Would net neutrality hurt the poor?

Eli Dourado has a new piece of note:

In much of the world, the net is not neutral, thanks to companies like Facebook and Google.Facebook Zero is an initiative launched in 2010 to give customers of 50 carriers, mostly in the developing world, access to a lightweight version of Facebook on their WAP-enabled feature phones at no charge. Users can post, like, poke, and comment to their hearts’ content, but if they want to view photos or access non-Facebook sites, they incur the usual data charge. The model has been so successful at growing Facebook adoption in Africa that Google followed suit with a competing offering, Google Free Zone in 2012. Lest anyone think that this is a cruel ploy by evil, for-profit corporations to trap the poor inside their walled gardens, the non-profit Wikimedia Foundation also copied Facebook’s idea with Wikipedia Zero, to great effect.

…non-neutrality can also help to fund necessary network buildouts on an ongoing basis. By giving access to Facebook, Google, and Wikipedia away as a loss-leader, carriers are serving with their basic tier of service those who can’t afford more, and habituating those who can afford to click beyond the walled garden to using the mobile web. This price discrimination not only increases access but also raises more revenue than a neutral strategy would. Developing-world carriers need that revenue if they ever intend to build the kinds of networks that will support widespread Internet use. Net neutrality, in other words, would not only keep the poorest offline, it would keep investment in poor-country telecom infrastructure down for longer.

A similar, but less stark, dynamic is playing out in rich countries. Anyone who has ever used their Kindle’s included 3G service has benefited from network non-neutrality; after all, you can’t use it to access non-Amazon services. Absent Amazon’s non-neutral arrangement with wireless carriers, you’d have to pay a nontrivial monthly fee to access books via the cellular network, which would mean that most people would forgo cellular and stick to Wi-Fi. Again, we observe a non-neutral arrangement expanding access and saving people money.

Read the whole thing.


I will read the whole thing but in an effort to be first poster three times I will simply say African telecoms are creative on fees: when it was discovered that some people ring others a certain number of times then hang up, without the receiver picking up, as a sort of code, the African telecoms started charging for such tactics, so the caller would get a small fee added for making a call.

as a child of the 80's I used a similar tactic with collect calls:

"Hello, would you like to accept charges from "HEYDADCOMEPICKMEUP" ?

Explicit government backing of Fannie and Freddie "expanded access and saved people money." And yet libertarians are against that but ok with net nonneutrality. But then that's because big government distortions are bad, big business distortions are good.

Libertarians are against gov't handouts to cable companies.


I'm against all those cable monpolies granted by cities.

I'm also against the FCC saying "you can't charge for data transfer", which is really what Google and Netflix want from "Net Neutrality" (which is not "neutrality", no matter what impassioned Facebook graphics about "charging you to use Facebook!!!!" say.)

The ISPs are already charging for data transfer. Moreover when they take my monthly service fee, they are contract to transfer data from whatever part of the Internet I choose. If I choose Netflix, they cannot justly turn around and say "Hey, NF. you also have to pay up or else I will break my contract with AR".

There are plenty of non-neutral arrangements which are just:

(1) My service contract is for only 300MB / month -- not enough for video, so NetFlix et. al. pay to have their data not counted in the cap.

(2) I buy a special purpose "NetTV", a widget which only uses the internet for downloading from NetFlix.

All examples in the original article seem to belong to (2). If ISPs don't have to honour their contracts, then every network connection will be some variation on that too.

When a residential customer pays their ISP for last mile Internet access services, they are not paying for the delivery of licensed content. They are paying for access to the Internet at a particular "speed." Their ISP is not selling them cable tv services. When a residential customer buys content from a content delivery service, they are paying that content delivery service for the delivery of that content.

The residential Internet access services customer does not have "a contract [with their ISP] to transfer data from whatever part of the Internet [the residential customer] choose[s]."

*Slam dunk* You just annihilated that straw man, like it was made of... ugh... straw.

I would argue that the physical differences between the African cell carriers and the wire-based last mile net monopolies in this country are too great to generalize the observations. We will not get bigger wires from the monopolists--they are just extracting rents.

I would also point out that in these instances the telecos are still "single" dipping. Facebook is paying for the internet access for the users and then limiting the access that Facebook is willing to pay for. It is not unlike the internet access my employer provides at work, while still blocking NSFW content. Net Neutrality is a more complex interaction between double dipping and monopoly, where I pay for interact access but then that access is degraded unless the content company also pays. People are upset that Comcast is charging them to get on the internet and then Amazon, Netflix and Hulu in exchange for not degrading the bandwidth. If Comcast wants to give me free access to their content services without me subscribing to their internet service, no one would be protesting.

The fun really starts when you get premium billed by your ISP for visiting xhamster. :)

It seems to me that the difference between something like Facebook Zero and net neutrality is that in the former Facebook is paying for their content to to be delivered to the end user (one side paying), whereas the latter demands that if the end user pays for internet at X/Y Mbps upload/download the provider of the content to them is not also charged for shipping them content to them within the rate limit that the end user is already paying for.

I wouldn't see a big problem if I paid for, e.g. 5 Mbps download, but when downloading software updates, e.g. Apple paid additional fees to momentarily boost my download rate to 1 Gbps. To me that doesn't seem to be something that "network neutrality" rules should prevent - it's trying to charge both sides for the thing the other party seems to be already paying for that I object to.

I'd be against it, too, if it was in fact "charging for what they're already paying for".

But it ain't.

Netflix (the biggest current context, but this applies to anyone in principle) charges you $8/mo "for streaming videos".

Your ISP charges you e.g. $50/mo for "up to 50 mbit downloads".

Net Neutrality does... what? Prevents the ISP from charging you extra to watch Netflix? They're already not doing that; if they were that would be something to complain about.

(Certainly that $8/mo does not somehow entitle Netflix, Inc. to get free data transfer over every network without paying for it; transit isn't free, and Netflix is in no position to do peering, precisely because it's not an ISP - it's a source of huge amounts of data. And that costs. Which is what they need to spend part of that $8/mo on; buying bandwidth, and not just "the cheapest someone can offer", but "bandwidth actually connected to subscriber networks in a decent way".)

What they are doing is not giving Netflix preferred local access to their network (colocation) for free, of giving their CDN free peering - which makes them "slower" than colocated CDNs (Akamai, etc.) or people on peered networks.

Netflix is now paying big ISPs for colocation (or using a fast peered network CDN) rather than using a non-colocated (or non-peered network) CDN, just like everyone else with serious data needs.

Net Neutrality is, in summary, Google/Netflix/etc. trying to sell "let us not pay data transfer fees like everyone has since the dawn of the internet" as "consumer protection".

(Contra other posters, "neutrality" is not why "Europe has fast internet". A map suggests why; most of Europe would fit in about a third of the US; dense areas are easier to serve. I can only speculate that national political differences and regulatory regimes have significant impact; they generally do. Local politics also suggests a reason - local monopolies have little incentive, "neutrality" be damned, to provide Extra Fast Anything.

And contra Jeremy I'm not aware of any actual abuses by the ISPs, even monopoly ones - apart from the monopolies themselves, which were of course granted by local governments, not the ISPs.)

In terms of abuses, its pretty clear that Comcast, etc. were intentionally degrading end-users' Netflix experience in order to extract rents from Netflix.

Not to me. Evidence?

JLV has already presented the evidence to the FBI.

I don't trust the cable companies any further than I can throw them, but that doesn't mean that net neutrality is the right regulation.


I wouldn't trust the conclusions arrived at in that Mattvukas link. I would point you instead to his follow-up blog-post where he's furiously backpedaling and equivocating like a storm. I qoute in part from his later blog post (which is located http://mattvukas.com/2014/02/17/venturing-comcast-netflix-rabbit-hole/):

"First, let’s make clear what Comcast is not doing

After my last article became popular, I received some criticism from people who took issue with my assertion that Comcast was purposefully blocking Netflix traffic. These naysayers are indeed correct: because of an Open Internet agreement that Comcast inked with the FCC as a stipulation of it’s NBC Universal merger, Comcast is prohibited from blocking certain types of Internet traffic until 2018. So, despite the recent Net Neutrality court case, Comcast would be violating Federal Law and violating the terms of their NBC Universal merger agreement if they blocked Netflix."

So... in other words, Comcast was *definitely not* throttling Netflix. The rest of the second blog post is basically wild, ungrounded speculation on his part.

This is actually verifiably a completely false statement. There was no degrading (or as some would say "throttling") of last mile services by Comcast with respect to Netflx content. The bottleneck was at the backbone level, on the interconnection(s) between Netflix's network services providers/content delivery networks and Comcast (and it appears to have been primarly on the Cogent-Comcast interconnection(s)). There was no intentional degradation of services.

It looks like Netflix is now in the process of shifting from paying third party intermediaries to transfer their content to ISPs, to directly interconnecting with certain ISPs. Netflix has always been paying for network services. Nothing new is happening here.

CW is right. Here is more: http://arstechnica.com/information-technology/2014/04/after-netflix-pays-comcast-speeds-improve-65/

Netflix pays their ISP, I pay mine. Now my ISP wants to charge Netflix extra to deliver the traffic that I've already paid them to deliver for me.

You're misrepresenting the commercial relationships. When you are buying residential Internet access services, you are not buying the same product that Netflix buys when it buys network services from its network services providers/content delivery networks. They are completely different. You haven't paid Netflix's network services costs (or bought the network services products Netflix requires) when you buy residential Internet access services.

No, your ISP is *offering* Netflix the opportunity to have their content delivered in a fashion that minimizes the amount of lag in its content.

>And contra Jeremy I’m not aware of any actual abuses by the ISPs


And 1980s Ma Bell breakup was bad for poor people as well!

Need to go further back... To when the federal government nationalized Bell in 1918, handing them the monopoly they'd always dreamed of. Of course, listening to the rhetoric of the time, telecommunications was a "natural monopoly" anyway, so the federal government was really just doing us all a big favor.

The situations aren't closely analogous. The much more analogous situation is Europe. Both sides have net neutrality, but we allowed more of a monopoly system. They got very fast internet, we didn't. You postulated in an earlier post that the quasi monopoly could lead to better investment in infrastructure because of guaranteed profit. It hasn't.

So in theory you might halve been right, but in practice you weren't. I would suggest that whatever the reason for that is likely to come into play here as well.

I don't think there is really a causal-relationship between whether or not a country has net neutrality rules in place (I'm assuming these would be regulations rather than commercial agreements between the parties to abide by net neutrality principles?) and the fact that it has faster or slower last mile Internet access services. Or, at least I'd like to hear a compelling argument for how the two are linked together.

Also, I think there needs to be a more precise terminology than just saying "very fast Internet" or "investment in infrastructure." Most of the big players who are selling last mile Internet access services to end user consumer customers are also selling network services beyond the last mile (all different kinds of network services), and they own and operate regional, national, or international backbone networks, with carriers interconnecting together as a matter of course. These backbone networks are being upgraded and enhanced on a continuous basis (this is clearly evidenced over the life of these privately financed networks). I seriously question whether European back-bone networks are not more heavily invested in (or better) than US backbone networks.

Sorry, typo in my last sentence. Should read: "I seriously question whether European back-bone networks *are* more heavily invested in (or better) than US backbone networks."

I keep on hearing this mantra about how Europe supposedly has faster broadband access than the US. Except it isn't true... See e.g. the Akamai "State of the Internet" reports (very easy to find on their website) which in its last version ranks the US 10th in the world both in terms of average (10Mbit/s) and average peak (43.7Mbit/s) bandwidth per connection (see figures 20 and 21). In terms of average bandwidth, there are six small European countries (out of c. 30 depending on the definition and not including any of the EU5) ahead of the US. in terms of the peak average bandwidth (which as Akamai itself acknowledges is the correct measure of capacity) there are only three, again very small, European countries ahead of the US.

Spending quite some time in U.S. and living in Europe the truth is Europe has much faster broadband access. The other truth is that vast majority of people don't need it because there is no content to use it (and that's why they don't pay for it and are not considered in official research). Even lonely farm up in the hills can have FTTH access 100/100 not to mention huge investments to FTTH networks all over Europe for people on the countryside. For example, because of such investments people in cities in my country has much lower broadband access than people living in middle of nothing.

It says Germany runs on 10 Mb/sec connections. They need to sink 80 billion euros for faster optic fiber network and no one knows where that money is coming since strict price regulations make impossible to ISPs save money for infrastructure development.


We think we're opposed to non-neutrality, but what we're really opposed to is abuses by Comcast and the other monopolies.

Non-neutrality might lead to a viable business model for online media. Which could lead to creation of better content and the rebirth of popular music.

I don't follow how that could come about unless the content providers are also the same as the content delivery services.

That couldn't possibly be Comcast/NBC/Time Warner/Universal Co.'s goal, could it?!

Neutrality only matters when you have a provider monopoly. If someone creates an internet provider service that anyone can reproduce, then that's not the same thing.

Permitting additional revenue sources to owners of wired internet connectivity services has, in the context of the United States and Canada, almost invariably resulted in increased rent extraction from people under quasi-monopolistic utility providers. Usage-based metering, for example, if implemented efficiently should result in a global decrease in flat-rate service fees - thus decreasing prices for the vast majority of users - and higher fees for heavy users, resulting in an overall higher revenue stream for the provider under the same infrastructure; instead, several implementations in Canada invariably placed an arbitrary cap on traffic and then extracted fees from heavy users with no changes or improvements in service. And increases in subscription fees came regularly as usual.

Similarly, advocates of non-net neutrality frequently resort to using efficiency arguments analogous to, say, premium courier services versus the mail to show how these systems can bring benefits. The question that has to be asked is whether the ideal result of non-neutrality, which presumably is better access on the part of popular content providers, better infrastructure and a lower proportion of costs borne by noncommercial consumers, is more or less likely than an implementation where price levels for consumers are maintained, the fees for content providers merely become another revenue stream, and new infrastructure is minimal. The past behavior of ISPs and their quasi-monopolistic nature in the US suggests the latter scenario rather than the former.

More generally TC seems to have a very myopic focus on downplaying the monopolistic power of ISPs. I am very surprised (perhaps I shouldn't be) that he hasn't ever mentioned the many instances of cities with poor quality and expensive internet service starting publicly run, nonprofit ISPs only to have state legislatures ban them in the name of "competition" and "protecting jobs and investment by regulating local government competition with private business," ignoring that this is a clear instance of local monopolies seeking to further entrench their market power when faced with better-run local competitors.

Or, in case both of those are too lowbrow, http://www.gwlr.org/wp-content/uploads/2013/03/Stricker1.pdf

Since Tyler loves Quora, I'll just leave this here...


Money quote: "... The objection to this is that ISPs may not have their users' best interests in mind when making these active management decisions."

The answer is obviously "yes" because net neutrality = government regulation of a government-produced product (the internet), so it's like evil squared. We need competition between websites to battle it out over who gets to go through the pipes first, which can be done through a free market and ideally monetized with bitcoins. Then the poor will benefit...somehow.

Show me a model were this scenario would play out.

Poll data, from the earlier post, shows that people don't care about the poor, so I don't see why we are even discussing this.

Comments for this post are closed