Why the status quo in the U.S. cable business can't hold

The more that I look at the U.S. cable, satellite and IPTV business, the more I realize that "business as usual" is eventually doomed. Cable companies' core business for more than 50 years has been to sell access to bundles of broadcast and cable-only channels, which are accessed through the use of proprietary set-top boxes, to consumers. Starting in the late 1990s, cable operators started adding access to high-speed Internet services, which ride in and out of consumers' homes on available bandwidth not used for video. A few years later, cable operators added Voice over IP telephony services, which use the same bandwidth as high-speed Internet. IPTV companies offer the same service, but in the reverse order: First came analog voice telephony, more than 100 years ago. Then, DSL came in the 1990s for high-speed Internet service, and finally, AT&T, Verizon and others added broadcast and cable-only television channels, accessible through proprietary set-top boxes.

Today's cable and IPTV operators look very similar so far as consumers are concerned, and they both face the same business challenges: Retransmission and carriage fees. Retransmission fees are intended to compensate broadcasters for the use of their programming by video operators. Carriage fees provide compensation to cable network operators. It used to be that some cable networks would pay video operators to carry their programming, in order to sell advertising that would reach the widest possible audiences. Today, however, almost all cable networks charge video operators to supply their programming to consumers.

Until 2008's Great Recession, broadcasters and cable networks got most of their revenues from advertising. Broadcasters kept their retransmission fees low, or waived them altogether if video operators agreed to carry cable channels provided by the broadcasters' parent companies. Cable networks generally also kept their carriage fees relatively low, in order to get into the widest possible number of households. After 2008, however, all that changed. Broadcasters' advertising revenues dropped (in some cases, dramatically,) so they needed to make up for lost income. In addition, broadcast networks, which had been paying television stations to carry their programming, began charging stations for programming or demanded a portion of the stations' retransmission fees. Similarly, cable networks started increasing their carriage fees to replace lost advertising revenues.

Early on, video operators absorbed the price increases from content providers as best they could, knowing that they couldn't pass the increases on to customers in the form of higher rates during a recession. Now, however, not a week goes by where a video operator isn't threatening to drop a broadcast station or cable network because it's too expensive, or a broadcaster or cable network isn't threatening to cut off a video operator. Video operators are trying to disguise consumer rate increases as things like "concierge" services, where they charge for services that consumers used to get for free. And today, Cablevision filed an antitrust lawsuit against Viacom, charging the company with forcing cable operators to license a bundle of 14 unpopular cable networks in order to get access to popular ones such as Comedy Central and Nickelodeon.

This situation can't persist for much longer. In many markets, subscription prices have reached the maximum that consumers are willing to pay, and consumers have gotten wise to video operators' pricing tactics: Offer low "teaser" rates to get consumers to switch, and then start raising rates frequently, and often silently, once their introductory deals expire. Consumers respond by cancelling services, switching video operators, and in the worst case, dropping video services altogether and switching to over-the-air broadcasts and over-the-top Internet video.

Within a decade, I believe that most cable and IPTV companies will be well on the way to dropping their video services. Consumers will purchase their own set-top boxes, and similar functionality will be built directly into new televisions. Some set-top box vendors will also aggregate content. Rather than the plethora of formats for publishing video that work on set-top boxes from Apple, Google, Intel, Roku, etc., a single standard protocol will enable content providers to publish channels and on-demand video that will work with most set-top boxes, and will show up in the devices' program guides. Cable and IPTV companies are likely to partner with set-top box vendors and receive a portion of their revenue from consumer subscriptions.

Consumers would get the "a la carte" cable channel choices that they've been asking for--but at a price. For example, Disney's ESPN might make its primary ESPN channel available by itself to subscribers for $6.95/month--but price the entire ESPN channel lineup at $12.95/month, thus making it more attractive to pay more but get everything. This strategy would work for the rest of Disney, as well as Viacom, CBS, Discovery, Fox, NBC Universal and others.

The cable and IPTV operators would compete on other services and benefits--who offers the fastest and most reliable high-speed Internet service, the simplest and most useful home networking, the best home automation and security packages, etc. All of these would be services that the cable and IPTV operators would provide themselves--thus, they wouldn't be subject to ever-increasing financial demands from cable networks and broadcasters. By literally wiring their services deep into households, it would be much harder for consumers to switch from one service provider to another, which should decrease churn levels.

The war among video operators, broadcasters and cable networks, with consumers in the middle and paying the bills, can't go on for much longer. At some point, a critical mass of consumers will stop paying the bills, and video operators will have no choice but to spin off their video services.

Part 1: Birth, Death and Transformation

We’re in the midst of a massive shift in media consumption patterns. People consume more news than they ever did, but they don’t read newspapers anymore. Magazines, even on tablets, are slowly dying. And, as for books, The New Yorker published an article titled “Twilight of the Books”…on December 24, 2007, before eBooks were even a significant part of the business. Statistics in the article show that the market for books has been declining for at least 30 years. U.S. movie theater ticket sales peaked in the 1950s; the only things that have kept the industry going have been home video sales and higher ticket prices. But, home video sales are also dropping—they’re being replaced by rentals from Redbox, and online streaming from Netflix, Amazon and others.

Let’s be clear: Movie attendance has been declining for half a century, but no one seriously expects the movie business to disappear. The same is true for books; readership will continue to decline, but it’s hard to visualize a world without books, even if most of the remaining books are digital instead of paper. Nevertheless, the balance has shifted. Consumers want their media faster and cheaper. Readers want their news from the Internet, as it happens (if not sooner, leaked out via Twitter.) One can argue that attention spans have gotten shorter—look at the popularity of viral videos on YouTube—but videogames, both casual and complex, can engross players for hours or even days.

The transformation of media in the 21^st Century is being driven by three forces: The Internet, mobile devices and wireless broadband. The Internet provides a conduit for every kind of content. There’s no need to ever leave your house to purchase any kind of media, and it makes possible entirely new types and combinations of media that didn’t exist prior to the rise of the World Wide Web. Mobile devices and wireless broadband make that content available anywhere, anytime, and open the digital world to hundreds of millions of people who can’t afford personal computers or high-speed Internet connections.

Episode 4 of the Feldman File videoblog is live!

It's Sunday night, and that means that I've posted a new episode of the Feldman File videoblog on YouTube! Here's the rundown for this week's edition:

• Apple's less-than-earthshaking announcement about adding the Beatles' music catalog to iTunes
• Sony follows up on its Super 35MM camcorder, the PMW-F3, with yet another Super 35MM camcorder, the 35MM NXCAM
• U.S. cable operators lose 741,000 subscribers in Q3--are consumers really cutting the cord?
• The Obama Adminsitration is looking for 500MHz of additional broadband bandwidth, and the National Telecommunications and Information Administration found 2.2GHz of bandwidth available within ten years. So, do we really have a bandwidth shortage?

CableLabs announces 5Gbps cable data technology

Last week, Alcatel-Lucent's Bell Labs announced that it had developed a new technology that would allow DSL high-speed Internet speeds to reach 300Mbps at distances up to 400 meters, and 100Mbps at a mile. These are hypothetical numbers, and the real-world numbers are likely to be quite a bit lower. Nevertheless, a number of analysts got good and frothy about it; one said that it could "reshape the whole next-generation broadband competitive environment."

The problem is that the competitors aren't standing still. Today, CableLabs, the U.S. cable industry technology consortium, announced that it's working on a new, very preliminary technology that will enable downstream data speeds as high as 5Gbps--more than 16 times faster than the new technology announced by Bell Labs. The CableLabs technology is a radical rethink of how to send data over coax, and isn't compatible with the current DOCSIS standard. That means in order to use the new approach, cable operators would have to throw out all their cable modems and much of their head-end data hardware.

However, even the current DOCSIS architecture still has a lot of headroom; Cisco and Broadcom have demonstrated a DOCSIS 3.0 cable modem with a maximum downstream speed of 300Mbps. According to Broadcom, such a device could download a two-hour-long HD movie in approximately two minutes.

Of course, once you can download a two-hour HD movie in two minutes, do you really need any more speed? Probably not today, but new applications always arise that take advantage of whatever bandwidth is available. That's why the high-speed Internet services will continue to leapfrog each other; you can never be too rich or have too much bandwidth.

Reed Hundt did something "naughty" with the HDTV Transition

You may remember Reed Hundt. He was the Chairman of the FCC under Bill Clinton, at the time that agency was planning for the transition to HDTV. The original plan was to implement HDTV in the same way as Japan had done it--one resolution, one refresh rate, one single standard. The FCC instead mandated a convoluted standard that required some 17 combinations of resolutions and refresh rates to be supported; even what we now consider Standard Definition was included in the HD standard, in order to support data applications over broadcast bandwidth. Chairman Hundt invited Microsoft into the process, and Microsoft demanded a series of changes to the standard in order to support its plans. What was originally a clean, straightforward and reasonably inexpensive upgrade became complex and expensive for everyone--broadcasters, cable and satellite operators, and consumers.

Earlier this month, at the Columbia Business School, Mr. Hundt gave a speech where he admitted that he and his associates deliberately logjammed the HDTV transition with the intention of killing it entirely. Here's a direct quote: "This is a little naughty...we delayed the transition to HDTV, and fought a big battle against the whole idea but we lost." If Mr. Hundt and his associates were so dead-set against HDTV, why didn't they fight it out in the open instead of waging a passive-aggressive war of changing the specifications and moving the goalposts? If he and his associates had been honest, we wouldn't need this National Broadband Plan--there would be plenty of bandwidth for broadcasters and for broadband wireless services. Instead, we've got the world's most expensive and convoluted HDTV system, many consumers still aren't getting true HD, and tons of bandwidth is lying fallow in subchannels mandated by the FCC.

We have to dramatically increase the availability of broadband access at prices that U.S. consumers can afford. As I've written previously, the National Broadband Plan says precious little about how it's going to lower costs to consumers, and puts few demands on the incumbent telephone and cable providers. It's biased to take its "pound of flesh" from broadcasters. Is what we're seeing a real attempt to make broadband available to everyone at an affordable price, or is it merely the continuation of Mr. Hundt's now almost 20-year-long war against broadcasters?

The most important thing left out of the National Broadband Plan

The Executive Summary of the National Broadband Plan was released today by the FCC (PDF link), and what's been proposed is largely in line with what was leaked in the last few weeks. The FCC wants to get affordable broadband service with 100 Mbps download and 50 Mbps upload speeds into at least 100 million homes by the end of this decade. The Plan proposes lots of ways to get there, from subsidizing expansion of wired Internet connections in rural communities to reallocating 500MHz of wireless bandwidth to broadband service. However, for all the platitudes in the plan about lowering costs and making broadband service more affordable, there's precious little (at least in the executive summary) proposed to actually back up that intent.

A few years ago, when I was an industry analyst covering the IPTV market, I regularly traveled to Europe. My visits to France were particularly enlightening. There were multiple service providers offering IPTV, high-speed Internet and local & long distance phone service at the equivalent of around $30 a month; less than one-third the price of the $99 triple-play deals offered in the U.S., with similar or better channel selections and Internet speeds. As you can imagine, the French services were wildly popular.

So why was triple-play service so much less expensive in France (and so much better)? The French Government required France Telecom to make its lines available to competitors at wholesale prices. Far from slowing down the growth of broadband access, the French Government's decision caused usage of broadband services to explode for everyone, including France Telecom.

Telephone companies in the U.S. were once required to provide their lines available to competitors on a wholesale basis, and the U.S. had a healthy competitive market for DSL services. Cable operators, on the other hand, were never required to make their networks available to competitors. In the Clinton Administration, the same team that drafted the National Broadband Plan allowed telephone companies to stop making their networks available to competitors as part of the Telecommunications Act of 1996, ostensibly to encourage phone companies to make bigger investments in their networks and further the growth of broadband.

We saw what happened: Most of the phone companies that were originally part of AT&T got consolidated back into AT&T. As a practical matter, most U.S. households have a choice of broadband service from two suppliers--the incumbent telephone company and cable operator. There's no meaningful price competition. U.S. households pay much more for much slower Internet service than do households in many other countries.

I'm not saying that the National Broadband Plan doesn't have merit, or that it shouldn't be taken seriously. What I am saying is that the thing that's most likely to increase competition, lower costs and make more broadband access available is to open up the existing cable and telecom networks for wholesale availability. That idea is nowhere to be found in the National Broadband Plan, and there's no surprise why.

Update, March 22, 2010: The Berkman Center for Internet & Society at Harvard University did an extensive study of broadband in the U.S. and the rest of the world for the FCC, and released the report, "Next Generation Connectivity" in February, 2010. The report speaks to the points that I made in this blog post, plus much more. Click here to access the entire study, or visit this article at the New York Times for a summary of the relevant points.

Taking the competition to the next level

A few posts ago, I wrote about the difficulty that IPTV operators are having with differentiating their services from cable, satellite, and (in some markets) over-the-air broadcasters. Triple-play packages (landline telephone, high-speed Internet and video) are old news in most markets, and cable operators are using VoIP to compete head-to-head with the telcos. Quadruple-play packages, adding mobile service, are less common, and are more difficult for non-telcos to compete with. In order to get access to mobile services, cable and satellite operators usually have to resell services from a telco, which puts them at a pricing disadvantage.

In most places, even when you buy a triple- or quadruple-play package, you get a bundle of services that don't talk to each other. Consumers purchase on the basis of price and features; brand loyalty doesn't exist. In this situation, you can get a runaway "race to the bottom", as is happening in several countries in Western Europe, with France being the best example. There, operators are piling on more and more features while maintaining the price at 30 Euros a month. Packages that would sell in the U.S. for $99 a month or more are going for the equivalent of under $45 at current exchange rates.

The real opportunity is in what I call silo-busting--tearing down the walls between services in order to unlock consumer value and provide the opportunity to increase prices (or at least maintain prices in the face of competition.) Telcos are just now starting to let consumers get Caller ID on their television when the phone rings. Instead of running to the phone when it rings, your television can tell you who's calling, so that you can make the decision of whether or not to take the call. If that service has value to you, and the competition doesn't offer it, you're more likely to stay with your service provider. Let's take it to the next step: Add a speakerphone to the set-top box's remote control, and you can take the call without picking up the phone. Now the television and the phone service are tightly linked. That adds value and increases differentiation.

Let's take a quad-play example: PCCW in Hong Kong enables consumers to look up movie showtimes and buy tickets, right from their televisions. The tickets are sent to their mobile phones in the form of a barcode. At the theater, the barcode is scanned for admission. PCCW can do this because they control all the elements. They're now the biggest seller of movie tickets in Hong Kong. They not only generate transaction fees every time they sell a movie ticket, they offer a desirable service that their competitors can't match.

Or consider a location-based service that ties the television and mobile phones together: A parent can see where her children are by plotting the position of their GPS-enabled mobile phones on a map on her television. It's a service that non-telco competitors can't match, and it's of considerable value to a section of the market.

The last point has to do with paying for these new services. In highly price-competitive markets, there's a fear that telcos won't be able to increase their prices, and these new services will simply get sucked into consumer expectations. Most video providers, whether IPTV, cable or satellite, package their services into tiers: For the basic price, you get a basic tier. If you want more channels, you have to buy another tier at a higher price. If you want premium channels, you pay even more. Consumers are familar and comfortable with this model. The integrated (or converged) services that I'm proposing can be priced into service tiers, just like video programming. Service providers can offer a basic service at a low price to retain subscribers, and offer unique services on tiers to bring in more revenue.

In short, I believe that the real key to unlocking value is to integrate services through applications. As consumers see the power of tying these services together, they'll migrate to the service providers that let them do the most, not just to the providers that are the cheapest.

Comcast limits bandwidth

Comcast announced today that starting October 1st, it will limit bandwidth usage for its high-speed Internet customers to 250GB/month. The first time that a subscriber exceeds the limit, they'll get a warning; if they do it again within six months, their high-speed Internet service could be turned off for a year. That's right, a year.

It's no surprise that Comcast is implementing bandwidth caps. The company was penalized by the FCC for interfering with BitTorrent traffic, so it's looking for alternative ways of limiting its network load. However, its decision raises two issues: First, is 250GB/month an appropriate limit, and second, how will consumers measure their bandwidth use to make sure that they don't use more than the maximum?

Comcast's press release gives examples of what can fit into 250GB: 50 million emails or 124 standard-definition movies. The problem, of course, is that people don't use their Internet connections for only one purpose, such as email--they use them for many different things. Someone who uses an online backup and restore service for their hard disk could use up most of their monthly limit in one session. If you use Vonage, Skype or some other VoIP service, that's going to count against your monthly limit. (One assumes that you'll be able to use Comcast's own VoIP service as much as you want, however.) And, if you've got something like a Slingbox or AppleTV, you could use as much as 15MB per minute of video. So, 250GB could get used up very quickly.

That brings us to the second issue: Comcast is providing no way whatsoever for subscribers to see how much bandwidth they've used. They recommend that customers install bandwidth monitoring software on their computers. That's all well and good for PC applications, but it won't track usage by a TiVo, Squeezebox, Slingbox, Vonage VoIP adapter, or similar devices. Mobile phone companies can tell subscribers their phone usage down to the second--why can't Comcast provide a webpage that tracks subscribers' bandwidth usage? After all, they have to be measuring it in order to enforce their 250GB limit.

Perhaps Comcast thinks that so many people will complain about this limit that they'll get the FCC to agree to content-based throttling. I think that it's more likely that the reverse will happen--Comcast will tick so many people off that the FCC will once again intervene and force the company to adopt a much higher limit. When subscribers who are using their Internet connections for perfectly legitimate purposes start to see their service cut off, the feces will hit the fan.