Greenfield

An important Wall Street analyst has publicly written what many have thought offline for the past six months — the chances of regulators approving a merger of Comcast and Time Warner Cable are growing less and less each day that passes.

Rich Greenfield from BTIG Research has grown increasingly pessimistic about the odds of Comcast winning approval of its effort to buy Time Warner Cable.

Despite the unified view from the executive suites of both cable companies that the merger is a done-deal just waiting for pro forma paperwork to get handled by the FCC and Department of Justice, Greenfield has seen enough evidence to declare “the tide has turned against the cable monopoly in the past 12 months,” and now places the odds of a merger approval at 30 percent or less.

“Since we realized the inevitability of Title II regulation of broadband in December 2014, we have grown increasingly concerned that Comcast and Time Warner Cable will not be allowed to merge,” Greenfield wrote.

The claim from both cable companies that since Comcast and Time Warner Cable do not directly compete with each other, there in no basis on which the government could block the transaction, may become a moot point.

There are three factors that Greenfield believes will likely deliver a death-blow to the deal:

• Monopsony Power
• Broadband Market Share & Control
• Aftershocks from the Net Neutrality Debate

Monopsony power is wielded when one very large buyer of a product or service becomes so important to the seller, it can dictate its own terms and win deals that no other competitor can secure for itself.

Comcast is already the nation’s largest cable operator. Time Warner Cable is second largest. One would have to combine most of the rest of the nation’s cable companies to create a force equally important to cable programming networks.

As Stop the Cap! testified last summer before the Public Service Commission in New York, allowing a merger of Comcast and Time Warner Cable would secure the combined cable company volume discounts on cable programming that no other competitor could negotiate for itself. That would deter competition by preventing start-ups from entering the cable television marketplace because they would be at a severe disadvantage with higher wholesale programming expenses that would probably make their retail prices uncompetitive.

Even worse, large national cable programming distributors could dictate terms on what kinds of programming was available.

The FCC recognized the danger of monopsony market power and in the 1990s set a 30% maximum market share limit on the number of video customers one company could control nationally. That number was set slightly above the national market share held by the largest cable company at the time — first known as TCI, then AT&T Broadband, and today Comcast. Comcast sued the FCC claiming the cap was unconstitutional and won twice – first in 2001 when a federal court dismissed the rule as arbitrary and again in 2009 when it threw out the FCC’s revised effort.

Comcast itself recognized the 30% cap as an important bellwether for regulators watching the concentration of market power through mergers and acquisitions. When it agreed to buy Time Warner Cable, it volunteered to spin-off enough customers of the combined company to stay under the 30% (now voluntary) cap.

Greenfield argues the importance of concentration in the video programming marketplace has been overtaken by concerns about broadband.

“While Comcast tried to steer the government to evaluate the Time Warner deal on the old paradigm of video subscriber share, it is increasingly clear that DOJ and FCC approval/denial will come down to how they view the competitive landscape of broadband and whether greater broadband market share serves the public interest,” Greenfield wrote.

If the Comcast merger deal ultimately fails, the company may have only itself to blame.

Last year Comcast faced intense scrutiny over its interconnection agreements with companies that handle traffic for large content producers like Netflix. Comcast customers faced a deterioration in Netflix streaming quality after Comcast refused to upgrade certain connections to keep up with growing demand. Netflix was eventually forced to establish a direct paid connection agreement with the cable operator, despite the fact Netflix offers cable operators free equipment and connections for just that purpose.

That event poured gasoline on the smoldering debate over Net Neutrality and helped fuel support for a strong Open Internet policy that would give the FCC authority to check connection agreements and ban paid online fast lanes.

Seeing how Comcast affected broadband service for millions of subscribers across dozens of states could shift the debate away from any local impacts of the merger and refocus it on how many broadband customers across the country a single company should manage.

Comcast will control 50% or more of the national broadband market when applying the FCC’s newly defined definition of broadband: 25/3Mbps.

That rings antitrust and anticompetitive alarm bells for any regulator.

Greenfield notes that changing the definition of broadband will dramatically reshape market share. It will nearly eliminate DSL as a suitable competitor and leave Americans with a choice between cable broadband and Verizon FiOS, community owned fiber networks, Google, and a small part of AT&T’s U-verse footprint. If those competitors don’t exist in your community, you will have no choice at all.

Even Comcast admits cable broadband enjoys a near-monopoly at 25/3Mbps speeds. controlling 89.7% of the market as of December 2013.

“If regulators take the ‘national’ approach to evaluating broadband competition, the FCC’s redefinition would appear to put the deal in even greater jeopardy,” Greenfield writes. “Beyond the market share of existing subscribers, the larger issue is availability.  Whether or not a current subscriber takes 25/3Mbps or better, the far more relevant question is if a consumer wanted that level of speed do they have a choice beyond their local cable operator?  As of year-end 2013, Comcast’s own filing illustrates that in 63% of their footprint post-Time Warner Cable, they were the only consumer choice for 25 Mbps broadband (we suspect even higher now).”

“With Comcast’s scale both before and especially after the Time Warner Cable transaction, they become ‘the only way’ for a majority of Americans to receive content/programming that requires a robust broadband connection,” Greenfield warned.

Even worse, to protect its video business, a super-sized Comcast will be tempted to introduce usage caps that will deliver a built-in advantage to its own services.

“Over time, the fear is that Comcast will favor its own IP-delivered video services versus third parties, similar to how it is able to offer Comcast IP-based video services as a ‘managed’ service that does not count against bandwidth caps, while third-party video services that look similar count against bandwidth caps,” he wrote. “The natural inclination will be for [Comcast] to protect their business (think usage based caps that only apply to outsiders, peering/interconnection fees, etc.)”

“With the overlay of the populist uprising driving government policy, it is hard to imagine how regulators could approve the Comcast Time Warner Cable transaction at this point,” Greenfield concludes. “Comcast continues to try to get the government to look to the past to get its deal approved.  But the framework is about not only what is current, but what the future will look like – especially in a rapidly changing broadband world.”

Neil Smit, CEO, Comcast Cable (left), Ryan Lawler, TechCrunch (right)

Of all the tech companies to turn up at TechCrunch’s Disrupt New York 2014 event, Comcast Cable seemed the least likely to qualify as the kind of innovative start-up TechCrunch loves to cover.

But there sat Comcast Cable CEO Neil Smit with TechCrunch’s Ryan Lawler, discussing Comcast’s mega-merger with Time Warner Cable, its peering agreement with Netflix, broadcast TV streamer Aereo, and Comcast’s legendary dismal customer service.

Smit’s arrival on stage to a smattering of tentative applause was a clear sign there was no love for the cable giant in the audience, particularly from many New York area Time Warner Cable customers dreading a future with Comcast.

Smit was immediately confronted with the fact Comcast was recently voted the Worst Company in America by Consumerist readers, prompting yet another promise that improving customer service was Comcast’s “top priority,” the same promise Comcast gave in 2007, 2008, 2009, 2010, 2011, 2012, and 2013.

“I think if there’s one thing to disrupt in our business, it’s customer service,” Smit added.

Smit defended Comcast’s merger with Time Warner, relying heavily on video subscribers to downplay the concentrated market power Comcast would have after the merger. Smit pointed out Netflix has the largest subscriber count of any pay television channel or platform and denied Lawler’s contention that a merger would give Comcast more than 50% of the American broadband market.

“I think the number is a little less than that — it is closer to 40% but if you include wireless than it would be less than 20%,” Smit responded, referring to the LTE 4G wireless networks from wireless carriers that come with very low usage caps and very high prices.

Smit also promised major broadband speed upgrades and other improvements for Time Warner Cable customers, but nobody mentioned Comcast’s gradual reintroduction of usage caps on residential broadband accounts.

Comcast Cable’s CEO also addressed several other hot button issues:

Smit claimed Comcast has a good working relationship with the FCC and is providing advice on whatever changes to Net Neutrality FCC chairman Tom Wheeler will propose later this month.

Despite the fact Comcast could ultimately benefit if Aereo is found to be legal by the U.S. Supreme Court, Smit recognized Comcast also owns NBC and other broadcast programmers and was concerned about the economic impact if cable operators stopped paying for over-the-air programming.

“We pay $9 billion a year for content,” Smit said. “One of the things that I question in the Aereo solution is: are they paying for content? The spend for that content has to come from somewhere.”

Smit also noted Comcast is increasingly targeting younger audiences by signing deals with college campuses to bring Comcast service to students to hook them as future subscribers. Comcast is also creating new packages with fewer channels to appeal to millennials. Smit also acknowledged many younger family members are accessing cable programming using passwords associated with their parent’s cable account.

[flv]http://www.phillipdampier.com/video/TechCrunch Interview with Neil Smit 5-6-14.mp4[/flv]

Here is the complete interview TechCrunch conducted with Comcast Cable CEO Neil Smit. (22:20)

Level 3’s global network: Orange lines represent Level 3-owned infrastructure, yellow lines show leased or co-owned connections.

Five of the largest Internet Service Providers in the country are intentionally sabotaging your use of the Internet by allowing their network connections to degrade unless they receive extra compensation from content companies they often directly compete with.

Mark Taylor, vice president of content and media for Level 3, wrote a lengthy primer on how Internet providers exchange traffic with each other across a vast global network. While clients of Level 3 are likely to have few problems exchanging traffic back and forth across Level 3’s global network, vital interconnections with other providers that make sure everyone can communicate with everyone else on the Internet are occasional trouble spots.

Every provider has different options to reach other providers, but favor those offering the most direct route possible to minimize “hops” between networks, which slow down the connection and increase the risk of service interruptions. These connections are often arranged through peering agreements. Level 3 has 51 peers, minimized in number to keep traffic moving as efficiently as possible.

This oversaturated port in Dallas cannot handle all the traffic trying to pass through it, so Internet packets are often dropped and traffic speeds are slowed.

Taylor writes most peering arrangements were informal agreements between engineers and did not involve any money changing hands. Today, 48 of the 51 Level 3 peering agreements don’t involve compensation. In fact, Level 3 refuses to pay “arbitrary charges to add interconnection capacity.” Taylor feels such upgrades are a matter of routine and are not costly for either party.

Peering agreements have been a very successful part of the Internet experience, even if end users remain completely in the dark about how Internet traffic moves around the world. In the view of many, customers don’t need to know and shouldn’t care, because their monthly Internet bill more than covers the cost of transporting data back and forth.

Because of ongoing upgrades the average utilization of Level 3’s connections is around 36 percent of capacity — busy enough to justify keeping the connection and providing spare capacity for days when Internet traffic explodes during breaking news or over the holidays.

However, Taylor says more than a year ago, something suddenly changed at five U.S. Internet Service Providers. They stopped periodic upgrades and allowed some of their connections to become increasingly busy with traffic. Today, six of Level 3’s 51 peer connections are now 90 percent saturated with traffic for several hours a day, which causes traffic to degrade or get lost.

“[The] congestion [has become] permanent, has been in place for well over a year and […] our peer refuses to augment capacity,” Taylor wrote. “They are deliberately harming the service they deliver to their paying customers. They are not allowing us to fulfill the requests their customers make for content.”

Taylor adds all but one of the affected connections are U.S. consumer broadband networks with a dominant or exclusive market share. Where competition exists, no provider allows their Internet connections to degrade, said Taylor.

Taylor won’t directly name the offenders, but he left an easy-to-follow trail:

“The companies with the congested peering interconnects also happen to rank dead last in customer satisfaction across all industries in the U.S.,” Taylor wrote. “Not only dead last, but by a massive statistical margin of almost three standard deviations.”

Taylor footnotes the source for his rankings, the American Consumer Satisfaction Index. The five worse providers listed for consumer satisfaction:

• Comcast
• Time Warner Cable
• Charter Communications
• Cox Communications
• Verizon

AT&T has also made noises about insisting on compensation for its own network upgrades, blaming Netflix traffic.

In fact, Netflix traffic seems to be a common point of contention among Internet Service Providers that also sell their own television packages. They now insist the streaming video provider establish direct, paid connections with their networks. Level 3 is affected because it carries a substantial amount of traffic on behalf of Netflix.

Ultimately, the debate is about who pays for network upgrades to keep up with traffic growth. Taylor says Level 3’s cost to add an extra 10Gbps port would be between $10-20 thousand dollars, spare change for multi-billion dollar Americans cable and phone companies. Normally, competition would never allow a traffic dispute like this interfere with a customer’s usage experience. Angry customers would simply switch providers. But the lack of competition prevents this from happening in the United States, leaving customers in the middle.

This leaves Taylor with a question: “Shouldn’t a broadband consumer network with near monopoly control over their customers be expected, if not obligated, to deliver a better experience than this?”

Netflix has reached an agreement with Verizon Communications for a paid-peering interconnection between the two that will help assure Verizon’s broadband customers can watch Netflix content without repeated buffering slowdowns.

It is the second major deal for Netflix where money has changed hands to assure a more error-free experience for customers, coming two months after a similar deal with Comcast.

Both Verizon and AT&T have told their respective shareholders they anticipate new revenue streams from interconnection agreements with Netflix, which now constitutes a large percentage of evening Internet traffic in the United States.

By establishing a more direct connection between Netflix and Verizon, customers should experience fewer streaming problems. Netflix traditionally pays third parties to handle its traffic and some of those shared connections have grown increasingly overcongested. ISPs are becoming increasingly reluctant to upgrade them without financial compensation.

The FCC does not consider peering arrangements part of the Net Neutrality controversy, but for customers the result is the same — without a financial arrangement with a provider, content from certain websites may not be dependably accessible. Earlier this year, Netflix viewing was increasingly disrupted for many Verizon, AT&T, and Comcast customers.

Netflix CEO Reed Hastings has repeatedly complained Netflix is being subjected to an “arbitrary tax” to assure dependable service to its paying customers. Hastings now wants the FCC to treat the issue as an “end run” around Net Neutrality and include peering agreements and financial compensation issues in the new Net Neutrality rules expected to be introduced in May.

Most analysts do not expect FCC chairman Thomas Wheeler to oppose or regulate the financial arrangements between ISPs and content producers.

Netflix kicks off the summer rating the streaming video performance of some of America’s largest ISPs, and the results deliver only a few surprises.

Google Fiber is the runaway leader, but Verizon FiOS — also a fiber-based network — is lagging behind several cable operators, notably Charter Cable and Suddenlink.

In mid-June, GigaOM reported Verizon was engaged in a battle with Cogent — a bandwidth provider Netflix depends on to help it reach Verizon customers.

Cogent promptly blamed Verizon for the slowdown:

Cogent and Verizon peer to each other at about ten locations and they exchange traffic through several ports. These ports typically send and receive data at speeds of around 10 gigabit per second. When the ports start to fill up (usually at 50 percent of their capacity), the internet companies add more ports. In this case, through, Verizon is allowing the ports that connect to Cogent to get crammed. ”They are allowing the peer connections to degrade,” said Dave Schaeffer, chief executive officer of Cogent said in an interview. “Today some of the ports are at 100 percent capacity.”

“Think of it as the on-ramp to the freeway being log-jammed,” Schaeffer said. And that means your Netflix content, especially content sent by Netflix’s content delivery network, slows down, and you get pixelated pictures and buffering.

Verizon just as quickly shot back at GigaOm and Cogent:

Cogent is not compliant with one of the basic and long-standing requirements for most settlement-free peering arrangements: that traffic between the providers be roughly in balance. When the traffic loads are not symmetric, the provider with the heavier load typically pays the other for transit (see our ex parte filing [PDF] from the 2010 Comcast/Level 3 spat for more info on peering and transit agreements). This isn’t a story about Netflix, or about Verizon “letting” anybody’s traffic deteriorate. This is a fairly boring story about a bandwidth provider that is unhappy that they are out of balance and will have to make alternative arrangements for capacity enhancements, just like any other interconnecting ISP.

Customers don’t care. They just know their efforts to watch Arrested Development are being stymied, and Netflix’s June ISP results illustrate the degraded performance customers are getting.

Cablevision, the top performing cable operator, can likely thank its recent investments in network upgrades for improved performance, not its participation in Netflix’s OpenConnect Content Delivery Network, designed to improve streaming performance for participating ISPs. Cablevision is a member, but so are Frontier Communications (#14) and Clearwire (#17 and dead last).

OpenConnect couldn’t help Frontier DSL or Clearwire wireless customers achieve good results — the technology in use and the upstream connections both companies maintain with the Internet backbone mattered much more.

In general, fiber performs best when everyone is getting along, cable comes in second, DSL third, and wireless last.

But if you want the best performance possible, and Google Fiber is not in your neighborhood, your best bet is to move to Sweden, where the top-six providers all outperformed every American cable, DSL, and wireless provider. In Finland, the top-four beat everyone but Google Fiber. The nine best-rated ISPs in Denmark also outclassed their American counterparts, while in Norway a half-dozen providers did better.

But many ISPs in the United States can still be proud: the top eight beat Mexico. Mediacom, AT&T U-verse & DSL, Bright House, CenturyLink, Windstream, Frontier, Verizon DSL and Clearwire have some work to do… if they want to keep up with those speed mavens in Guadalajara.