As reported by Broadband Reports and now confirmed elsewhere, a Time Warner internal memo indicates Time Warner will pilot a program where it has an explicit bandwidth cap, and users that exceed the cap will pay additional explicit fees — rather like what happens now with your standard cell phone package where you buy a bundle of minutes and then pay for any overages. The pilot will include a website to allow customers to track their usage, moderate their behavior, or buy additional capacity if they wish.
I agree with Dave Isenberg that this is the best way for Time Warner to handle its network capacity constraints and address the supposed 5% of users gobbling 50% of the bandwidth. We can expect some heavy users to move to other networks without caps, but also expect that users that use much less capacity and frustrated by congestion caused by heavy use by others to prefer plans like Time Warner’s because it should produce a less congested pipe overall.
I would be remiss if I failed to note that I was just musing about this the other day, giving me a chance to do another Stephen Colbert I CALLED IT!!! dance.
O.K., shameless gloating over. Analysis below . . . .
As I have stated numerous times in the past, it is far better to be upfront about your network limitation problems than try to lie about them. Yes, it means admitting you built a crappier network than your rivals, but it will keep you from getting investigated by the FCC and will avoid pissing off your customers.
Unlike charging third parties to override user preferences (what I call Whitacre Tiering to avoid confusion with selling different service packages to the customer), this behavior at least links the problem (lack of capacity) with the cure. While it will to some degree encourage operators to ration scarcity, it also includes incentive to increase capacity to keep with up with demand (because users will only pay so much additional for use of capacity above their existing bundle). The cell phone network provides a reasonable example of this dynamic. By contrast, allowing providers to charge third parties to reach users creates a “two-sided” market where network operators have incentive to maintain scarcity and the capacity to capture monopoly/duopoly rents. (See this paper by Economedes and Tag for the proof.)
On the other hand, while I think this beats outright lying about your limitations, it’s not exactly a win-win for folks. As Ben Scott at Free Press observed, this highlights the overall crapiness of U.S. networks compared with other industrialized nations that offer faster, cheaper packages and don’t have to resort to metered pricing to address capacity issues. And while I hope this has a pro-competitive/shaming effect that will continue to spur investment upgrades and fiber to the home, it does allow network providers to forestall upgrades by incenting users to cut back on their own use.
Why is that a problem? Isn’t that more efficient and puts less strain on the network? Sure, but that has ripple impacts across the economy that suck for everyone else. Ecommerce is made possible in no small part because users have unlimited access. I can go online, browse, try stuff out, and stream or download content easily. I can view a teaser clip of a movie before deciding to buy it, or upload my video to Youtube, or watch someone else’s video, without worrying that I’m using “my minutes” (or, in this case, “my capacity”).
Switch to a metered pricing system and that calculus changes. Yes, networks will experience less congestion. But this isn’t like electric power, where conservation has benefits beyond dealing with network capacity. Conserving power has overall benefits for the environment by reducing consumption of fossil fuels and so forth. It also doesn’t have nearly the same ripple impact on the economy. Whether my thermostat is at 65 or 72 does not effect if I try a new service or go shopping for additional goods. By contrast, my willingness to spend time on line has a major impact on my willingness to engage in e-business or try new online services. At flat rate, I browse Ebay as much as I want. At metered, I only go when I am sure I want something. At flat rate, I can subscribe to Yahoo! Radio or similar services. At metered, I have to factor in the additional cost of the connection.
So while I like Time Warner’s approach better than Comcast, I don’t pretend it solves the real problem. I also suspect that the folks who hope to make money off the internet, whether it’s the Ebays and Yahoos and others that rely on users staying online, airlines or banks or others trying to shift certain operations to much cheaper online processes, or even average workers trying to telecommute, will find themselves unhappy. And, as we spiral into a recession caused by the collapse of the housing bubble and the ripple effects of the credit crunch, we really don’t need to see people retreat from their ecommerce activity because metered pricing encourages “more efficient” use.
The real solution, of course, is policies that build out more capacity so that it becomes “too cheap to meter.” I expect a rash of experts will tell me that such a thing is ridiculous, despite the fact that we see it happening in other countries. Of course, the same experts told me on Monday that the idea of moving to metered pricing with user-defined traffic shaping was impractical, so I shall continue to press for solutions that solve the real problem of capacity rather than solve the problem in a way that conforms to the chosen business model of the broadband access providers.
Stay tuned . . . .
Harold, we over at ThirdPipe have the same concerns. Fact we see a most ugly scenario where TW would do nothing essentially other than rake in the extra fees but do nothing about the congestion. It smacks of hush money for the abusers all because TW set the bar too low on the rates.
The other problem is futures. We have a nascent TVoIP industry attempting to take hold. Metered services would squash that in a heartbeat. But again if the bar is set too low then folks pay a tad more and now 95% of the users are using 100% of the network. So you’re back to square one.
We need the equivalent of the Railroad Act of the 1800’s. Only for telecom.
From a Time Warner Statement:
“Company spokesman Alex Dudley said the trial was aimed at improving the network performance by making it more costly for heavy users of large downloads.
Dudley said that a small group of super-heavy users of downloads, around 5 percent of the customer base, can account for up to 50 percent of network capacity.”
So Time Warner plans to make it “more costly”, for what? Using the capacity paid for? A common download broadband speed is 4 Megabits per second. Although the fine print explains that the connection is provided on “good faith” terms – which means it can be interrupted or downgraded at will by Time Warner, the marketing and advertising designed to sell the connection is clear – the connection is made intentionally attractive as having 4 Mbps of available, maximum capacity.
It’s like buying a car with a top speed of 90 mph. One may rarely drive that fast, but that’s what was paid for and expected to be available as desired.
So what’s the problem with using the full capacity of the broadband connection paid for as well? Why are these customers instead classified into the “heavy user” category?
The answer is this is not the customers’ problem. It’s Time Warner’s problem for overselling the capacity in the first place. As stated above, Time Warner complains that 5% of the customers use up to 50% of the capacity. So why didn’t it build the system network large enough to provide what was sold?
Do 5% of the customers of a 90 mph car that actually drive 90 mph cause the car company a problem? Even if all car customers drove at 90 mph, why is that a problem for the car company? The problem occurs only if the cars don’t perform at 90 mph as sold, and even then, it’s not the customers’ problem.
Now that users are starting to use more capacity per customer, Time Warner is backpedaling from its marketing promises and claiming “heavy users” must “pay more”. Why? Because they’re using all of what was sold to them, while before they were using only part of it?
Instead of Time Warner providing the original broadband speed necessary to serve ALL customers on the network at the MAXIMUM SPEED SOLD per customer, it is now planning to single out customers who actually use the maximum speed and raise the price – again, just for using what they were sold, like driving the car at 90 mph instead of 50 mph.
Why shouldn’t the Federal Trade Commission or the Federal Communications Commission intervene in on the basis of deceptive business practices? If all cars were sold as 90 mph cars but only 5% of them could actually go that fast, wouldn’t that be deceptive for the other customers who cannot go 90 mph as the cars were sold? And if the car company attempted to charge extra, after the fact, so all cars could go 90 mph, wouldn’t that be deceptive as well?
If certain customers are indeed exceeding the maximum 4 Mbps capacity of the connection as originally sold, that’s a different question and perhaps they should pay more. If so, then that should be the starting, floor point for new, higher metered rates above 4 Mbps – not a retroactive starting point that penalizes existing customers who are just now starting to use the full speed of the connection as sold, up to 4 Mbps.
Another way to ask the question is, if Time Warner insists that “heavy users” should pay more, then why shouldn’t “light users” get a refund – or lower rate – for not using all of what they paid for in the first place?
Finally, it is essential to understand that Time Warner is a landline monopoly or duopoly in many places, which explains why it can casually announce price increases that violate existing contracts by raising prices rather than providing the capacity sold.
While not an explicit violation of net neutrality by direct, selected content manipulation, this policy could become a serious implicit violation of net neutrality by restricting certain high-volume uses through prohibitively high prices assessed retroactively in areas where there are no competitive alternatives.
The other problem is that this is a very blunt instrument. The p2p bandwidth hogs are largely a problem during peak usage times. In the middle of the night, light users are not affected by bandwidth hogs. A better approach would be to have a BW cap that only applies during certain times (e.g. any traffic between midnight and 8 AM is uncapped). This is a better way to optimizing the service experience for light users and bandwidth hogs.
I wonder if another way to look at this is for the broadband providers to charge ‘carriage fees’ to those ecommerce companies who increasingly use rich media as part of their offerings? Today, cable charges carriage to QVC, HSN, ShopNBC etc; cable pockets that money. Why should EBay, Amazon etc have a ‘free ride’ into the home when the other traditional home shopping company’s have to pay? Wouldn’t those broadband carriage fees be sufficient to fund the broadband providers networks?
Barry, that is the argument for Whitacre Tiering. You can see my economic rationale for opposing it here:
http://wetmachine.com/i…
Specific to ecommerce, with expanding bandwidth and video capability available via broadband, perhaps the TV ShopNets should abandon cable carriage all together and become broadband home shop nets. Together the big 3 TV home shopnets pay $500-$600M annually for carriage. They shut that off and watch local cable bills increase
Harold, your assertions that carriers have built “crappy networks” is way off base. Bandwidth costs money! And any network, no matter what its capacity, can be pushed to the limit by bandwidth hogging and P2P (which, as I mentioned in my testimony before the FCC, has no limits at all to its appetite for bandwidth). If a user wants bandwidth, he or she cannot expect the provider to sell it below cost. ISPs aren’t charities; we are hard working people who are, quite frankly, undercompensated rather than overcompensated for the long hours and hard physical labor we put in. (I expect to be climbing a half dozen roofs today alone, doing site surveys for wireless Internet.) And Time-Warner’s overage charges are not excessive, as anyone can calculate on the back of an envelope. (Our own ISP’s incremental cost for an additional gigabyte, at peak periods — which is when subscribers tend to stream movies — is about $6.) You wouldn’t expect the power company or the gas company to give you free product just because you self-righteously decided that you had a “right” to it. Why are you doing the same for bandwidth?