Does cable really have a 97% profit margin?

Font Size:

Let’s face it, everyone loves bashing their cable company. It’s become an American pastime.

So it’s not surprising that the Internet exploded with cable hatred when the MIT Technology Review’s David Talbot claimed last week that “cable distribution giants like Time Warner Cable and Comcast are already making a 97 percent margin on their ‘almost comically profitable’ Internet services.”

The quote spread like wildfire and quickly made it to the front page of Reddit the next day. Predictably, comments poured in decrying the cable companies.

But the claim is false. The “97 percent margin” assumes cable infrastructure materializes out of thin air, ready for broadband use and requiring no upgrades. It’s another misleading statistic all too easily accepted by those who insist that cable is a rapacious monopoly requiring public utility regulation.

Talbot cited the highly respected telecom analyst Craig Moffett as his source. But Moffett was referencing “gross profit margin” (GPM), a statistic that doesn’t take into account the infrastructure-heavy models of broadband providers. Talbot conveniently left out Moffett’s next sentence: in which he said that “this is not as crazy as it first appears,” and explained GPM’s pitfalls.

GPM is a simple equation: (Revenue – Cost of Goods Sold)/Revenue. It works well for calculating the profit margin on selling tangible goods like televisions and computers where the Cost of Goods Sold (COGS) is easy to calculate.

But GPM doesn’t properly evaluate the cost calculation for any business model dominated by high infrastructure costs, like broadband because the costs are not included.

The cable industry has invested nearly $200 billion in building networks since 1996, and it continues to invest more than $10 billion more each year in maintenance and upgrades. Yet cable’s COGS takes into account only the much lower day-to-day costs of running the network as opposed to building it. This produces GPMs that make cable companies’ profits look artificially high.

GPMs exaggerate broadband profits in a second, subtler way. Cable companies provide not only broadband but voice and video services as well–over the same networks. Many of their costs, such as network maintenance and customer service, can’t be broken down by service. These costs are thus considered operating costs for the entire business, and likely were left out of the equation when determining the gross profit of providing broadband services alone. This, again, overstates the margin of revenue over costs.

“Return on invested capital” (ROIC) paints a much more accurate picture of how the cable industry is performing. Unlike the gross profit numbers, ROIC actually attempts to incorporate long-term investment in infrastructure, giving a better sense of how a company is using its money to generate returns.

Using ROIC, we find that until very recently cable companies were earning small returns, still trying to recover their colossal initial investments. It often takes years of positive profits for these companies to make up for that initial investment and start seeing a return. So, what may appear to be a massive annual profit using GPM is really just recoupment of a tiny piece of past costs. Returns for cable companies range from negative to quite small.

Comcast, supposedly the greatest cable monopolist, has averaged just a 4.5% ROIC over the last five years. Time Warner Cable’s 5-year average is -1.3%. Compare those with Apple (32%) or Google (16.1%).

If cable companies were really making 97% profit, they would be among the most profitable businesses in history. Other companies and investors—sitting on huge stores of cash today—would be rushing to compete with cable, or simply funding the expansion of Verizon and AT&T’s fiber networks.

An immediate influx into the market is not happening because the amount of capital required to build, upgrade and maintain a cable network tilts the calculus. Others are testing the waters, like Google’s fiber project in Kansas City, but even it isn’t expected to recoup its investment for several years.

As broadband analyst Dave Burstein pointed out, about Google’s fiber network, “The problem is it costs a lot of money to climb all those poles and dig all those trenches to make it happen. You don’t make money in three years, but you make money in 10 years.”

In short, a more careful look at Craig Moffett’s work would have revealed that cable isn’t exceptionally profitable. But that wouldn’t have fit into Talbot’s narrative: that cable companies are already so profitable, they have no incentive to improve their service to match high-speed offerings like the Google Fiber deployment in Kansas City.

The truth is much more complicated. Cable companies just spent billions deploying DOCSIS 3.0 technology to increase their download speeds from 38 to over 100 megabits per second. They’re now trying to recoup that investment.

As Talbot notes, Google claims that its fiber service is profitable today. But it’s not clear how many consumers would pay the higher rates Google would have to charge if it were required to connect every home in its franchise area, as cable companies must do. Most consumers don’t seem willing to pay for the faster speeds Google Fiber offers now, as there are few compelling applications for such high speeds.

Cable companies aren’t holding back because they’re just raking it in today, but because cable franchising law bars them from deploying ultra-high speed services selectively to those willing to pay for them. That’s a harder story to write, but it’s the real answer to the question posed in Talbot’s headline: “When Will the Rest of Us Get Google Fiber”—or something like it?

Matt Starr and Will Rinehart are fellows at TechFreedom, a dynamist tech policy think tank.