[orthecreedence]

> There are four major wireless carriers, none with more than about a third of the market.

Can that be said for wired internet carriers? And can that be said at the municipal level and not just the national level? My point is that in many areas, people generally don't have a choice in who their internet provider is. So if you look nationally, yes, the coverage is split between several companies, but at the local levels the areas serviced are often carved up such that there are local monopolies.

[mjevans]

Where I live, in a suburb well within the metro area around the city of Seattle (WA), I am LUCKY to have even one provider that offers at least 25Mbit (per second, down). They thankfully offer 1 Gigabit down packages (very recently, YEARS behind when Seattle got this 'city wide').

HOWEVER, there is ZERO competition so:

    * data-caps
    * horrendous over-charges
    * UPLOAD (on a 1000Mbit line) ~= 30 Mbit

Yes, that's right, the upload is about ONE THIRTIETH, a very small fraction, of the download. Even the BUSINESS plans don't go much higher (while costing tons more, and really only having a support that will say sorry and give a small credit with no actual rush on support; in past experience).

So very much no. The wired-line sector is a natural monopoly that should be like roads to the airport. I should pay a very small bit in taxes for good connections to the global hub, and at said global hub I should have my choice of services from many providers.

[StudentStuff]

70% chance if your in Centurylink's territory they'll offer you 1 Gig for $65 a month (no contract, free modem, price for life, no BS fees) that you can sign up for here: https://www.centurylink.com/home/

If Frontier is your RBOC, they have 300Mbps for $50 iirc in 90% of the non-rural areas they serve. Check https://frontier.com/order-online/plan/package

Comcast is hot garbage, both the offerings I mention are fiber, and the gaps seem to be caused by poor enforcement of the franchise agreements in the relevant locality, eg: Frontier was supposed to have 100% fiber coverage in Kenmore a decade ago, but the city isn't consistently assessing them fines so there is no reason to meet the contract.

Seattle has no universal coverage mandate, but a complex customer income weighting metric to prevent redlining. IMO we negotiated a worse agreement than Kenmore: https://www.seattle.gov/tech/services/cable-service/cable-fr...

[mjevans]

Putting in the address of the HOUSE (that I am renting, but still a house in an OK area of town)...

"Speeds up to 12 Mbps are available in your area!"

[gremlinsinc]

wow, I'd think Seattle being a major tech hub would have better internet. I live in a farming community 15 minutes from the nearest small city (Google: Paragonah Utah), and we have few competitors but for $90 I get 1 gig fiber. There's not many places in utah where you can't get at least 100-300mb.

The state had the forsight to lay tons of conduit a decade or two ago whenever they do road work so that there's plenty of space for bandwidth upgrades and what not across the entire state.

[giobox]

If you’d expect that of Seattle, try looking at domestic broadband options in Silicon Valley.

For huge swathes you can have lovely ancient old ADSL on your copper phone line or Comcast via cable. Oh and all Comcast plans in the Bay Area, regardless of what the marketing says, have a 1 TB bandwidth cap and charge for overages up to 200 dollars on top of your monthly bill, unless you pay them an additional 50 bucks a month to get the _actually_ unlimited option on all of the plans offered. As you can imagine, many of us living here routinely blow through a terabyte of bandwidth - just one cloud connected security camera can eat a third of it easily.

That’s literally it for most people.

[rayiner]

I wonder how much tech peoples’ views are shaped by the fact that Seattle/the Bay Area have such awful broadband. (After all, broadband construction is mostly regulated at the municipal level, and so varies greatly depending on where you live.) Here in Maryland, 60% of people have access to fiber. Verizon in urban/suburban areas, and in rural areas, the state government is building a fiber backbone for municipal fiber networks to hook into. Almost everywhere also has cable. 4G coverage maps are a sea of color. For the most part, our housing prices also aren’t crazy, because we don’t have crazy anti-development/historical preservation laws. I don’t think these things are unrelated. We have pretty good municipal government here, outside Baltimore. I paid well under $500k for a 4BR house on the water, less than 30 minutes to four train stations, and with two fiber providers. Silicon Valley, by contrast, seems like some parody of American municipal mismanagement.

[ethbro]

I try to avoid reaching for comedy, but I couldn't read...

> Silicon Valley, by contrast, seems like some parody of American municipal mismanagement.

And not hear "You were saying something about best intentions?" from Jules in Pulp Fiction.

[dmode]

I am sure there are rural areas in Bay Area that have coverage issues, but definitely not true for urban areas. San Francisco itself has a ton of providers including Monkey Brains, Webpass, Sonic etc. I am on AT&T Fiber and it straight up 1gig up and down at all times.

[rconti]

If by "rural areas" you mean "not literally San Francisco proper", or, in other words, "where almost everyone actually lives".

[StudentStuff]

70% of the city proper has gigabit for $65 a month from Centurylink, OP is out in the fringes where they allowed VDSL2 deployment years ago (whereas Seattle was on ADSL), thus they missed out on upgrades in the last few years. Coverage map for the city proper: https://www.seattle.gov/tech/services/cable-service/cable-fr...

ex-Verizon areas that aren't rural generally have fiber coverage, excepting Finn Hill. They committed to a bunch of smaller suburbs to have 100% fiber coverage by 2008, which mostly got done. Sadly, they only go up to 300mbps symmetrical for $50/month iirc.

[rconti]

1.5 miles from Facebook HQ, in the middle of silicon valley, we JUST got FTTH a few weeks ago.

Cable internet is the only game in town in most of the valley unless you want 40mbit DSL.

[PaulHoule]

Big Tech is global, however, so it faces regulatory pressure from Europeans and other people who are more keen to regulate than the U.S.

Thus the U.S. feels pressured to keep up.

[khuey]

Seems like US regulators could just as easily decide that if the tech companies do anything really egregious the Europeans will smack them down so they don't need to do anything.

[TehCorwiz]

Let's take an example of Google being sanctioned and regulated for anti-competitive monopolistic practices abroad, but not broken up.

European or Asian (broad rhetorical buckets) regulations have no direct impact on US domestic regulations. Meaning that even if the EU held Google accountable they could continue to operate at home without changing their practices. While the argument can be made that regulation is likely to follow domestically I don't think that can be held as a truism under current geopolitical landscape.

Even if domestic regulations followed there would undoubtedly be a gap, during which companies would leverage that imbalance. Google would be pressured to make the most of its advantage in its shrinking monopolistic market while smaller companies would press their advantage abroad where they have better access to the market, investors, and customers.

Domestic inaction would be a net loss in the face of foreign regulation.

[rayiner]

Obviously there is more market concentration in telecom than say in frozen yogurt. But telecoms face competition (from DSL, cellular, and satellite) in almost every market. Comcast, for example, has 47% broadband market share even in its own footprint: https://www.bizjournals.com/philadelphia/news/2019/04/24/wha.... They also face competition across local markets. Verizon followed Google Fiber to the $70-80 for gigabit price point, even though it competed in no markets with Google.

[esoterae]

I have comcast service in one of the most dense urban environments in the US.

I have no alternative. I am being bled dry because there is no competition of similar service.

Please, if you're going to compare things at least stick to fruit. This is apples and basketballs.

[rayiner]

50% of folks in Comcast’s footprint don’t subscribe. Those folks aren’t going without Internet access. For a significant portion of the market, those other alternatives are viable. That creates competition, though obviously less competition than an equivalent competitor. (I’ve been lucky to mostly live in FiOS territory, and now I’ve got two fiber lines to my house. But back when I lived in Wilmington, I had just Comcast. Instead of subscribing, we used a T-Mobile hotspot the whole time. It was fine. With 5G, that phenomenon is only going to increase.)

[snowwrestler]

Pew reported in Jan 2018 that only 65% of adults are home broadband users. That's down from a peak of 73% in 2016. Segmenting by age, fully 50% of people 65 or older do not use broadband internet at home.

So what are they using? Mobile phones--Pew says 20% of adults do not have broadband at home, but own a smartphone. Or nothing--Pew reported that 11% of U.S. adults do not use the Internet at all.

https://www.pewinternet.org/fact-sheet/internet-broadband/

[esoterae]

And what % of people that actually have alternatives like FiOS (unlike me) don't subscribe to comcast? Saying 50% of subscribers in Comcast's footprint is prevaricating about the bush AF. Why, you ask? If 50% of potential Comcast subscribers have an alternative to Comcast, that would mean 100% of people with an alternative are taking advantage of the free market. Which would mean that 100% of the people remaining are likely forced to subscribe through a lack of access to alternatives.

I don't even have working voice cell coverage at my house, much less data. Again, almost certainly due to a an entrenched monopoly limiting the market outreach of competition, and thereby obviating the requirement that they themselves perform adequately enough to retain customers.

Q.E.D.

[rayiner]

For your hypothetical to hold, you’d expect to see some providers with near 100% market share. (You’d also have to assume Comcast builds and maintains infrastructure to millions of households in areas where it has no market share.) Who are they? FiOS’s market share is 40% in its footprint. AT&T’s fiber is around 25% now, hoping to reach 50% by 2023.

The only other way to get your math to work out is to assume that only a very small portion of Comcast’s footprint has no other cable/fiber competitor.

As to your cell situation—what mechanism do you think creates this competition-limiting monopoly? And why doesn’t that same mechanism apply to say my house (in the DC exurbs), where all of the big 4 have a decent signal? (Or the majority of the country, where at least three providers have coverage?)

[erikpukinskis]

So you’re saying the upper bound on the number of municipalities where Comcast has no competition is half of them?

Not saying much at all there.

[DuskStar]

What percentage of people in Comcast's footprint have broadband internet? 50%/55% is a rather different market share than 50%/95%.

[dillonmckay]

This is such a disingenuous argument.

DSL, cellular, satellite, cable broadband, and fiber broadband are not equivalent.

By your reasoning, as long as POTS exists, dial-up ISPs are a competitive threat.

[icebraining]

If changing the product is evidence of competition, then Google Search and Android must have some, since they are continuously being worked on and new versions released.

[rayiner]

Responding to a specific price point is the sign of competition, not general improvements in the product. AT&T built a gold-plated, state-of-the-art telephone network in the “Ma Bell” days, without facing competition. (And gave us the transistor and C and UNIX while they were at it. Google in many ways fits the same “benevolent monopoly” mold. I think it would be unwise to take antitrust action against Google for that reason.)

[esoterae]

You never used the old long-distance telephone system, did you? The only thing that was gold plated were the executive bathrooms at AT&T.

[silotis]

Verizon may have price matched Google, but Verizon fiber service is only available to a small fraction of the country. The cable companies, which have a much larger footprint than Verizon's fiber, have mostly limited price cuts to markets where they face competition from fiber. Comcast for example still charges $150 for gigabit (1000/35 Mbps) in most markets. By contrast markets were a competitor has built out fiber get the price cut in half. If this isn't a tacit admission that they are abusing their lack of competition by price gouging, I don't know what is.

[ashelmire]

> So if you look nationally, yes, the coverage is split between several companies, but at the local levels the areas serviced are often carved up such that there are local monopolies.

It's not just that they are monopolies, but that they are anticompetitive monopolies. In some cases, they are even granted by local governments.

[javiramos]

Yes - in my building I only have one provider option. And I pay $60/month for crappy internet.

[oneplane]

You'd think people would be up in arms about that. Restricting a 'building' to 1 provider seems ridiculous to me.

[partiallypro]

Wired internet carriers having monopolies are almost all because of local laws and agreements. The only way to bust it is to say those "rights of way" restrictions are illegal...which isn't going to happen.

[MagnumOpus]

> the only way to bust it is to say those "rights of way" restrictions are illegal

Or to impose compulsory sub-licensing of those wires at fair and reasonable terms defined by a government agency. Like other countries with wire monopolies are doing.

[rayiner]

Do you have concrete examples of these “local laws and agreements” that you can cite?

[partiallypro]

Yes, it's called local franchising. It's incredibly common. That is why most/all cities only have one cable provider and one telecom provider over land lines.

https://en.wikipedia.org/wiki/Cable_television_franchise_fee

https://www.rand.org/pubs/research_memoranda/RM6309.html

https://www.wired.com/2013/07/we-need-to-stop-focusing-on-ju...

https://consumerist.com/2014/05/10/why-starting-a-competitor...

https://www.mackinac.org/10118

https://en.wikipedia.org/wiki/Local_franchise_authority

[rayiner]

The Posner article is from 1970 when exclusive local franchises were legal. Congress outlawed them in 1992: https://transition.fcc.gov/Bureaus/OSEC/library/legislative_...

> SEC. 7. AWARD OF FRANCHISES; PROMOTION OF COMPETITION. (a) ADDITIONAL COMPETITIVE FRANCHISES.- (1) AMENDMENT.-Section 621(a)(1) of the Communications Act of 1934 (47 U.S.C. 541(a)(1)) is amended by inserting before the period at the end the following. ; except that a franchising authority may not grant an exclusive franchise and may not unreasonably refuse to award an additional competitive franchise. Any applicant whose application for a second franchise has been denied by a final decision of the franchising authority may appeal such final decision pursuant to the provisions of section 635 for failure to comply with this subsection.

The 1996 Telecom Act also mandated non-discriminatory competitive access to poles and rights of way: https://pubs.naruc.org/pub/635DF852-FCB3-B3A2-D381-9A307381F....

Most big cities--which have the density to have multiple viable wire-line competitors--don't have competition for the same reason they have high housing prices. They make it onerous to build infrastructure. Baltimore, for example, would be a great Google Fiber city. It has city-owned conduits everywhere that would make deployment relatively easy. But nobody wants to build fiber there, because the city insists that every neighborhood has to be wired up, without regard to potential demand: https://www.wypr.org/post/why-comcast-one-and-only-cable-and...

> “We’ve, in fact, asked other cable operators if they’re interested in coming into the city and, so far, nobody else is,” says Minda Goldberg, a chief solicitor in the city’s Law Department.

https://www.wired.com/2014/06/holding-back-high-speed-intern...

> But don’t pop the champagne. Many factors could derail the progress. One is a complaint sometimes raised when Google, in particular, starts negotiating with a community. The complaint? The new investment will create a digital divide because Google does not commit to connect every neighborhood. The argument, while sometimes well-intentioned, ignores history, economics, and the reality of the digital divide. Moreover, its proponents fail to acknowledge the consequences of their arguments.

> As we have seen in Gig.U negotiations, must-build requirements make projects unsustainable. Communities can wish for anything they want but if the economics don’t work, it won’t happen. Recently, Los Angeles put out a request for proposal (RFP) seeking a guaranteed citywide fiber build out. Broadband policy expert Harold Feld correctly noted, “I look forward to their RFP for a unicorn supplier, because I think it's about as likely under these terms.”

[paulddraper]

The FCC tracks this (well, they track availability, not market share). [1]

98% of census blocks have multiple ISPs offering 10Mbps+

82% of census blocks have multiple ISPs offering 25Mbps+

(This doesn't count wireless offerings.)

[1] https://docs.fcc.gov/public/attachments/DOC-355166A1.pdf

[jedberg]

Yeah, but census blocks are too big. Also, if one person in a census block has that level of access, the entire block is counted. The ISPs know this and game it. They will make their service available to one house in all the blocks that neighbor a block they actually serve, so they can get five blocks instead of one.

[dillonmckay]

Exactly. There is legislation pending at the federal level to gain better insights.

‘Measuring the Economic Impact of Broadband Act’ was introduced in the senate recently.

[paulddraper]

There are 6 million inhabited census blocks, averaging 52 residents per block. [1]

Even the larger ones still only have a few hundred people (and those are dense, apartment-heavy areas where multiple ISPs are common anyway).

I don't doubt there is some polishing the numbers, but it's still a very granular measure.

[1] https://en.wikipedia.org/wiki/Census_block

[liara_k]

Unfortunately, the FCC's tracking data for broadband availability is notoriously poor.[1]. Think "regulatory capture" and, if you'e seen The Wire, "juking the stats".

[1]: https://www.theverge.com/2019/5/30/18644726/fcc-broadband-re...

[paulddraper]

There are -- I am sure -- legitimate criticisms of the report, but that link isn't doing those arguments any justice.

> But the 2019 version of the report has come with an unusual amount of political baggage

What? Why baggage? How? Zero explanation, just FUD.

> The census block system relies on data that only shows where providers could easily offer service, not where they actually do.

Yes, that's how it works. Also why every ISP I know charges a connection fee: they're not going to connect unless you first agree to be a customer.

> Starks’ statement questioned the figures. “It’s incredible to me that an error this large — approximately 62 million in overstated broadband connections — didn’t materially change the report,” he said.

BarrierFree is a New England service provider [1], in urban centers where customers have numerous internet options. [2] Without altogether leaving their geographic regions, there's simply no way for their results to affect the FCC report. Probably why their mistake initially went unnoticed.

> Starks points out in his statement that some blocks are larger than 250 square miles.

Maybe 0.01% of the time. No wonder this man is perplexed at statistical averages. Look at any census map and you'll see that census blocks are very focused.

And to top it all off, the article includes zero suggestions for improving the admittedly imperfect measurement. Comes off as just a hit piece.

[1] https://broadbandnow.com/BarrierFree

[2] https://businessinternet.com/barrierfree

[jedberg]

> And to top it all off, the article includes zero suggestions for improving the admittedly imperfect measurement. Comes off as just a hit piece.

It's really easy to fix, but just more work.

How about the actual number of homes that can get high speed data? The ISPs already know this to the single address, because when you go to their website to sign up, they tell you right there if you're qualified based on your address.

All they have to do is be forced to share that database with the FCC.

[liara_k]

The FCC has a map app.[1] Go and plug in some random addresses.

The link below is for a NJ suburb , and lists 6 ISP competitors. But three of them are just satellite services and two are Verizon. In practice, there are exactly two competitors here: the phone monopoly and the cable monopoly, but the FCC would point to this to say that there's robust competition!

Here's another one I just picked at random, in a suburb of Oklahoma City.[2] The FCC lists 9 different providers. 3 of them are Satellite, 2 are Cox Cable, and 4 are AT&T. Again, there are exactly two competitors here: the phone monopoly and the cable monopoly. Look around yourself.

Two long-term competitors are not enough for meaningful competition. Duopoly competition has the same strategic profile as the Prisoner's Dilemma. When you have the same two players in the game over a long period of time, they learn to cooperate. See Axelrod (1980) etc...

Regulatory capture, juking the stats, crooks and liars all.

[1]: https://broadbandmap.fcc.gov/#/location-summary?version=dec2...

[2]: https://broadbandmap.fcc.gov/#/location-summary?version=dec2...

[SmirkingRevenge]

I just check my block, and it's claiming 5 providers. Realistically, we have Spectrum, and that's really it.

It even claims we have AT&T fiber! Nearby neighborhoods do have it (along with google fiber) but not this one.

[paulddraper]

Everything I quoted was 2+ providers. I never claimed 5+ was common.

"Two long-term competitors are not enough for meaningful competition."

That is a legitimate argument.