The story moving through the media press this week is that cable television has stopped dying.
Pay TV fell from roughly 100 million American households in 2016 to about only just over 60 million now, the rate of decline is easing, and analysts think the business settles somewhere north of 50 million by 2030. All of that is accurate. The part almost nobody is saying out loud is that the 62 million figure is not cable subscribers, and the stabilization everyone is describing is not cable stabilizing.
Here is the number underneath the number. MoffettNathanson’s second-quarter count puts the sector at 61.51 million subscribers, and it breaks down like this:
- Cable video: 27.5 million
- Virtual bundles such as YouTube TV, Hulu + Live TV, Fubo and Sling: 21.42 million
- Satellite: 9.56 million
- Telco TV: 3.01 million
Traditional cable video, the thing the headlines are calling cable TV, is 27.5 million homes. That is roughly a quarter of the 2016 peak, not two thirds of it. More than a third of the surviving bundle is people who already cancelled cable and bought a streaming replacement, and they are being counted in the total that is used to argue cable has found its footing.
The Quarter’s Good News Is an Arithmetic Illusion
The improvement everyone is citing is real and it points the other way. The sector lost 885,000 subscribers in the second quarter, a clear improvement on the 1.34 million it lost a year earlier. But traditional pay TV alone shed 953,000 over the same three months. Subtract one from the other and the virtual bundles added roughly 68,000 net customers, which is the only reason the headline loss looks smaller than the underlying one.
Put plainly: traditional distributors lost more subscribers than the industry did. The gap between those two numbers is the entire good news story. YouTube TV alone is estimated to have picked up 50,000 in the quarter, and it is close to the only line on the page that grows.
Craig Moffett, whose note produced the 50 million floor that generated all of this coverage, was not coy about the mechanism. His projection holds, in his words, as long as growth from virtual providers can mostly offset the declines in traditional ones. He also expects the composition to keep flipping, from 38 percent virtual at the end of this year to 49 percent by 2030. That is not a forecast that cable steadies. It is a forecast that by the end of the decade the bundle is close to half streaming, and that the total holds because one half is eating the other.
This Is a Change of Landlord, Not a Recovery
We think calling this a floor is a category error, and a convenient one. What the data describes is a transfer of the American television bundle from Comcast, Charter and the satellite operators to a small set of technology companies, with Google the largest single beneficiary. The product is similar. The channel lineup is similar. The bill is often similar, which is the part cord-cutters keep rediscovering. What changes is who owns the pipe, who sets the price, and who holds the customer relationship.
That is a bigger consumer story than cord-cutting ever was, and it is getting almost none of the attention, because “cable decline slows” is a tidier headline than “the company that owns YouTube is becoming the largest pay TV distributor in the country.” The complaints people had about the cable bundle, which were that it was expensive, that it bundled channels they did not want, and that a single company had too much leverage over what they could watch and what it cost, are not resolved by moving the bundle to a firm that also owns the world’s dominant video platform and the advertising system attached to it. They are concentrated.
The pattern is visible in the individual operators too. Charter’s losses have narrowed sharply, and we looked at Spectrum’s position in June when the video business was still bleeding customers on both the TV and broadband side. Slower losses at a company with 27.5 million industry peers is not evidence that the product got better. It is evidence that most of the people who were going to leave have left, and the remainder are the ones with a reason to stay.
The Reason They Stay Is Sports, and That Is Not Permanent Either
The retention argument holds up, as far as it goes. Live sports are the last thing the bundle does that substitutes poorly, and most marquee American leagues still require a pay TV subscription of some form. That is genuinely why a floor exists at all.
It is also the least stable part of the thesis. Rights packages have been steadily migrating to Amazon, Netflix, Apple and YouTube itself, and each migration moves a load-bearing wall out of the traditional bundle and into the virtual one. A floor that rests on sports is a floor that rests on the assumption that sports rights stop moving, and there is no evidence for that assumption. Cable penetration has already sagged to about 32 percent of American TV homes, a figure that would have been unthinkable during the years when the bundle’s grip on live sports was the industry’s whole answer to streaming.
The analysts numbers are sound and their floor may well arrive. It makes the coverage wrong, because a floor for “pay TV” defined to include YouTube TV tells you almost nothing about the business people mean when they say cable, and quite a lot about who is going to be collecting the money in 2030.
The question worth tracking is not whether pay TV stops shrinking. It is whether anyone regulating or reporting on this notices that the bundle everyone spent fifteen years escaping is being reassembled, at similar prices, by fewer and larger companies than the ones they left.