« Assorted links | Main | How much better is local food for the environment? »
How far behind is the U.S. in terms of broadband?
In response to Paul Krugman and others, FCC commissioners Robert McDowell writes:
The OECD says the U.S. has dropped from 12th in the world in broadband subscribers per 100 residents to 15th. The OECD's methodology is seriously flawed, however. According to an analysis by the Phoenix Center, if all OECD countries including the U.S. enjoyed 100% broadband penetration -- with all homes and businesses being connected -- our rank would fall to 20th. The U.S. would be deemed a relative failure because the OECD methodology measures broadband connections per capita, putting countries with larger household sizes at a statistical disadvantage.
Here are statistics on household size; I am suspicious that McDowell cites only a polar point (which in essence is ranking *only* household size, and not how much household size contributes to the current rank order) in support of his case. Not every argument in his rebuttal succeeds.
More fruitfully, should we should compare Europe to the whole U.S. or to individual states?:
...if we compare many of our states individually with some countries that are allegedly beating us in the broadband race, we are actually winning. Forty-three American states have a higher household broadband adoption rate than all but five EU countries. Even large rural western states such as Montana, Wyoming, Colorado and both Dakotas exhibit much stronger household broadband adoption rates than France or Britain. Even if we use the OECD's flawed methodology, New Jersey has a higher penetration rate than fourth-ranked Korea. Alaska is more broadband-saturated than France.
Maybe federal policy is not mainly at fault, though more contestability is still a good idea. By the way, here are some alternative broadband rankings, the U.S. comes in twelfth.
The pointer is thanks to Ben Davis.
Addendum: Over at Mark Thoma's, Peter Schaeffer has a good comment at July 23, 2007 at 11:05.
Posted by Tyler Cowen on July 24, 2007 at 09:57 PM in Data Source | Permalink
Comments
You've missed the main point of McDowell's article, which is that delivery
platform innovation drives broadband buildout. That's why Krugman's New York Times
op-ed was not just wrongheaded but irrelevant.
As McDowell says:
"Consumers don't buy fat pipes for their own sake; they buy applications
and content that require fat pipes. As consumer demand for more bandwidth-
intensive applications and content increases, so does the incentive for network owners to provide more bandwidth.... We are creating more competition through the construction
of new delivery platforms."
Consumer demand leads to deeper and wider investment to meet that demand,
as per Menger.
Krugman also overlooked the state franchise laws that make it costlier for
broadband firms to offer apps like television; these don't exist in France and
Japan. Neither McDowell nor Krugman mention the baneful effects of intellectual monopolies
(like the one that derailed Napster) on the development of delivery platforms.
Abolish these regulations and restrictions, and the monopoly formerly known
as intellectual property, and watch the market soar--for both platforms
and bandwidth.
Posted by: Bill Stepp at Jul 24, 2007 10:23:04 PM
Er, I was going to say "population density" and then decided to read Schaeffer's comment.
When I was in Paris not too long ago, my host had [cough] dial up (28.8 or worse) and I could only find one internet cafe there (vs. many in London and Madrid). Paris was a digital wasteland.
I'm not sure I understand the point of Krugman's later correspondence with Thoma. He says that except for satellite, cellular, and cable service he had no broadband. Then he and several commenters proceed to indict free markets based on this failure of regulated utilities.
Posted by: Eric H at Jul 24, 2007 10:36:11 PM
You're more of a man than I am for finding Schaeffer's post. I don't even dare wade into the comments section on Thoma's blog. (Seriously how has "Cut'Anne'Paste" not been ip banned from the blogosphere? All she does is spam news links into the comments section of numerous blogs. She has to be a NYTimes bot.)
Posted by: Dan in Euroland at Jul 24, 2007 11:19:59 PM
Wow, McDowell is a real idiot. This gives new meaning to "special pleading." Yea, and you know what? Minnesota is almost as pacific as Norway. You see, if we just x out this Southern part of the US and don't count the black folk and the immigrants...
Posted by: david at Jul 25, 2007 7:09:08 AM
Bill Stepp: Krugman also overlooked the state franchise laws that make it costlier for broadband firms to offer apps like television; these don't exist in France and Japan.
Quite wrong, BS.
Why don't you ask someone in France before writing such nonsense? France offers "triple play" ISP services over DSL: Internet, Telephony and Television.
Posted by: Lafayette at Jul 25, 2007 7:14:58 AM
Clearly in the U.S. we need federal subsidies to provide broadband service low-density rural areas -- otherwise, how else am I going to get cheap broadband service up at the cottage? ;)
Posted by: Slocum at Jul 25, 2007 9:03:58 AM
You are talking about counting methodology and I want you to talk about
economics.
When I read Matthew Ygelias's piece and to a lesser extent Krugman's my first thought
was wait.......
I can see how competition is good for prices. I love it. Allowing Verizon to compete with Comcast is saving me 50 or 60 dollars a month via the new triple play packages. Yea!
But how does competition result in more infrastructure investment and/or innovation. I thought monopolies might have an incentive to over invest in R & D.
I just don't see how more competition necessarily means better pipes. And Ygelias and to a lesser extent Krugman seem to imply this.
Maybe public investment and open access is the best way to go. But, I don't think laws that help promote competition, will necessarily result in more investment/better pipes.
Posted by: SomeGuy at Jul 25, 2007 9:35:33 AM
Eric H.
Where the heck were you in Paris? I was just there, and on certain streets downtown there were
scads of internet cafes.
It is ridiculous that after all the huffing and puffing, if there were a terrorist attack on the
New York subway system, people would not be able to phone out. They have been able to do so for
quite some time from the much bigger Tokyo metro system. How much are we spending on building up
al Qaeda in Iraq?
Posted by: Barkley Rosser at Jul 25, 2007 10:17:24 AM
Bill Stepp: "Krugman also overlooked the state franchise laws that make it costlier for broadband firms to offer apps like television; these don't exist in France and Japan."
Quite wrong, BS.
Why don't you ask someone in France before writing such nonsense? France offers "triple play" ISP services over DSL: Internet, Telephony and Television.
I think that was Stepp's point. State franchise laws make these packages costlier to offer in the U.S. than they are in France.
Posted by: Steve Miller at Jul 25, 2007 10:41:52 AM
"Consumers don't buy fat pipes for their own sake; they buy applications
and content that require fat pipes.
Of which there are none, because application and/or content that require fat pipes don't exist if there are no fat pipes.
Posted by: Back to the real world at Jul 25, 2007 10:43:14 AM
So which is it - do you want more competition or not?
The current infrastructure owners (the phone and cable companies) want to protect their oligopoly. In addition they are proposing to add new fees to boost profits using the argument that they need the revenue to pay for "innovation".
However free market theory would seem to argue that the best way to get innovation is to increase competition. Well to increase competition in this market means that the market power of the existing players will have to be limited in some fashion. This can be via legislation or government funding of new technology or allowing new players into the field as is being attempted by several municipalities and other local entities.
To make the issues more stark there is the comparison with the services being offered in other countries. This shows that many services are not available because of market considerations, not technological ones.
It does not make for a strong argument when the comments center on ad hominem attacks on one's opponents or calling into question some details of the comparison statistics. Anyone who favors the status quo or the arguments of the existing major players needs to explain how this is a "good thing" in its own right. Belittling others isn't adequate.
Posted by: robertdfeinman at Jul 25, 2007 11:03:38 AM
"To make the issues more stark there is the comparison with the services being offered in other countries. This shows that many services are not available because of market considerations, not technological ones."
This is where the argument is stalled, for many of us.
Comparisons with other countries are rarely as legitimate as Dr. Krugman wishes them to be. In this case, several reasons have been offered by Mr. McDowell and other commenters as to why the study (and Dr. Krugman's personal anecdotes) may not be an accurate portrayal of reality.
Posted by: Captian Hook at Jul 25, 2007 1:38:05 PM
Discussions of telcoms seem as good a place as any to link to The Myth of Natural Monopoly for some history of the industry.
Posted by: TGGP at Jul 25, 2007 1:57:12 PM
Wouldn't it be more interesting to rank countries on high-speed Internet affordability, rather than penetration? Why penalize countries whose citizens have better things to spend their money on then Internet access?
-Kevin
Posted by: Kevin Postlewaite at Jul 25, 2007 1:59:26 PM
Mr. Cowen,
Thank you for posting the reference to my post over at Economist's View.
In my opinion, Krugman has succumbed to the temptation of very smart people to think they can knowledgably comment on subjects (frequently technical) that they don’t know much about.
Posted by: Peter Schaeffer at Jul 25, 2007 5:35:50 PM
I know this is a website, and probably 90%< of us have some type of broadband. But what's so great about broadband vs, um, narrowband? I've had dial-up, DSL, 3 cable carriers and ethernet. None of these seemed to make as much difference as which computer I was using. Besides, if I spend 5 minutes reading this page, does it make much of a difference if it loaded in 2 seconds or 20?
Posted by: kingwood kid at Jul 25, 2007 5:52:59 PM
Back to the real world writes (quoting my quotation from the article):
"Consumers don't buy fat pipes for their own sake; they buy applications
and content that require fat pipes.
Of which there are none, because application and/or content that require fat pipes don't exist if there are no fat pipes.
Real world, roads don't get build until there are vehicles existing to travel on them. The car (and truck) preceded the highway. Passenger excursions and goods to transport by horse and buggy preceded cars and trucks. The interstate highway system would be the equivalent of broadband; the local two-lane system would be like telecom's narrowband.
More generally, consumer demand drives investment, not vice versa, as Carl Menger elaborated in his Principles. So competition at the platform delivery level logically precedes and drives broadband buildout. If there were no Napster, YouTube, etc., there'd be little demand for broadband, so it wouldn't exist, just as if there were no cars and trucks (and no demand for their services), there'd be no highway system.
The mistake is understandable, given how horrible the teaching of basic economics is, unless you're schooled by an Austrian or fellow traveler.
Posted by: Bill Stepp at Jul 25, 2007 6:22:51 PM
Please check out this report:
http://www.freepress.net/docs/shooting_the_messenger.pdf
"The major critiques leveled at the OECD data simply fall apart upon closer examination. The coordinated attempt to “shoot the messenger” cannot hide critical failures in the U.S. broadband market. These failures are chiefly due to poor policy decisions that have fostered an anti-competitive marketplace. Our European and Asian counterparts are outperforming us because they have policies that foster vigorous competition in the broadband marketplace, offering consumers more
choice, faster speeds and lower prices."
Posted by: Frank at Jul 25, 2007 7:26:13 PM
Frank,
I looked at the freepress article. Once again, it didn't really address the issue of density. All urban areas are not alike. The urban areas of the US are far less dense than those in Japan, Korea, or Europe.
Canada and Australia are reasonable matches for the US. Canada is modestly ahead, Australia is trivially behind.
Note that according to the ITIF rankings (http://www.itif.org/files/BroadbandRankings.pdf) France has a lower household broadband penetration rate than the US. Average speed in France is considerably higher than the US. This fits with the higher density of French urban life, but somewhat lower per-capita GDP.
Posted by: Peter Schaeffer at Jul 25, 2007 8:31:43 PM
Peter,
So why don't I get the good stuff when I moved from the suburbs of Allentown to downtown DC. It is the same crappy cable broadband at the same price.
Posted by: RobbL at Jul 25, 2007 9:14:29 PM
Robbl,
At last a valid question. I think the answer amounts to standardization. In other words, the Telcos pick technology that matches most of their network. Even if they could offer VHDSL in some areas, they adopt an infrastructure that can be widely deployed. In Korea, VHDSL comes close to being lowest common denominator. Not in the US.
Of course, even American cities aren’t that dense. I worked on connecting a non-for-profit to the web in Houston, TX. Southwestern Bell refused to provide service at all, because we were more than 17,000 feet from the central office. That was at the corner of Kirby and 59, inside the 610 loop.
Fortunately, Covad was willing to provide service (at double the Bell price). As you can guess, I do favor broadband competition.
However, you aren't using DSL. Cable modem technology is (apparently) not nearly as distance/density sensitive as DSL. I get around 6 mbps where I live. The ITIF study suggested somewhat higher cable data speeds in Canada (7.6 mbps).
Why is downtown DC so bad? I don’t have any idea. I worked on DSL in the Capitol Hill area and was appalled by the PPOE setup Verizon uses. Where I live DSL is available without PPOE. I tell everyone to dump PPOE based DSL and switch to cable if it is available. I switched to cable, not because of PPOE, but on reliability grounds.
Posted by: Peter Schaeffer at Jul 25, 2007 10:20:51 PM
Bill Stepp:
Delivery platform innovation drives broadband buildout....As McDowell says: "Consumers don't buy fat pipes for their own sake; they buy applications and content that require fat pipes.
So, like tens of millions of other people, I'm sick of going to the video store and I'm ready to start downloading movies over the Internet.
That's my obvious application; Apple and Netflix have the delivery platforms and IP agreements; now where's my fat pipe?
Oh, that's right, it doesn't exist, because the Telco has no interest in providing me with said pipe until they can require me to sign up to their "exclusive," inevitably monopolistically designed, movie downloading service.
And the Cableco has no interest, because they'd be competing with their own video-on-demand services.
I wonder what Menger would say about that. Adam Smith would call it "rent."
Posted by: theo at Jul 25, 2007 10:49:28 PM
Peter Schaeffer,
Your density of urban areas argument is cleverer than anything the FCC hack came up with, but it still doesn't explain why the Telcos can't seem to apply different technologies to areas of different density.
Verizon and AT&T both serve areas substantially larger than any European country. In fact, if you just considered the urban areas within Verizon's service area, you'd have a country the size of Germany -- of equal urban density within select areas, and much denser overall. So why can't they offer decent broadband to that market? It's baffling.
I wonder if it has anything to do with the fact that, for the last ten years, the telco leadership have been entirely obsessed with (1) reconstituting Ma Bell, and (2) anticompetitively buying up the wireless industry. (Not to mention (3), illegal surveillance of their customers' phone and internet traffic.) Perhaps a reasonable regulatory regime at the FCC could have forced the Telcos to pursue market growth rather than oligopoly.
Posted by: theo at Jul 25, 2007 11:13:11 PM
Bill:
> If there were no Napster, YouTube, etc., there'd be little demand for broadband, so it wouldn't exist,
Careful - you're oversimplifying to the point of being wrong. Broadband was on the market well before either existed Same dealie with the roads - cars/trucks came before highways, but AFTER roads.
As you know, the reality is that new products follow supply/demand spirals. Mostly - see the rest of the comment.
McDowell also misses an important fact:
> As consumer demand ... so does the incentive for network owners to provide more bandwidth.
Yes, but, big as it may be, we often don't see that demand satisfied in monopolies. DEMAND for bandwidth greater than sa single modem over a single phone line did, in fact, precede the breakup of AT&T and MCI's historic agreement to provide at reasonable terms at long last.
Similarly, we saw telco-side broadband provision delayed for decades after its planning and creation, but it didn't really show up until AFTER the cable companies started providing the service. You gonna tell me the telcos happened to wait for exactly the right demand moment? Or maybe Time-Warner out-innovated the telcos (no, it's NOT just a question of demand curves, because early promised broadband versions had correspondingly less promised bandwidth).
Posted by: Jon Kay at Jul 26, 2007 1:59:01 AM
theo,
I really can't answer your question. However, let me ask one. Have you ever worked for a large corporation?
Doesn't picking one DSL technology platform and standardizing on it sound familiar?
The other point is that telcos frequently offer tiered DSL pricing these days. More for higher speeds, less for lower speeds.
If delivered speed was tied to physical capacity, how would this work? People living within 2000 feet of a CO (Central Office) would get 40 mbps at $100 month? People living within 5000 feet would get 20 mbps at $70 a month? People living within 10,000 feet would get 5 mbps at $50 a month? People living within 15,000 feet would get 3 mbps at $40 a month?
Note that the actual cost of service (to the telco) would be almost identical in each case. Would consumers understand such pricing? Would they accept it? Should they?
Alternatively, telcos could simply offer the maximum speed workable for each residence at a fixed price. If you lived closer to the CO you would get more bandwidth. If you lived further away you would get less. Would consumer accept this approach?
This would eliminate the tiered pricing structures currently offered. I don’t see a gain here. Tiered pricing enables lower income consumers to get some broadband at a more affordable price. Somewhat like airlines charging business travelers more.
My guess is that telco broadband capacity is driven by both the need for technology standardization (what speed will work on 85% of our network…) and marketing standardization.
By the way, as I have stated several times, I favor competition and have personally benefited from it. However, America will never be Japan or Korea when it comes to broadband delivered capacity and pricing. Something like FIOS might give Americans Asian speeds, but at a considerably higher price.
Note that all of my numbers above are by way of example.
Final note. The right comparison (in my opinion) is the de facto density of Verizon’s territory versus the UK, France, or Germany. For example, the New York SMSA versus London or Paris. The New York SMSA is downright empty compared to Europe. The Boston SMSA is emptier still (I would have never guessed this, but it is quite true).
Posted by: Peter Schaeffer at Jul 26, 2007 2:21:07 AM
