Should Verizon be allowed to charge Internet content providers?

I have long feared this development:

Verizon, Comcast, and their ilk have been lobbying Congress to
transform the Internet into a two-tiered system. By tagging content,
broadband providers would ensure that their own packets (or those from
companies paying them protection money) get preferential treatment and
reach subscribers faster than second-tier content. This would give
companies like Verizon a tremendous advantage as they roll out their
own television and VoIP telephone services.

Telco-cable companies have spent billions to lay down broadband pipe
and want a return on their investment. They are tired of bandwidth hogs
like Google, Amazon, and Microsoft getting a free ride. This was fine
when the Internet consisted mostly of e-mail and static Web pages. With
the advent of online video, Internet telephony, and IPTV, Verizon, AT&T, and BellSouth want content providers to share the cost.
Their reasoning: If Google is going to introduce a video service,
shouldn’t it have to pay for some of the bandwidth it scarfs down?

…If the telcos and cable companies get their way, we’ll have a
Balkanized Web. Content providers who can afford to pay for premium
service will market superior products to consumers with fast
connections. Everyone else will make do with second-class companies at
second-class speeds.

There is much more in this fascinating article.  In purely economic terms, the idea of charging Google or other "bandwidth hogs" does not sound outrageous.  (What would the incidence of such a price hike be?  Would cable connections become cheaper, or do the cable companies have too much mononpoly power?)  But in public choice terms, this would bring politically-influenced pricing.  Don’t expect porn or blogs to get a break.  The net would become much more corporate.  The perils of regulation aside, Verizon probably would favor its own products, and no, Harold Demsetz never disproved this tendency. 

The beauty of the status quo is that web sites compete on the basis of consumer surplus alone.  The bandwidth costs end up as a fixed charge on net access as a whole; I suspect this hits many inelastic demanders, a’la the Ramsey rules for optimal taxation.  Admittedly it may be a bad deal for the poor who cannot afford to connect, but the overall arrangement enhances the long-run "competition of ideas" feature of the net.

One second-best solution is to charge users for bandwidth per se, while not discriminating across differing uses of that bandwidth.  In essence this would tax file-sharing while leaving most content decisions unaltered.  Alternatively, a tiered net could lead to more Wi-Fi networks, whether at the municipal level or constructed by Google.  If that is where we are headed anyway, this apparently troubling development could rebound to our collective advantage.  We might end up bearing the fixed costs of the transition sooner than is optimal, but again the dynamic benefits of the new arrangement might swamp that problem.

Comments are open…will my free market readers defend Verizon’s right to charge Google bandwidth fees?


I read that something like 70% of global internet traffic in terms of bandwidth is file sharing. It is probably less in the US but still it's a significant portion. If bandwidth is taxed, it might be huge for the MPAA and RIAA

This is a twist on the old AOL model when they were the dominant access provider. AOL controlled the pipe for 50+ % of the market; they did not block access to any website but prominently featured those that were willing for pay for placement. AOL still uses a variation of this "carriage" model on and their other web brands (Netscape, ICQ), but they have far less leverage. AOL's model was a twist on the TV network distribution model - whoever owns distribution is ultimately the big money winner not the content providers. Cable put a huge dent in that model. If the Telco's get this to work, I suspect that other access alternatives will quickly emerge sponsored by the big content providers. Isn't Google already experimenting with free wide-area Wi-Fi in San Francisco? The Telco's need to avoid killing the golden goose of access fees, even if they are not as profitable as they'd like to be.

"Isn't Google already experimenting with free wide-area Wi-Fi in San Francisco?"
No. That's the talk but nothing at all is in place at the moment. Google is actually one of many companies competing to offer free wi-fi here. I live in San Francisco, and trust me, I'll be one of the very first people getting free wi-fi.

Perhaps I'm mistaken but wouldn't this, if allowed, be
more or less a natural monopoly? And isn't monopoly
the antithesis of a free market?

P.S. - it is my understanding that a fraction of every peripheral connection fee (small like mine or big like Google's) is "kicked up" to infrastructure and backbone providers.

I wonder how municipalities will deal with this. The companies already hate them.

This is just regulated monopolies trying to grab some more money. And regulatory issues aside, it's not at all clear to me that it'll work. Google getting slow service for 10% of it's customers is one thing, but if they say "We can't provide an acceptable service to customers of Verizon, so we're not going to serve them at all." I bet that Verison would blink first.

Computer Scientist/Lawyer/Economist Barbara van Schewick has a fascinating paper on this on SSRN: "Towards an Economic Framework for Network Neutrality Regulation". The long abstract is worth looking over, even if you're not interested in the article:

Mark - the antithesis of the free market is not monopoly, it's government regulation. At least with a monopoly one can buy a share of the company and capture some of its profit. There is no way to recapture the deadweight loss due to regulation.

DAn, monopoly deadweight loss disappears into the air, it isn't captured by the monopolist.

That aside, i sure look foward to being forced into AOL2.0!

Verizon and others aren't thinking straight, and arguably never have. They built up their networks on various waves of euphoria that claimed all sorts of crap about the levels of profit they'd be able to achieve. The latest idiocy is the so called triple or quadruple play (or quintiple, sextuple, how many services do they think they can sell you?).

Unfortunately the status of their residential business of highly capitalised, low margin - that is to say a utility - doesn't suit their CEOs images as 'Masters of the Universe'.

Offer services and try to move up the stack, by all means. Understand this, the attractions of a Verizon only walled garden are much smaller than that of access to amazon, google PLUS a thousand other small and useful suppliers who can sell you virtually anything. Stomp on the internet and less people stay connected.

Website owners are already charged for bandwidth usage. If anyone doubts this, go to or any of the other hosting services, look at their prices, look at their bandwidth limits, and look at whether or not they charge you if you go over your limit. Nor are the hosting services themselves getting some kind of free ride.

People are missing a key point. Verizon is not proposing to charge Google for the first time. They already do that via peering agreements. What Verizon is proposing is to modify their peering agreements to provide different levels of quality of service and speed, at different rates.

This is simply adding price discrimination to existing business relationships. Whether or not it is fair depends on whether or not Verizon has monopoly pricing power in this relationship, and whether that monopoly is natural or due to existing FCC interventions.

The essential fact is that bandwidth does not have a differentiable cost function. A pipe that can handle X Mb/s costs the same to build and maintain whether it's used at 0% or 100% capacity. So using more bandwidth costs nothing until you reach your physical limit at which point costs flow to users in the form of congestion or the provider incurs the cost of upgrading the pipe to, for example, 2X Mb/s. Note also that what that provider sells the customer (and in this sense, a Google is identical to you or me) a pipe of a certain size connected to the network at large. They do not sell the delivery of individual packets, because controlling the network at that level is extraordinarily costly.

So taking off my I.T. hat and putting on my economist hat, it's hard for me to see how charging for priority is efficient. It seems to be an attempt to introduce scarcity where none naturally exists, an endeavor that can only succeed with the assistance or protection of the state. However, the interconnected nature of the network makes competition difficult. If Google's provider throttles their bandwidth, what they're really doing is degrading the service that all of Google's customers are paying their providers for (likely other firms). The people that are unhappy then cannot voice their objection because the quality degradation comes from someone they don't have a commercial relationship with.

Though, when I put it that way, it seems like ultimately what's happening is providers deciding to charge higher rates for the same bandwidth to certain customers, and I can't levy an objection against that except on spurious "fairness" grounds.

It will be interesting to see the fallout. There's a famous phrase, "the internet treats censorship as damage and routes around it." I suspect if the network retains any of its original robustness, this bandwidth mafia nonsense will fall into that same bucket.

I'm gonna agree with the statements that if tried, this would ultimately backfire on Verizon. The fact of the matter is, this could only work in a natural (or granted) monopoly scenario, but in general, that situation is increasingly. Two major broadband technologies (DSL and cable) already exist and are available in most markets, and in many markets, there is competition for DSL service. Moreover, if the Internet has taught us anything, it's that more competitive markets for information (due to lower barriers to entry) are remarkably efficient. Verizon's proprietary content will never be as valuable to consumers as freely available content, and they don't have a sufficient control of the market to extract additional rents from websites to receive preferential treatment. I would bet that Verizon is betting on users not caring because they prefer the convenience of one bill or some other such argument, but as far as I'm concerned, it's doomed to fail.

However, I'd stop short of short selling Verizon as one person suggested. After all, Verizon's done a great job in the cellular phone market, despite monopolizing content distribution, because multimedia cellular phone services are not yet that mature a market in the U.S. In fact, even as these services continue to become more popular, I would venture that Verizon will do well in that market, because the market for multimedia services is fairly inelastic and consumers make choices on quality of service and other factors. Its competitors have sunk substantial costs in non-CDMA technologies and will be reluctant to switch. I'd expect Verizon to be the market leader here for the near term.

Charging for bandwidth isn't the issue - Google already pay for their bandwidth. (the mistake over what Verizon, SBC etc. want is one they are anxious to encourange) Verizon etc. don't want to charge Google more because they are taking more traffic, they want to charge more because they are taking more valuable traffic - in particular, traffic which competes with Verizon's other current or potential businesses - businesses that have higher margins than the internet bandwidth business.
Now you can argue that Verizon are entitled to charge extra for this more lucrative traffic, but the problem is that they are probably technically not able to do so effectively without breaking the internet. The internet is able to support innovation because carriers do not need to know anything about the traffic they carry but its source and its destination. Once you are required to identify to each carrier on the route what your packets are going to be used for, the era of innovation is over.

There is a tricky argument over whether the market can keep the end-to-end internet going without regulation, but there should be no argument that charging differently for different kinds of internet traffic (as opposed to different amounts or different routes/times) is a bad thing.

I would be happy to accomodate them as soon as they return the value of the monopolies they were granted to fund just this broadband expansion. After all, that was their argument at the time.

A réplica replica Rolex é compo das peças as mais finas da qualidade; tampa de cristal original do seletor da prova do corpo de aço inoxidável e do risco.

ưén van de populairste merken en de modellen van de horloges van de Replica van replica Rolex watches is de Omega modellen.

This subject keeps resurfacing every couple of years. My fear is that eventually the ISP's will get their way...

Comments for this post are closed