Net Neutrality, Libertarianism, and Free Information

Publius/John Blevins at Obsidian Wings has written quite a bit over the years about the issue of net neutrality.  For the most part, I’ll admit, the discussion has largely made my eyes glaze over and hasn’t much interested me.  Then I saw that in Publius’ Clark Kent persona, he recently filed an amicus brief on behalf of several other law professors, including Jack Balkin of Balkinization.   For whatever reason, this has piqued my curiosity about this issue, which I think presents something of a quandary for libertarian thought.*

The standard libertarian position on net neutrality seems to be that it’s a bad idea that will hinder the development of networks, and that it’s a solution in search of problem (made with much more nuance by Julian Sanchez here, and with some caveats here, amongst many other places, I’m sure).  For the moment, I think Sanchez’ arguments are pretty persuasive, in that existing regulations and market forces thus far have done a pretty good job preventing the kind of discrimination about which net neutrality proponents are worried.  Meanwhile, the distinct possibility of new networks developing that usurp the existing duopoly and allow for even more innovation on the internet is very real; that a net neutrality regime would be a mechanism for that duopoly to prevent such competition and innovation from occuring cannot be easily dismissed. 

And yet….the kind of situation that net neutrality proponents fear hardly seems far-fetched.  Although existing regulations and market forces may, for the moment, be enough to prevent this from happening on a significant scale, there is no guarantee that this will continue.  Existing regulations may indeed be enough for existing providers to block any real competition from developing, and those providers already have plenty of incentive to undertake measures that block applications, unduly discriminate against particular types of traffic, and more or less force users to use applications in which the provider has a vested interest. 

There, I think, is where libertarians have a real dilemma.  Obviously, the boilerplate libertarian position would be to remove the regulations that ensure and enforce the duopoly.  This, however, will be an unsustainable position in that it will have a near-zero chance of becoming reality.  As a result, the practical dilemma will be that libertarians will have to choose between increased regulation of networks (helping to ensure that alternative networks will not be able to develop)  and de facto heavy regulations on application and traffic innovations, implemented and enforced by the duopoly rather than the government. 

At that point, one of the great predicaments of libertarianism becomes clear – more government intervention may well be the only pragmatic way of correcting the flaws of a previous intervention.  To be sure, further shutting the door on network innovation and competition would not be a particularly happy or freedom-enhancing endeavor; but neither would permitting the drastic stifling of application-level innovation and information exchange implemented to advance the duopoly’s anti-competitive interests.

In such an instance, I would argue that the proper way of viewing the situation would be to view the service providers as quite literally creatures of the State.  While they may not necessarily be acting on behalf of the State, the effect of their State-enforced duopoly asa gate-keepers to the internet would be functionally indistinguishable from State action to suppress free speech and innovation.  Indeed, because of the nature of the internet, I would argue that their role as gatekeepers would put them in the position of overseeing the commercial and ideological marketplace that is the internet in much the same way that government is charged with overseeing the commercial and ideological marketplace that is the physical world.  A requirement that the providers maintain neutrality – abide by a sort of equal protection principle, if you will – seems to me necessary should it turn out that the duopoly is unbreakable under the existing regime. 

A quick hypothetical should explain why this would be necessary and essential even from a libertarian perspective.  Imagine that the government turned all of the Interstates over to ABC Corp., all the US Highways to DEF Corp., and all the state and county roads over to GHI Corp., but retained jurisdiction to regulate those three companies.  Wouldn’t the government have not only a right, but a responsibility to ensure that those companies did not discriminate in a way that the government itself would be prohibited from discriminating? 

This is why I think the arguments made by Blevins, Balkin, et al in the Comcast case are important.  Indeed, one of the many interesting components of the brief is that they cite to Schumpeter, of all people, in support of their argument that the FCC acted properly in prohibiting Comcast from blocking BitTorrent and other peer-to-peer uploads.  Although it appears that public outcry over Comcast’s actions was critical to Comcast ending this practice, it also seems likely that this outcry never would have reached such levels had the FCC not treated this as a proper regulatory matter in the first place.  Merely by investigating this issue, the FCC played a valuable role in informing market participants about Comcast’s actions that allowed those participants to exercise their market power in such a way as to force Comcast to put an end to its actions. 

Of course, the central issue in the case isn’t whether net neutrality is good or bad policy, even though that is the central issue addressed by amici, and an issue that can properly inform the DC Circuit about the wider significance of this case, which is particularly relevant given that the central issue is whether the FCC acted within its jurisdiction.  That central issue, while perhaps not as “sexy” as the question of whether net neutrality is good policy, is uniquely important in and of itself, especially given that the FCC’s decision came about via adjudication rather than formal rulemaking. 

One reason why this jurisdictional issue is important is that a decision in the FCC’s favor would largely obviate the need for formal net neutrality legislation, which would necessarily be byzantine in addition to at least partly shutting the door on network-level innovation.  Such a decision would make clear that the FCC has the power to investigate and publicize behavior such as Comcast’s.  Moreover, an entirely favorable decision for the FCC may well encourage the FCC to delay implementation of formal net neutrality regulations.  In that situation, the FCC may be willing to develop its net neutrality rules more organically through adjudications, dealing with neutrality problems if and when – and only if and when – those problems arise, rather than creating one-size-fits-all rules for problems that merely could hypothetically arise (rules that naturally would most severely affect small competitors and alternative service providers). 

On the other hand, a ruling against the FCC in this case would give service providers carte blanche to tightly regulate internet use and access and to “pick winners and losers,” as Blevins writes, without any fear of oversight unless Congress chose to act.  The mere threat of an FCC investigation for violation of neutrality principles would be lost, and with it much of the FCC’s capacity to disseminate information about those violations to market participants.  Safeguards that would potentially guarantee the continued evolution of internet technologies and applications would be wholly dependent on either the development of new network-level technologies and providers or Congressional action, neither of which is at all a certainty, and the latter of which carries with it the problem of at least partly shutting the door on the former. 

*HUGE, GIANT caveat to this post: I’m just now starting to learn about this issue, so I reserve the right to be very, very wrong in my interpretation of this case and of the importance of net neutrality in general.

Please do be so kind as to share this post.

27 thoughts on “Net Neutrality, Libertarianism, and Free Information

  1. “Solution in search of a problem” sounds about right to me. Let’s rewind a dozen years.

    “What if AOL decided to do something really evil and stupid??? We’d want the government to intervene, wouldn’t we?”

    Well, let’s look at the last 12 years. Seems almost quaint to see the string “aol”, doesn’t it? Here’s another fun one for you: “bundling internet explorer with windows gives an unfair market position to Microsoft and is discriminatory against Netscape and we need to have the government intervene to make sure the marketplace remains fair.”

    Any regulation of this issue will, by the time the lobbyists are done writing the legislation, and by the time Delaware and South Dakota and Nevada senators wet their beaks, and by the time everything else is done… this will be more time than it took for the phone companies to brag about having “the first 3G network” to bragging about “the first 4G network”.

    Personally, I’m wondering if there isn’t also a little bit of regulatory capture going on here. Will this legislation allow Comcast to skate on its current capabilities for a little while longer while making it tougher for young Upstart companies with new/better capabilities to compete?


    • Right, which is why for at least the short-term I’m definitely opposed to actual legislation or, for that matter, comprehensive regulation. That said, I think there’s a very strong argument in support of the FCC’s actions in acting via adjudication in this particular instance, since the effect of adjudications is much narrower and may obviate the rationale for comprehensive legislation/regulation.

      One thing that worries me in the long run a bit is that in the broadband age the service providers are now almost entirely utilities that effectively have natural monopolies over the delivery of services in no small part thanks to the state’s powers of eminent domain and the like, which is why I analogized it to government selling the roads system to just a handful of companies. I guess my point is that the existing system may (or may not) already be so corrupted by regulatory capture as to effectively prohibit any real competition on the provider level. If, over time, this proves to be true, then we’ll face a very real tradeoff between a free market on the application-level and a nominally (but perhaps not actually) freer market on the provider level, unless we can successfully remove the barriers that prevent competition from arising on the provider level.


      • Didn’t we think that AOL was a monopoly though?

        I worry that the first instance of intervention will be exceptionally good, I guess, (hurray, young men trading copyrighted movies through bittorrent can still trade movies!) but as capabilities naturally expand and the thinking about such things evolve, the legislation will be a collar chained to a fencepost.

        Compare the web from 1997 to today. If the web of 2021 is not as different (or, at least, within an order of magnitude of difference), it will be in no small part due to legislation surely passed with the best of intentions.


        • Oh, totally, although the AOL situation is I think different in kind since it did not have control over the infrastructure. Whatever else you could say about AOL, I don’t think that at any point you could say that it was able to hide behind many regulations to keep competition at bay.

          Still, for now I’m very much opposed to comprehensive legislation or regulation since I do think that it is currently a solution in search of a problem. It’s more that in the long run the existing regulatory hurdles to new service providers could very well prove too high for competition to enter the market; if that’s the case for any appreciable period of time, and there’s no possibility of removing those existing hurdles, then the only solution may be to, in effect, treat the primary existing service providers as arms of the government themselves.

          This would not be an ideal situation by any stretch of the imagination, but I think it would be a better situation than the alternative of having two or three state-protected monopolies that are free to control access to the internet as they choose.


          • One other thing – if you have a little while, I strongly recommend reading Blevins’ brief. Even if you don’t find it persuasive, it’s an excellent piece of legal writing that is nonetheless accessible and lays the overarching issues out quite nicely, although it doesn’t really touch on the jurisdictional issues.

            The brief is here:

            You can start reading at page 12 of the PDF.

            He touches on a bunch of other issues that I don’t touch on here (and which I wish I had). It’s interesting, at the very least.


            • Here’s something I saw on page 22 that made me scratch my head:

              First, singling out specific applications on a network reduces
              developers’ incentives and abilities to develop new applications.

              My response was “really? It seems to me that that would result in a host of new applications that would spring up organically. Bytetorrent. Flitorrent. Torrentator. Torrentisimo. Torrentino. (Makin’ Copies.)


              • I think the point is that by having the ability to arbitrarily and without warning single out specific applications, providers reduce incentives to develop new applications. So, if someone wanted to develop a competitor to BitTorrent, Comcast’s actions in singling out BitTorrent would signal to the potential competitor that it could be singled out as well should it attempt to enter the market. It’s the arbitrariness and uncertainty that creates the disincentive.


                • Hrm. This would create problems for Amalgamated Bitorrent Competitor, I agree, but it wouldn’t create problems for Anonymous Basement Citizen.

                  The second ABCer would merely change this bit, that bit, and the other bit, announce it to 30,000 of his closest friends via one of the messageboards out there, and the world would know about it by Wednesday next. Comcast wouldn’t be able to pick a name for a Tiger Team in that period.


                  • Ahh, now you’re getting into the technical side of things, where I’m particularly unqualified to comment. But my understanding (and there’s a p>.5 chance I’m wrong here) is that the way in which this sort of thing is done wouldn’t necessarily have to be application-specific – it could look for certain markers or certain traffic patterns, and block that traffic.


                  • That’s a true point, but a great number of applications have serious networking effects.

                    If Bittorrent is blocked off by Comcast, yes you and I and five other guys can download the source code for bittorrent, re-write a chunk, and resume sharing stuff with almost no downtime.

                    However, this stops Grandma from using bittorrent to get to her daughter’s pictures on the web.

                    Now, you can easily say that Grandma isn’t using bittorrent to get to her daughter’s pictures anyway, because she’s using Flickr or Facebook or whatever. But the point is; sharing data between N users effectively requires N-M of them to have a standard for sharing files in order to leverage a network effect.

                    If I can arbitrarily decide that I don’t like the amount of pipe they’re using, as their ISP, and I’m allowed to throttle stuff arbitrarily, that’s bad.

                    I’m okay with bandwidth caps (although I wouldn’t ever sign up for a service *with* one), but traffic shaping seems to be a very bad power imbalance.


  2. I’m not understanding how net neutrality could limit the development of networks. Doesn’t having the roads/intertoobz open for everybody , using open standards and without a company playing gatekeeper give more space for innovation. If I want to start a business that involves driving I don’t have to ask permission to use the roads or pay for access to standards, I can just get on the road. It seems like what Comcast was doing is bad for the toobz and free flow of porn torrents info.

    Edumacate me.


    • Here’s my understanding (please, please, please, someone else please correct me if I am wrong).

      At its heart, this is a commons issue. 80% of all of the bandwidth out there is consumed by 20% of the users. This 20% are trading all kinds of data back and forth and, if there are hangups caused, these hangups affect everybody which means that the super-duper majority of people who aren’t torrenting hi-def, erm, movies are paying the price (through slower bandwidth, etc) of the actions of that 20%.

      This legislation, among other things, would protect the majority of folks (whose ideas of “heavy use” is checking Drudge, watching the Boxxy video again, then playing WoW for an hour or three) out there while limiting the 20% who are ruining it for everybody. It has the added benefit of protecting the intellectual property of, ahem, movie makers.

      (Did I screw that up?)


      • I think you messed this up a bit. You’re right that this is at least in part a commons issue, but the position you’re describing is more the position of Comcast, et al, not of any hypothetical net neutrality legislation.

        What net neutrality legislation, if passed in its idealized form (and obviously, it would not get passed in its idealized form) would do is to prohibit service providers from blocking the voluntary download of applications or favoring one sort of traffic over another. In a nightmare scenario, it would prevent, say, Comcast from blocking access to Hulu or from entering into agreements with Rhapsody to block access to iTunes (or vice-versa). It would also prevent providers from creating requirements for application operability such that application designers would need to design different programs for each service provider.

        So far as I can tell, a key feature of net neutrality legislation would also be a disclosure issue so that customers know when their provider is restricting their access in some fashion or another.


        • (In my defense, I kinda figured that Greginak knew the arguments against the sort of thing that Comcast was doing.)

          The hypothetical legislation you’re talking about is legislation that I would probably agree with, for the record. The problem is that the actual legislation will have have a name like “The Net Neutrality Act Of Internet Freedom And Liberty And Justice For All 2010” and it will be a bill that will lay out, specifically, exactly the circumstances under which Comcast is allowed to block Bitorrent users, slow Hulu users, and give priority to Rhapsody users… for the sake of the commons.


      • Close.

        Typically all telecommunications lines are *massively* over-provisioned. That’s because almost *nobody* uses the capacity to which they have contracted access.

        Here’s a quick example: a modern corporation of, say 300 people does not maintain a phone switch with 300 lines (cough, assuming they’re using a PBX and not VoIP). That would be complete insanity, you are *never* going to have 300 people on the phone at once, unless you’re a call center.

        So it’s up to the telco guy to monitor the call usage and figure out how many peak phone calls you make in a day, and sign telco contracts such that you can make your peak (assuming it’s business necessary to be able to hit your peak). This might be as low as 10 lines or less, nowadays.

        Okay, now imagine that there’s one department (the IT guys) who don’t make a lot of phone calls, but each call takes over 40 minutes, because they’re on hold with tech support waiting for their call to be processed. If we ever have 3 IT guys on 3 calls, all sitting on hold, we’ve parked three of the incoming/outgoing lines and nobody else can use them. Suddenly, it’s possible for the CEO to pick up the phone and be unable to get an outgoing line. We have a (phone) network saturation event.

        This is more or less exactly how the data provisioning goes with your ISP. You contract for a maximum bandwidth (say, 768 up/1.5MB down). But you and your other data users in the same neighborhood are all getting DSL service through the same central office. That central office itself also has a pipe limit.

        If you and I and N other people in our neighborhood are all downloading HD movies over Bittorrent (or AT&T’s Uverse or whatever), we’re always at or near our pipe limit. But the DSL provider *expects* the vast majority of their users to come *nowhere near* that limit. If we represent a statistical anomaly in our user group, we can cause problems for the provider.

        Even though we’re both still using our contracted amount of pipe.

        A lot of backhaul networks do traffic shaping (they have to do so, to provide mandatory call service). So once all the lines get to the CO, everything is converted to ATM packets or IP packets and shoved through the same pipe to the large telco hubs, where it’s routed wherever it is supposed to go. All of *that* traffic is assigned QoS, and the phone traffic is typically guaranteed, so all the data is a lower priority (unless I have a contractually mandated pipe, like a DS3 or whatever).

        Basically, the anti-net neutrality folk want to be able to continue to bullshit their contract statements: “We offer up to 1.5MB for $19.95 a month, or 4MB for $25.95 a month, or the incredible deal of 10MB for $43.95 a months!”

        You’re unlikely to get *any* of those speeds. They’re certainly not going to give you 10MB *all the time*, unless they are legally required to do so. And most contracts for private telecommunications aren’t written that way. In fact, in many cases it’s happened that users have been consistently limited to a lower tier of service than they’ve signed up for.


    • It adds an extra regulatory hurdle, which always increases costs of entry to begin with. But beyond that, it forecloses the possibility of valuable networks that violate principles of net neutrality in rather benign ways that would in fact undermine the existing duopoly.

      I am admittedly speaking very much on the meta-level here without giving any kind of specifics, since I’m just starting to get interested in this issue. I guess the quintessential hypothetical example would be something like a situation where a new service provider figured out a way of offering high-speed service at a significantly better price than the existing duopoly, but only if it blocked access to certain corners of the internet or to certain applications (a flaw about which it was open from day one). This service would provide a means of breaking the duopoly but would be illegal under a net neutrality regime.

      It seems to me that in the short run, taking steps that may break the duopoly by removing restrictions that may protect that duopoly is the low-risk, high-reward course of action, while taking legislative (rather than adjudicative) steps that would guarantee net neutrality but further entrench the duopoly is a high-risk, low-reward course of action.

      That said, should Comcast win in this litigation, it would seem that the risks of having no formal net neutrality regime increase substantially, while the reward of doing so also increases substantially since at that point the duopoly would have carte blanche to not only violate neutrality, but to do so secretively. That’s not to say that it would be immediately worth doing – just that it would be a closer issue since both neutrality and deregulation would then be high-risk, high-reward scenarios. Finally, though, if it turned out a few years down the line that the duopoly was persistently violating neutrality, with no signs of meaningful competition developing, then I think a net neutrality regime would become the low-risk, high-reward option, and deregulation or, more likely, status quo would be the high-risk, low-reward option.


  3. What exactly would be the method of breaking the “duopoly” be?

    Giving out a gigantic cash grants to local municipalities to develop wireless infrastructure?

    I’m not quite sure if there’s a feasible method for actually increasing access that doesn’t involve favoring one or two very very very large firms that in the long run would have the same sort of market power that current landline providers do.


    • I honestly don’t know beyond my general inability to imagine that which does not exist, especially in an arena where my understanding of the technology is limited at best. That said, the first Sanchez link makes this suggestion:
      “Instead of giving wireline incumbents a new regulatory stick to whack new entrants with, the FCC could focus on facilitating exploitation of “white spaces” in the broadcast spectrum or experimenting with spectral commons to enable user-owned mesh networks. The most perverse consequence I can imagine here is that you end up pushing spectrum owners to cordon off bandwidth for application-specific private networks—think data and cable TV flowing over the same wires—instead of allocating capacity to the public Internet, where they can’t prioritize their own content streams. It just seems crazy to be taking this up now rather than waiting to see how these burgeoning markets shake out.”


  4. We hit on something here and ran right over it. Debating this topic from a libertarian point of view is problematic. We don’t live in a libertarian society, many assume that we are slowly headed towards socialism. If I may make an oversimplified analogy, net neutrality is like deciding on what flavor of ice cream to order in a world where there is only vanilla and chocolate. I’m sorry to those that think this is a valid libertarian argument, it is not, in a libertarian world Net neutrality would be a mute point. You cannot argue a libertarian philosophy inside of a socialist regime and not look like the fool you seem to me. Its like being in a quantum mechanic course and arguing a point based on Einsteins’ general relativity. Does this make sense to you psudolibertarians or libertarian-lite types?


Leave a Reply

Your email address will not be published. Required fields are marked *