AT&T announced that it will withdraw its 5G home Internet product in New York rather than comply with the law that requires it to offer broadband rates as low as $15.
The law went into effect recently when the U.S. Supreme Court refused to hear the appeal for the New York law approved by the New York legislature in 2018. The Affordable Broadband Act requires ISPs to offer broadband rates to low-income households of no more than $15 for 15 Mbps (rumored to soon to be 100 Mbps) or $20 for 200 Mbps. Earlier this year, the 2nd U.S. Circuit Court of Appeals in Manhattan ruled that federal telecommunications law does not stop states from regulating broadband rates, and when the Supreme Court refused to review the case, the law went into effect.
It’s pretty extraordinary when a huge company like AT&T walks away from a state over reduced profits. The company has $122 billion in revenues for the year ending September 2024, and it’s impossible to believe that the company can’t afford to give a discount to a few of its customers.
What is most extraordinary about AT&T’s decision is that the company has been touting its FWA wireless technology as the replacement for customers who lose copper lines. The company announced recently that it intends to retire all copper nationwide by the end of 2029. This new announcement means AT&T is willing to tear down rural copper in New York and provide customers with no alternative. To some degree, they were going to do that anyway since many rural areas don’t have adequate cellular coverage to support the AT&T wireless product.
It’s hard to think that New York regulators won’t quickly react to AT&T walking away from existing FWA customers. This decision might ultimately cost the company more in fines than what it would lose from customer discounts.
It’s hard to see AT&T’s decision as anything other than a political decision and not a monetary one. The FWA products likely has high margins since there are no wires involved. AT&T could have reacted differently. For example, they could have required FWA customers to buy the wireless receiver, which would have eliminated the biggest cost of offering the service. This feels more like a warning to other states about implementing similar laws.
In a related story, Starlink wants an exemption from the new law. The company says it has less than 20,000 customers in the state – which would qualify for an exemption. Other ISPs, like Windstream, are asking for the same exemption. It’s a little hard to accept Starlink’s story of having less than 20,000 customers in the state since the company claims 1.4 million customers in the U.S., but if the company is under that threshold, it should be exempt.
What does that mean for Starlink’s future in the state? Will it permanently cap customers at 20,000 to stay out of the discount program? This seems like it would preclude Starlink from taking any New York BEAD funding to add more customers in rural areas.
One interesting feature of both FWA wireless and Starlink to consider is that these products don’t offer different speed tiers. The networks of both ISPs are not equipped to selectively choke customers to slower speeds, so a $15 customer would get the same bandwidth as everybody else. Both products deliver all-you-can-eat broadband, with no caps on speeds or the amount of broadband used during a month.
It will be interesting to see if other states consider putting the burden for customer discounts on ISPs. There is no guarantee that if this issue spreads to other states that the Supreme Court won’t decide to hear the issue.
Perhaps the biggest problem with the law is that it could bankrupt small ISPs that build expensive networks. 20,000 might seem like a lot of customers, but it’s not. The New York legislators seemingly picked the $15 rate of out the air with no consideration of what that might mean for ISPs. This feels like regulating with a sledge hammer rather than doing the research to develop a long-term solution that can work for both ISPs and customers.
AT&T isn’t an incumbent carrier in New York, is it? I believe Verizon/Frontier had the copper wire infrastructure back from the pre-divesture days.
Given that as a premise, along with your other points about Starlink and Windstream, this appears to be a text book example of government price control creating supply shortages.
for sure. It also creates a bit of a trap for verizon etc in that they still have some mandate to provide DSL on the old copper plant but can’t meet gov. standards for broadband on it, and at discounted rates can’t/wont upgrade it.
I don’t feel at all sorry for Verizon, don’t get me wrong, but I believe strongly that this model hurts consumers on the whole AND hurts the low income people it ‘helps’.
Low income internet service should be handled via the existing low income support models. Providers shouldn’t be involved in this, shouldn’t have to devote resources to vet low income, shouldn’t be demanding that data from end users in the first place, and shouldn’t be ‘avoiding’ low income markets that they’d have to lose money in.
$15-20/m is a loss, and this is a business.
There’s virtually no qualification for ‘low income’ and if AT&T were to deny people they could face legitimate fines. That makes this a wide open wound to lose money without recourse.
I highly doubt AT&T can actually be fined to ceasing operatings in a state by that state. That would be quite a leap. fining a company for not operating in your state… or not offering specific services in your state.
It might be an interesting reason to revisit Wickard v. Filburn which has caused all kinds of mischief since it was decided.
IDK, might be a stretch to use the commerce clause for low income services and states don’t have access to the commerce clause to apply an interstate economic effect law. I see basically zero overlap in these things.
We are building four rural towns in NY state right now. When they are done, we are done in NY. We will never have more than 19,900 customers in NY state because everyone will want the $15 deal and how can I have my CSRs determine eligibility? Use a third party for another month or two of losses? At $15/month, I can’t pay the interest on the money I borrowed to build fiber in front of that home. The principal will never be repaid. NY already has the highest cost to build of any state we have encountered. This will ensure that rural broadband networks, which are more expensive to build per customer, will never be built. The rural areas that need better broadband the most are the ones that have the higher concentrations of people at the bottom of the economic range. These are depressed towns with little industry and few jobs. They need our investment more than anywhere else and now they are guaranteed to not get it.
“It’s hard to see AT&T’s decision as anything other than a political decision and not a monetary one.”
Clearly making a policy statement here. As regulation shifts to states in the absence/unlikelihood of affirmative federal regulation and states impose Title II style regulation to increase access, reliability and affordability, there will be industry push back.
I’m still going to refute the ‘not a monetary one’
There seems to be a very strange idea that these services and this tech has no cost associated. $15 literally doesn’t cover a single customer support call, it doesn’t cover one call in for someone to pay their late bill (we’re talking low income here, be realistic). Even assuming all the 5G gear and fiber and backhauls and tower rents etc were all free $15 doesn’t make sense.
I mean, eggs are almost that price (jokes!!!… sort of).
Leaving behind many thousands of paying customers to make some political statement seems like something that would get a CEO fired for violating corporate law, shareholders come first not politics.
I believe Doug’s point is this is for a relatively small portion of the potential market based on household means. I agree such a low rate would not likely support the overall capex and opex.
“I believe Doug’s point is this is for a relatively small portion of the potential market based on household means.”
That might be true if it was true, but what’s the qualification process here? AT&T doing a means assessment?
I can tell you confidently that people in 1/2 million dollar homes were getting ACP service because there was no effective qualification process or validation process, and it would be exceptionally hard to do so.
Basically, if you claim you can’t afford it, you’re in. Any guidelines to the contrary can’t really be proved.
So how small is the portion of services that have to be delivered at a loss and what right does the state have to force them to sell a single service at a loss? Sounds a whole lot like a 4th amendment issue to me.
https://broadbandbreakfast.com/isps-press-supreme-court-to-reconsider-n-y-internet-affordability-law/?ref=alerts-newsletter
I suggest that states start requiring that electricity costs have a max cap of $0.10 kwh, vehicles have a max purchase price of $25,000, rent can’t be more than $800 / month, and gas not over $2.50 / gallon.
Those are things required to live in 2025. Why pick on just ISP’s. $15 / month? Is someone out of their freaking mind? One hour, at minimum wage, per month, and 100% of your gross income from that client is gone. They are free for that month. If it wasn’t so sickening that this idea actually made it to the breathing stage of life it’d be laughable.
*only if they say they are poor*
I certainly can’t afford a brand new $80,000 truck so I’m loving this $25k cap idea!
~
What’s going to happen if they actually fully enforce this, will you get a $200 fee if you call in ? How bad MUST the service be if you provide it at a loss? Or do they really expect a couple hundred dollars in customer equipment plus a couple hours of labor to install with a 22 month ROI assuming there are no service calls or phone calls or anything?
This is why there was a red wave. People didn’t move ‘right’, they ran away from these out of touch laws that are so hostile to business AND society.
Is there a way to validate low income? ACP had a validation process, while perhaps not perfect, was standardized and pretty simple to implement.
Regarding those that say AT&T cannot make money on these customers, I will point you to the fact folks like Mint Mobile offers $15 per month mobile plans and seem to be able to make it work. I do not cry for companies like AT&T that have happily taken billions in public money if they are forced to give back a bit.
ACP’s validation was a joke.
Mint mobile has a 3 month introductory offer, that’s not the rate. That’s also limited limited to 480p and a certain amount of data before dropping to 3G speeds. Ie, there’s backpressure to prevent heavy use.
Home broadband averages 10-50x more data and it uses that data during high use times, as in it’s one of the drivers for high use, in fact the mint mobile type plans are often off-peak use because people are anxious to get on home wifi to preserve their data.
It’s not remotely comparable services.
Someone can qualify for the low-cost plan in NY if they are in any of these programs:
* Free or reduced-priced lunch through the National School Lunch Program
* Supplemental Nutrition Assistance Program
* Medicaid
* Senior citizen rent increase exemption
* Disability rent increase exemption
* Affordability benefit from a utility
I’d be curious to know the number of households that would qualify for the Affordable Broadband Act (ABA) in New York under that means testing. And, what impact might that have on broadband availability in low-income areas where there would be a higher percentage of households that qualify. Could an ISP pull out of those areas only? Those are areas where they would be losing money on a high percentage of accounts.
Under the the means testing for the federal ACP program about 55 million households would have qualified. When the ACP ended there were 23 million signups. There are about 130 million households in the U.S, so 17% of all households were getting the subsidy, and that number was growing. The FCC had just started paying groups to promote the ACP and to help guide people on how to sign up for it. If every household that was eligible applied, 42% of the households in the U.S. would eventually be getting the subsidy. That is four out of every ten households in the U.S. That’s certainly not sustainable. AND, not long before the program ran out of money the FCC passed new rules that would have increased the subsidy for those in high-cost areas from $30 to $75.
Of course, a huge difference with the NY law and ACP is that the federal government was paying the ISPs a subsidy. The ISPs loved it, because they got more customers, and the government paid them for each one, so they got the same revenue. It was a win-win for everyone except for the U.S. taxpayers who were the ones actually footing the bill. Six out of every ten households were paying for the broadband of the other four. And a high percentage of households on the ACP signed up for a plan that cost more than the ISPs low-cost plan, so they could have afforded a $30 plan, but instead got a higher tier plan at a $30 discount.