4 ms·

Ajit Pai tries to kill San Francisco’s attempt to spur broadband competition

54 points by JaimeThompson 7y ago

deogeo 7y ago
> the Pai plan argues that ISPs "are less likely to invest in deployment" if they know they have to share network components with other providers. "We know any residence is unlikely to have broadband from multiple ISPs, yet we will continue to claim there is competition among ISPs."
sytelus 7y ago
This guy should be the prime target for journalist under cover investigation. Bribery is written all over his face. If you ever wanted to earn Pulitzer Prize for great journalism, this guy is a gift for you.
- cylinder 7y ago
  I think it's worse; he's actually an ideological fanatic.
rayiner 7y ago
The article distorts what the dispute is actually about: > The ordinance only applies when the inside wiring belongs to the property owner. Under the rule, property owners who have outfitted their buildings with Internet wiring cannot deny access to ISPs, making it harder for them to strike exclusive deals with Internet providers. If you read the NPRM (https://docs.fcc.gov/public/attachments/DOC-358068A1.pdf https://docs.fcc.gov/public/attachments/DOC-358068A1.pdf, starting at paragraph 11), you’ll see that you’re not necessarily talking about scenarios, as the article implies, where the building owns the wires because it paid to install the wires. In many scenarios, the building owner technically owns the wiring, but the service provider made the original investment in wiring the building and entered into a contractual arrangement to recoup the initial investment. Consider the Interstate 66 express lanes into DC. In that project, private investors paid to build the new freeways, but the State owns the road. What the investors got in return is the exclusive right to operate the toll system for 50 years, plus a cut of the toll revenue. If you eliminate the exclusivity provision as anti-competitive, you’ve totally overturned the original bargain. Maybe that sort of deal was a bad idea to begin with and shouldn’t have been allowed. But overturning those contractual arrangements after the fact creates real problems. (And note that no government entity pays or subsidizes companies to wire up apartment buildings in San Francisco, so that angle isn’t relevant here.) The solution is probably, as a matter of building codes, to require MDUs to be wired with Ethernet.
- jrnichols 7y ago
  thank you for pointing that out. it really seems that arstechnica purposely takes the negative road when there's anything related to the FCC or Pai in the news, which is disappointing to see happen.
- sagarm 7y ago
  > But overturning those contractual arrangements after the fact creates real problems. Can you expand on this a bit? Clearly there's a trade-off being made here. There are real benefits to ensuring that ISPs actually face competition. The main argument against is around property rights, but as you stated the wires are owned by the building owner. ISPs are potentially less likely to "subsidize" infrastructure in exchange for exclusivity, but IMO that's a win. Apartment owners shouldn't be able to undermine competition in this related but separate market.
- wahern 7y ago
  > In many scenarios, the building owner technically owns the wiring, but the service provider made the original investment in wiring the building and entered into a contractual arrangement to recoup the initial investment. Exactly, and that's why it's anti-competitive. Those deals permit ISPs to effectively purchase customers from developers. The future customer-residents have no say, and because of substantial information asymmetry in practice customer-residents aren't in a position to judge the relative costs of renting a building locked into a contract versus renting in another building. It's obvious on its face the dominate reason to enter these contracts is to extract additional profit from renters and distribute it to the developer, landlord, and the ISP. There are situations where exclusivity might be a necessary condition to make a service viable, but decades of evidence has disproven this when it comes to ISPs. As someone was quoted saying in the article, San Francisco has $40/month 1Gb/s fiber. Sonic.net didn't demand exclusivity to provide that. Some background facts in SF: for years AT&T's plan was to install FTTC (fiber-to-the-cabinet) throughout the city. Sonic.net decided to do the same--also install their own cabinets around the city using the same regulatory approval AT&T would get. Once AT&T got wind that Sonic.net was also going to do this, they began to drag their feet in the regulatory process and actively began to sabotage it. Sonic got frustrated and decided to take a real business risk, scrapped their own FTTC plan, and began to plan and build-out FTTH throughout San Francisco. This completely caught AT&T off-guard. Within substantially less time than AT&T spent petulantly dragging its feet to avoid facing competition, Sonic.net had built-out fiber to a substantial fraction of the city, charging only $40/month.[1] Eventually AT&T scrapped their own FTTC plan and started building FTTH, except AT&T is doing it haphazardly and charges twice per month what Sonic charges. We shouldn't be surprised that it was the pursuit of profits in a competitive environment that spurred Sonic and then AT&T to aggressively roll our fiber, rather than the promise of guaranteed, monopoly profits. (EDIT: I forgot to point out here that the SF ordinance was passed because AT&T and Comcast were already beginning to use exclusivity agreements to prevent Sonic.net and Monkey Brain from expanding their service offerings, particularly among the existing housing stock, not just the big, new apartment towers in SOMA and Mission Bay. Sonic.net had to keep their FTTH plans secret for as long as they could because those are the kinds of strategies AT&T and Comcast use to "compete". Thankfully the SF ordinance closed off one such avenue.) Policymakers like Pai have claimed for years that exclusivity arrangements are necessary and that government should refrain from prohibiting, or should even affirmatively support exclusivity, based on a theory that it's necessary to incentivize telecom investment. But the reason this is necessary, if at all, is because people like Pai continue to hold out the promise of these exclusivity arrangements. Pai and his ilk are literally manufacturing the opportunity costs upon which their arguments are premised. The promise of being able to achieve exclusivity is why AT&T spent years obstructing the regulatory process in SF. More to your point, your I-66 example is not comparable. AFAIU the SF ordinance doesn't prohibit existing exclusivity arrangements. I'm pretty sure that would be unconstitutional as states are prohibited from interfering with existing contractual rights. The SF ordinance was only prospective. And, yes, as a theoretical matter such "interference" in the so-called free market could lead to unintended, negative consequences. But it's unlikely to do that, and much more likely to have the salutary impact intended. In a free market nobody wants to compete; everybody is looking for a way to avoid competition. Among other ways to do this, market actors exploit information asymmetry to capture more profits than would be possible if their customer had full information. That has manifestly been the case when it comes to typical ISP exclusivity arrangements. As someone else said elsethread, Pai is a fanatic--he's committed to a theory about the way the telecom market works despite evidence to the contrary. Municipalities do dumb things and pass a lot of dumb regulations with unintended consequences. (Sonic loudly complained that SF wouldn't permit microtrenching, so for now they say it's only cost effective to build out FTTH to the parts of city using with utility poles, which fortunately is more than half, IIRC.) But simply because that happens doesn't mean all such regulation is bad. Such policies need to be considered on a case-by-case, sector-by-sector basis, and reassessed as the facts play out. It's exhausting but the only proper thing to do. There are no easy answers. [1] To be fair $40 is the teaser rate. The long-term rate is... wait for it... $50/year.

54 points by JaimeThompson 7y ago

> the Pai plan argues that ISPs "are less likely to invest in deployment" if they know they have to share network components with other providers. "We know any residence is unlikely to have broadband from multiple ISPs, yet we will continue to claim there is competition among ISPs."

sytelus 7y ago

This guy should be the prime target for journalist under cover investigation. Bribery is written all over his face. If you ever wanted to earn Pulitzer Prize for great journalism, this guy is a gift for you.

cylinder 7y ago
I think it's worse; he's actually an ideological fanatic.

rayiner 7y ago

The article distorts what the dispute is actually about: > The ordinance only applies when the inside wiring belongs to the property owner. Under the rule, property owners who have outfitted their buildings with Internet wiring cannot deny access to ISPs, making it harder for them to strike exclusive deals with Internet providers. If you read the NPRM (https://docs.fcc.gov/public/attachments/DOC-358068A1.pdf https://docs.fcc.gov/public/attachments/DOC-358068A1.pdf, starting at paragraph 11), you’ll see that you’re not necessarily talking about scenarios, as the article implies, where the building owns the wires because it paid to install the wires. In many scenarios, the building owner technically owns the wiring, but the service provider made the original investment in wiring the building and entered into a contractual arrangement to recoup the initial investment. Consider the Interstate 66 express lanes into DC. In that project, private investors paid to build the new freeways, but the State owns the road. What the investors got in return is the exclusive right to operate the toll system for 50 years, plus a cut of the toll revenue. If you eliminate the exclusivity provision as anti-competitive, you’ve totally overturned the original bargain. Maybe that sort of deal was a bad idea to begin with and shouldn’t have been allowed. But overturning those contractual arrangements after the fact creates real problems. (And note that no government entity pays or subsidizes companies to wire up apartment buildings in San Francisco, so that angle isn’t relevant here.) The solution is probably, as a matter of building codes, to require MDUs to be wired with Ethernet.

jrnichols 7y ago
thank you for pointing that out. it really seems that arstechnica purposely takes the negative road when there's anything related to the FCC or Pai in the news, which is disappointing to see happen.
sagarm 7y ago
> But overturning those contractual arrangements after the fact creates real problems. Can you expand on this a bit? Clearly there's a trade-off being made here. There are real benefits to ensuring that ISPs actually face competition. The main argument against is around property rights, but as you stated the wires are owned by the building owner. ISPs are potentially less likely to "subsidize" infrastructure in exchange for exclusivity, but IMO that's a win. Apartment owners shouldn't be able to undermine competition in this related but separate market.
wahern 7y ago
> In many scenarios, the building owner technically owns the wiring, but the service provider made the original investment in wiring the building and entered into a contractual arrangement to recoup the initial investment. Exactly, and that's why it's anti-competitive. Those deals permit ISPs to effectively purchase customers from developers. The future customer-residents have no say, and because of substantial information asymmetry in practice customer-residents aren't in a position to judge the relative costs of renting a building locked into a contract versus renting in another building. It's obvious on its face the dominate reason to enter these contracts is to extract additional profit from renters and distribute it to the developer, landlord, and the ISP. There are situations where exclusivity might be a necessary condition to make a service viable, but decades of evidence has disproven this when it comes to ISPs. As someone was quoted saying in the article, San Francisco has $40/month 1Gb/s fiber. Sonic.net didn't demand exclusivity to provide that. Some background facts in SF: for years AT&T's plan was to install FTTC (fiber-to-the-cabinet) throughout the city. Sonic.net decided to do the same--also install their own cabinets around the city using the same regulatory approval AT&T would get. Once AT&T got wind that Sonic.net was also going to do this, they began to drag their feet in the regulatory process and actively began to sabotage it. Sonic got frustrated and decided to take a real business risk, scrapped their own FTTC plan, and began to plan and build-out FTTH throughout San Francisco. This completely caught AT&T off-guard. Within substantially less time than AT&T spent petulantly dragging its feet to avoid facing competition, Sonic.net had built-out fiber to a substantial fraction of the city, charging only $40/month.[1] Eventually AT&T scrapped their own FTTC plan and started building FTTH, except AT&T is doing it haphazardly and charges twice per month what Sonic charges. We shouldn't be surprised that it was the pursuit of profits in a competitive environment that spurred Sonic and then AT&T to aggressively roll our fiber, rather than the promise of guaranteed, monopoly profits. (EDIT: I forgot to point out here that the SF ordinance was passed because AT&T and Comcast were already beginning to use exclusivity agreements to prevent Sonic.net and Monkey Brain from expanding their service offerings, particularly among the existing housing stock, not just the big, new apartment towers in SOMA and Mission Bay. Sonic.net had to keep their FTTH plans secret for as long as they could because those are the kinds of strategies AT&T and Comcast use to "compete". Thankfully the SF ordinance closed off one such avenue.) Policymakers like Pai have claimed for years that exclusivity arrangements are necessary and that government should refrain from prohibiting, or should even affirmatively support exclusivity, based on a theory that it's necessary to incentivize telecom investment. But the reason this is necessary, if at all, is because people like Pai continue to hold out the promise of these exclusivity arrangements. Pai and his ilk are literally manufacturing the opportunity costs upon which their arguments are premised. The promise of being able to achieve exclusivity is why AT&T spent years obstructing the regulatory process in SF. More to your point, your I-66 example is not comparable. AFAIU the SF ordinance doesn't prohibit existing exclusivity arrangements. I'm pretty sure that would be unconstitutional as states are prohibited from interfering with existing contractual rights. The SF ordinance was only prospective. And, yes, as a theoretical matter such "interference" in the so-called free market could lead to unintended, negative consequences. But it's unlikely to do that, and much more likely to have the salutary impact intended. In a free market nobody wants to compete; everybody is looking for a way to avoid competition. Among other ways to do this, market actors exploit information asymmetry to capture more profits than would be possible if their customer had full information. That has manifestly been the case when it comes to typical ISP exclusivity arrangements. As someone else said elsethread, Pai is a fanatic--he's committed to a theory about the way the telecom market works despite evidence to the contrary. Municipalities do dumb things and pass a lot of dumb regulations with unintended consequences. (Sonic loudly complained that SF wouldn't permit microtrenching, so for now they say it's only cost effective to build out FTTH to the parts of city using with utility poles, which fortunately is more than half, IIRC.) But simply because that happens doesn't mean all such regulation is bad. Such policies need to be considered on a case-by-case, sector-by-sector basis, and reassessed as the facts play out. It's exhausting but the only proper thing to do. There are no easy answers. [1] To be fair $40 is the teaser rate. The long-term rate is... wait for it... $50/year.