George Ou & Bret Swanson on Berkman Broadband Report

by on October 21, 2009 · 15 comments

Last night here on the TLF, Bret Swanson raised a number of objections with this FCC-commissioned report about international broadband comparisons, which was conducted by some folks at Harvard University’s Berkman Center. Meanwhile, over at the Digital Society blog, George Ou also offers a hard-nosed look at the Berkman broadband report and concludes “The underlying data cited by Berkman study is simply too flawed to be of any use.”  I recommend everyone check out both essays.  It will be interesting to hear how the Berkman folks respond.  Some of these international broadband comparisons are really fishy.  [Here's a podcast we did on that issue two years ago.]

One quick point… Like Bret, I also found it shocking that–even though the report reads like an ode to forced access regulation–the Berkman folks didn’t spend much time discussing the result of America’s previous open-access regime. “The gaping, jaw-dropping irony of the report,” Bret argues, “was its failure even to mention the chief outcome of America’s previous open-access regime: the telecom/tech crash of 2000-02. We tried this before. And it didn’t work!”  Indeed, America’s regulatory experiment with forced access regulation involved a lot of well intentioned laws and regulation, and too many acronyms to count–CLECs, TELRIC, UNE-P, etc– but it did not result in serious, facilities-based competition.  Instead it offered us the fiction of competition through network-sharing, or what Peter Huber once referred to as building “networks out of paper.” The results were disastrous for investment during that period since regulatory uncertainly led to a lot of stunted innovation.

In sum, sharing is not competing.  You can socialize and commoditize old pipes for awhile and get decent results in the short-term, but you’ll sacrifice long-run investment and innovation if you do.  [For more background, see my recent essay on "The Fiction of Forced Access 'Competition' Revisited" and this old Cato piece on "UNE-P and the Future of Telecom "Competition" as well as Jeff Eisenach's PFF white paper, "Broadband Policy: Does the U.S. Have It Right After All?"]

  • sleeplesssueinstlouis

    Why is Oboma so interested in Net Neutrality? Isn't there an important vote this week? Why has he set aside $76+ million for “net neutrality”? (Wasn't that money supposed to go to creating jobs!) Is net neutrality going to affect my emailing you in the middle of the night? I am suspect of anything the eighteen wheeler rushing down the hill with Obama's picture on it wants.

  • mwendy

    The Net Neuts are asking for Title 2-like subsidies for their content dissemination. They want the network effect without taking the risk to build it themselves. Is there an entitlement to a privately-created network effect, especially when no demonstrable harm has yet resulted from those who own that network effect? No.

  • http://www.wetmachine.com/totsf Harold Feld

    “Socilaize.” Oh please! Is this really the new buzzword.

    The Berkman folks did not cover the U.S. in the 1990s because that wasn't their mandate. They were asked by the FCC to do a specific thing: look at what other countries did and see how that worked out for them. This the Berkman folks did, extensively.

    You are certainly free to argue that it isn't a complete picture because of the sustainability issues — although Japan and S. Korea have been at this long enough that we would have expected the sustainability issues to become more urgent. You can also make the case that unbundling did not work here, although I think the data for that points the other way. But it seems foolish to me to criticize Berkman for doing the job they were asked as if the failure to go and do an entirely separate report to support your objections was a failure of their initiative.

  • mwendy

    I have a bigger question. If the FCC used one of the RBOCs consultants to perform the agency's analysis function, wouldn't that raise some red flags. Let's say, NERA writes this report instead – the choice of the author has likely predetermined the answer. And, I think in this instance, we see the same.

    So, isn't this a waste of tax dollars – we knew when Berkman was chosen by the agency, that it would get this answer. They could have $500-or-so Large as a result, and then just have Susan Crawford write the durned thing.

  • billgoat

    Harold (or should I say Che)… I love the shirts actually. Very funny, made me laugh out loud and am thinking about ordering one for friends — although I'd be supporting PK financially, and that gives me pause.

    Berkman is clearly advocating for structural separation and for them to hide behind a supposedly objective report is disingenuous. It was David Weinberger (also from Berkman) who in an op-ed a month ago said “We could require access providers to open up their lines to other companies at profitable wholesale rates. Users would benefit from a newly competitive market. Or, we could legally separate those who provide access from those who provide content and services.”
    http://www.npr.org/templates/story/story.php?st

    This op-ed was written and posted before Yochai's study was done.

    So, a reasonable skeptic could easily conclude that Yochai, Weinberger and the Berkman Utopians put together a “study” that supported their structural separation worldview and then loaded it up with data to support a conclusion they'd obviously already reached.

    So yes… I think “socialize” is the appropriate phraseology here. Instead of playing semantic games, have the courage of your convictions and own it.

    Now, having said all of that… I thoroughly enjoy reading your stuff and find you very intelligent and quite witty (although I obviously strenuously disagree with just about everything you write)…. How's that for the velvet glove treatment?

  • HIREDGUN

    The reality is that if you take out fibre deployments that haven't been made by (a) municipalities (Sweden, Holland), (b) consumer-owned energy trusts (Denmark), or made with (c) significant helping hands on both the supply and demand side (Korea), there's really not much commercial FTTH deployment outside the USA. Yet non-commercial investments are not impacted by access pricing and access regulation like commercial investments are, so their presence in my view has zero relevance to assessing the efficacy of access pricing and access regulation policies.

    It seems very clear that some European countries have done better on penetration than the US, although here too it's not really clear that prices are always lower there. But in terms of commercial fibre deployment, the US is far ahead of WesternEurope. Indeed, in terms of all fibre deployment, the US is still far ahead of Western Europe. Even if we took the sum of FTTH plus FTTN plus DOCSIS 3.0 subscribers, the US would be ahead of Europe although data on some of this is missing. These differences in penetration versus advanced services deployment are unlikely to be a function of just the regulatory variables, but they are consistent with what one would expect.

    I don't see how you can get around the fact that US telcos have made massive investments. Particularly Verizon.

    I found it troubling that Berkman don't even discuss pricing issues, because the moment you say “open access”, the next question is “at what prices?”. Factors like option values and asymmetric risks that conventional TSLRIC pricing didn't handle well have been addressed in European regulatory settings, but I am not aware of any great solutions that have been achieved. It seems like a really futile exercise in providing spurious precision. Simply translating regulation from copper to fibre seems pretty weird, given the amount of attention these issues have received in Europe.

    Re: data, they say that looking at household penetration rates are inappropriate because one needs to account for business lines. Well, actually counting business lines is one big issue. My firm (in the UK) benefits from a shared 100 Mbps Ethernet service to our building; but we probably just get counted as one connection, because we are one customer for our vendor. But prior to the recession, there were 110 of us, 100% of whom had our own Internet-enabled computers.

    Spain has almost as many enterprise DSL connections as the US according to one source that I've seen. With 1/7th the population. Shock, horror? I think not. The root cause of this is that enterprises in the US get their Internet access much more often via special access/dedicated access/T-1/DS-3. Corporate data revenue per head of population in the US is much higher than in Spain.

    The relevant metrics should be: proportion of households who have broadband access, proportion of employees that have broadband access at work, proportion of businesses that have broadband access, proportion of adults who have a 3G mobile connection (European mobile penetration numbers are massively inflated by counting inactive users and multiple SIM users). In any case, the evidence that the US is “failing” on these counts is very thin.

    Until the 1980s, the US was the clear leader in telecoms. However if you look at data such as fibre route deployment, ISDN deployment, and digitalisation of exchanges, the US was behind several OECD nations in the 1990s. Other countries opened up their markets, and smartened up their act. They also began producing more IP– e.g., innovation starting originating elsewhere. In Finland, in Japan etc. The world became much more level, because information technology enabled greater information flows. This is the big reason that the US is not an absolute leader in telecoms, although its outcomes are actually quite good, especially in wireless.

  • HIREDGUN

    The reality is that if you take out fibre deployments that haven't been made by (a) municipalities (Sweden, Holland), (b) consumer-owned energy trusts (Denmark), or made with (c) significant helping hands on both the supply and demand side (Korea), there's really not much commercial FTTH deployment outside the USA. Yet non-commercial investments are not impacted by access pricing and access regulation like commercial investments are, so their presence in my view has zero relevance to assessing the efficacy of access pricing and access regulation policies.

    It seems very clear that some European countries have done better on penetration than the US, although here too it's not really clear that prices are always lower there. But in terms of commercial fibre deployment, the US is far ahead of WesternEurope. Indeed, in terms of all fibre deployment, the US is still far ahead of Western Europe. Even if we took the sum of FTTH plus FTTN plus DOCSIS 3.0 subscribers, the US would be ahead of Europe although data on some of this is missing. These differences in penetration versus advanced services deployment are unlikely to be a function of just the regulatory variables, but they are consistent with what one would expect.

    I don't see how you can get around the fact that US telcos have made massive investments. Particularly Verizon.

    I found it troubling that Berkman don't even discuss pricing issues, because the moment you say “open access”, the next question is “at what prices?”. Factors like option values and asymmetric risks that conventional TSLRIC pricing didn't handle well have been addressed in European regulatory settings, but I am not aware of any great solutions that have been achieved. It seems like a really futile exercise in providing spurious precision. Simply translating regulation from copper to fibre seems pretty weird, given the amount of attention these issues have received in Europe.

    Re: data, they say that looking at household penetration rates are inappropriate because one needs to account for business lines. Well, actually counting business lines is one big issue. My firm (in the UK) benefits from a shared 100 Mbps Ethernet service to our building; but we probably just get counted as one connection, because we are one customer for our vendor. But prior to the recession, there were 110 of us, 100% of whom had our own Internet-enabled computers.

    Spain has almost as many enterprise DSL connections as the US according to one source that I've seen. With 1/7th the population. Shock, horror? I think not. The root cause of this is that enterprises in the US get their Internet access much more often via special access/dedicated access/T-1/DS-3. Corporate data revenue per head of population in the US is much higher than in Spain.

    The relevant metrics should be: proportion of households who have broadband access, proportion of employees that have broadband access at work, proportion of businesses that have broadband access, proportion of adults who have a 3G mobile connection (European mobile penetration numbers are massively inflated by counting inactive users and multiple SIM users). In any case, the evidence that the US is “failing” on these counts is very thin.

    Until the 1980s, the US was the clear leader in telecoms. However if you look at data such as fibre route deployment, ISDN deployment, and digitalisation of exchanges, the US was behind several OECD nations in the 1990s. Other countries opened up their markets, and smartened up their act. They also began producing more IP– e.g., innovation starting originating elsewhere. In Finland, in Japan etc. The world became much more level, because information technology enabled greater information flows. This is the big reason that the US is not an absolute leader in telecoms, although its outcomes are actually quite good, especially in wireless.

  • Pingback: man cave

  • Pingback: DDOS Protected Servers

  • Pingback: meubelen

  • Pingback: Cigarette electronique

  • Pingback: premier league fixtures

  • Pingback: Dentist Camberley

  • Pingback: prix de l'immobilier

  • Pingback: свещ

Previous post:

Next post: