that owns a monopoly on internet access)

Sons = Incumbent LECs

Great Treasure = Perceived local monopoly that yields a producer surplus above and beyond the free market surplus

No Treasure = deadweight loss or absence thereof (depending on whether or not you agree that overcoming a barrier to entry will result in a superior product in the long run)

Grapes = the riches that society and the producers reap from the factors of production

Now put the pieces together, using the decoder (decode the encrypted message): if the sons did not think there was a gold treasure, they would never have bothered to till every portion of the land, and there would be no riches to society. So TC is correct: the perceived monopoly is what drives producers to produce. That’s why I favor patents btw. Put another way: if MR = MC (commodity wheat farmers) then wheat farmers earn a mere subsistence wage, and a paltry life akin to some small scale rice farmers in southeast Asia.

Now let’s review the converse, which is *unbundling* the local loop. This is akin to a bunch of ingrate sons that are looking to “get rich quick” with little effort to overcome any barriers to entry, since it’s being handed to them–a mad scramble ensues, with the local loop suffering from overuse and neglect. It would be akin to the Ambrose Bierce version of Aesop’s fable, which is this:

The Farmer and His Sons (Ambrose Bierce version)

A FARMER being about to die, and knowing that during his illness his Sons had permitted the vineyard to become overgrown with weeds while they improved the shining hour by gambling with the doctor, said to them:

“My boys, there is a great treasure buried in the vineyard. You dig in the ground until you find it.”

So the Sons dug up all the weeds, and all the vines too, and even neglected to bury the old man.

BTW, the Mother Jones article article by Kevin Drum is correct insofar as if we don’t allow competition with local monopolies by law. But in TC’s scenario, such competition would be allowed by law. In the case of Ma Bell, it was expressly forbidden to compete because the theory was Ma Bell would use profits from long distance to subsidize consumers and/or get “universal coverage” (everybody would have a phone, even in rural Alaska). That was the idea behind giving Ma Bell a monopoly, and this fails in TC’s hypothetical.

john personna February 23, 2014 at 5:12 pm

Two things. First my cable operator started as a 16- channel analog service sometime in the distant past (more than 40 years ago). As far as I know they have never been forced by the city to roll out any particular service to keep their monopoly. They own it, or the company that has acquired the monopoly owns it, regardless of whether they decide to ever give us fiberoptic gigabit ethernet or not. Their pattern seems to be to lag best-speeds while constant raising prices. Possibly this means ever-increasing profit margins. Good monopoly deal, if you can get it.

Second, ATT enjoys the second half of the 2-party fake monopoly for my area. If that monopoly grant were enough to drive development and interoduction of new technologies they sure wouldn’t be stuck at 768Kbps.

The world does not work as well as your glass-half-full wish.

Curt F. February 23, 2014 at 5:42 pm

To solve this market failure, we need Congress to act. I support a three-pronged cable internet policy: First, the government should subsidize cable internet for those who can’t afford it themselves. Internet freedoms are essential in an open society, such as viewing House of Cards to learn about democracy. Second, to keep cable internet providers honest, all internet plans should be guaranteed issue: even if someone has a past history of viewing thousands of youtube and Amazon videos per month, they will pay the same price for their internet as their luckier neighbor, who just needs to check gmail a few times a week. But just these two policy elements would lead us back to death spiral: no one would buy “gold”-caliber internet access plans until they wanted to consume lots of video streaming services. Thus, to stop the death spiral, comes the third prong, and the masterstroke of the new policy: Americans will be mandated to buy high-speed internet for their home, or pay a tax penalty. That way, we can stop the death spiral and finally deliver a new, just internet policy to the next generation of Americans.

Curt F. February 23, 2014 at 5:47 pm

And of course, although the mechanism and details are murky, rest assured that my plan will lower internet access costs for all Americans.

Mark Thorson February 23, 2014 at 11:02 pm

Oh no! The Affordable Cable Act! It will require everybody to sign up for cable or pay a big penalty (small in the first year, but massively increasing every year thereafter).

Thomas February 23, 2014 at 9:40 pm

My little burg, in the middle of nowhere, flyover country, is served by two broadband providers, with a third–Google–rolling out a higher quality offering next year. Some neighborhoods adjoining mine are currently served by three providers, with Google on the way to making it four. The suggestion that competition is impossible without regulation requiring it seems less plausible to those of us who actually have competitive providers. Similarly, the suggestion from Salmon and others above that the broadband that most people have is perfectly fine cuts off the possibility of innovation that some of us are benefiting from. Speeds available from all providers are at least claiming to increase as Google builds out its network. Salmon apparently thinks the Google Fibers speeds are unnecessary; the first effect of Salmon’s regulatory regime is to end that kind of innovation.

Turkey Vulture February 24, 2014 at 9:00 am

How does monopoly mean infrastructure will be laid down more rapidly next time? At the least it reduces the number of current competitors who could apply their expertise to lay down the next generation of infrastructure.

john personna February 24, 2014 at 9:35 am

We have the demonstration in the “next time” roll-out of optical fiber. It’s been very slow, right? Which is why Google Fiber makes such a big impact.

Turkey Vulture February 24, 2014 at 11:33 am

So that would suggest TC’s story about the incentives produced by possessing monopoly is missing something important, right, since real world monopoly or near monopoly produces the opposite result? Or am I missing something?

o. nate February 24, 2014 at 11:38 am

Huber has the inklings of a point, but he misses the true story. There are quite good reasons why there are lots of local, municipal-level regulatory hoops to jump through in order to run cable to every building in town. It involves digging up streets, attaching wires to overhead poles, etc. The utility poles in my town are already unsightly enough – I can’t imagine what they’d look like if any company that wanted to hang wire off of them was allowed to because of “deregulation”. Once you acknowledge there are good reasons why local municipalities might want to limit the number of cables being run to every building in town, then the logic of local-loop unbundling becomes a lot more persuasive.

nike air max 95 March 13, 2014 at 4:27 am

Liu was one of the 31 migrant workers elected to the National People’s Congress last year. She is a foot masseuse in southeast China’s Xiamen City. A native of south China’s Anhui Province, Liu was a school dropout at the age of 14 and worked to support the schooling of her siblings.

Comments on this entry are closed.

Previous post:

Next post: