October 20, 2006  ·  Lessig

HOTI points to Scott Cleland of the respected “The Precursor Blog” who has posted a reply to my FT article. My “thin rule,” Cleland says, is just “thin gruel” (by which I take it he means he doesn’t like my rule; I personally find the best gruel to be thin gruel, and in fact, in the increasingly cold Berlin mornings, I think gruel is a very good start to the day. I wish people would stop picking on gruel, thin or fat.)

The thrust of Cleland’s one pager is that I’ve been “loose with the facts.” Let’s review the charges: (Cleland’s words are bolded below):

Loose Fact #1:

Lessig asserts: “In the U.S. at least, broadband competition is dying.”
Anybody who cares to check, will find that broadband prices are falling, broadband speeds are increasing, consumer choices are increasing steadily, broadband investment and deployment are strong, and innovation is vibrant.

This is Dick Cheney on the war. (Indeed, let’s give it a name: to Chenify): Put the prices issue aside till the next sentence: Are you kidding, Scott? Relative to every nation who should be considered to be competitive to us in this, we are worse off today than we were four years ago. When Bush said “10th is 10 spots too low” he was right (well, sort of. It’s actually 9 spots too low), and yet now the US is 16th in broadband deployment. The worse things go, the more a certain set simply denies reality.

Why can’t an anti-neutrality advocate begin with what everyone knows is true: US broadband sucks — it is too slow, it is too expensive, and it is too unavailable. The only question is what we are going to do about it.

For those who care to go more in depth on this subject, I have produced two useful one pagers to prove this point: “Debunking the Broadband Market Failure Myth” and “Debunking the Broadband Competition Can’t Work Myth.”

So here’s where I was worried. Though this is an issue I’ve been studying “in depth,” I hadn’t read Cleland’s “one-pagers” before. (I hate the word “debunking”. Sounds way too Marxist for my taste). So I took a deep breath and clicked on the one pagers, expecting to find a refutation of the data upon which I had based my understanding that in fact, prices have not fallen.

That data, again summarized well in Broadband Reality Check, suggests that first, cable prices have increased slightly, and second, while NOMINAL DSL prices have fallen, the $/MB has gone up, since the speed of the offering for the cheapy deals is significantly slower.

Here’s Cleland’s “debunking”:

Real DSL prices have fallen ~50% as speeds have roughly doubled over the last 2 years; introductory DSL prices have fallen ~70%in ~3 years; average monthly DSL prices fell ~15% from 2004-2005.

Every fact stated here could be consistent with the conclusion that the $/MB has gone up. (Well, almost: You’d have to be a bit charitable in interpreting “as speeds have roughly doubled over the last 2 years” — that’s plainly not true, as the “introductory” packages he points to offer less than 1 MB speeds). So where’s the beef, Scott? Where is the data to debunk “Broadband Reality Check”? I’d be happy (in the academic sense of that term) to be proven wrong about relying upon the data I relied upon. I’d be even really happy to learn that average $/MB prices for DSL have gone down. But notice the critical fact Cleland didn’t try to “debunk”: That prices in the US range from 6x (France) to 12x (Japan) the $/MB of the US. But don’t worry, every-thing’s great. The war’s great. Broadband in the US is great. The deficit is great. It’s just the best of all possible worlds…

For those who want to get the FCC’s analysis of broadband competition click here; or for the FCC’s analysis of Wireless competition click here.

I love the way anti-regulation types hate everything the government does, except data that supports their argument. Talk about “debunking”: The FCC’s analysis has been the subject of extensive criticism, including by the GAO. The problem is the FCC’s method for counting penetration with in a zip-code: They conclude that if 1 person within the zipcode has a broadband choice, the whole zip-code has broadband choice. As the GAO concluded about this obvious fudge: “the number of providers reported in the ZIP code overstates the level of competition to individual households.”

For those who don’t want to be bothered with facts and analysis, but just want anti-business assertions about what imminent peril our way of life faces from continued free and open competition on the Internet click here for SavetheInternet.com, of which Professor Lessig is a Charter Member.

This is the part of this debate that drives me nuts: As if this is a battle between “anti-business” sorts, and pro-business sorts. I understand how it’s easy to believe that if you spend your life thinking about other things, and spend 30 seconds thinking about this issue. But for anyone inside this debate, this claim is the most bogus sort of rhetoric there is.

This is not a pro vs. anti-business debate at all. The whole point of the Network Neutrality argument I’ve advanced (for almost 8 years now) is about what conditions produce the greatest growth in applications and content. The aim is to maximize wealth for the economy as a whole, and not just for the network owners. The whole argument is that a neutral network incentivizes more competition in applications and content than a network controlled by network owners. Think the cell phone network vs the Internet: This is not a battle between pro and anti-business sorts, it is a battle between cell-heads and net-heads.

There are those who continue to function normally in the world who believe, all facts to the contrary, that this is a debate about regulation vs. no regulation. I’ve spilled too many bits over that canard to believe wasting more space here makes any sense. But I wish those anti-regulation sorts would spend some of their effort getting the FCC out of the business of regulating (through property) spectrum.

Loose Fact #2

Professor Lessig asserts: “There are fewer competitors offering broadband connectivity today than there were just six years ago. The median consumer has a choice between just two broadband providers. Four companies account for a majority of all consumer broadband; 10 companies account for 83 percent of the market.”
What Professor Lessig fails to explain was that six years ago we basically had NO broadband competition, because we had a de facto monopoly for wholesale Internet access called dialup, which had lots of resellers of the underlying monopoly service, which Mr. Lessig likes to call competitors.

Yea, I’m old enough to remember those days. Many many businesses would try to get me to switch to their service by offering me lower prices and higher quality. I confess, I call that competition. But whatever you call it, we need more of it today.

Over the last six plus years, the free and open Internet that has been unfettered by regulation has created a steady increase in real inter-modal broadband competitors/choices for consumers.
What Mr. Lessig really laments is the decrease in the faux/artificial regulatory-favored Internet Access resellers that basically competed on brand; and the increase in REAL inter-modal competitors that can truly compete on price, speed, innovative features, and mobility among other differentiators that consumers value about competition.

What I “lament” is that the speed of broadband sucks in the US, and the prices are too high. Again, if the policies of the last 6 years had really produced the kind of prices and quality that other competitive nations around the world have, I’d be the first to admit I was wrong. But if you turn off the Cheney channel, and looks at the sorry (and increasingly sorry) state of broadband in the US, at some point someone has got to ask whether this policy is a mistake? Call me a cut-and-runner, Scott. Because I definitely want to cut-and-run from the FCC’s policy.

What Mr. Lessig conveniently omits from his assertion that “broadband competition is dying” is the pesky little truth that real broadband prices have fallen by over half over the last three years and that competitive supply is vibrantly increasing.

But then again, there’s the problem of those “pesky” data necessary to support the ultimate claim that needs to be made: That the DSL $/MB have fallen “by 50%.” Show me the data, Scott, not the made-for-TV-soundbites.

Maybe Professor Lessig should take some more classes in economics and antitrust to bone up on the fact that competitiveness of markets are truly measured by effective pricing, by the trend of competitive entry and by the amount of innovation. Only undergrad courses covering antitrust would consider it sufficient to count the number of competitors in a market and then declare a market not competitive. Responsible scholars of competition understand that the competitive facts can vary widely in various markets, and that the number of competitors alone is insufficient data to determine the competiveness of a market. I am sure there are any number of attorneys with “real world” experience in analyzing competition at the DOJ Antitrust Division or at the FTC who would be happy to give Professor Lessig a little tutorial on this before he opines on this topic again on the world stage.

This is no doubt true. A submission to the FCC or to a court about market power with the substance of an 800 word op-ed would be absurd. And indeed, to show fully whether competition is improving or getting worse, you would need to go much more “in depth” than even Cleland’s nippy one-pagers. But really, Scott. This is an op-ed. They don’t allow footnotes.

Loose Fact #3:

Lessig said: “Network owners now want to change this by charging companies different rates to get access to a ‘premium Internet.’” [bold added for emphasis]
This is the way the Internet has operated since it was commercialized in 1995. There have long been been three Internet backbone tiers of service. And companies have long paid for a “premium” Internet since they upgraded from dialup to broadband!

So at a debate with George Gilder, Peter Huber made this same move. Look, no one is arguing about the backbone. No one is arguing for regulation of the backbone. This is a debate about last mile broadband, and the effect certain business models for the last mile will have on competition.

What planet has Mr. Lessig been on that he didn’t notice that companies pay for a “premium” Internet every day? Has he ever heard of the Akamai “premium” service which has been used by most all the biggest online companies to get “premium” Internet service?

And of course this is exactly the criticism I was trying to preempt by my original post on this matter. Obviously, companies do whatever they can to make their content on the net run well as it can. Google must spend millions around the world on caching servers. Everyone spends what they can to get the fastest servers they can.

But again, this is exactly the sort of competition we should celebrate — businesses spending money to add real capacity and functionality to the network, by going to a (relatively) competitive market to add that capacity.

My complaint is not against that. My complaint is about (relatively) uncompetitive markets, and about the consequence of them exercising power over the next YouTubes of the world. No doubt, as they extract rents from these businesses, they make Wall Street happier about them. But as my focus is not the net wealth of a handful of companies, but instead, the wealth of the economy as a whole, what’s good for them is not the end of the matter.

Indeed, this is exactly why my position on Network Neutrality is not as extreme as some. As I testified, for example, I’m all for “consumer tiering” by network providers — where network providers offer higher quality to consumers for more money. That again is the sort of business model that creates an incentive to increase capacity.

“Access tiering” doesn’t. Or at least, I’m looking for the economic analysis to show it does. What I’ve seen so far is that in an relative uncompetitive market, “access tiering” creates an obvious (and perverse) incentive: with relatively limited competition, if you can charge a premium for a “fast internet,” you don’t have much incentive to make the rest of the Internet very fast at all.

  • three blind mice

    I wish people would stop picking on gruel, thin or fat.)

    well, it’s nice to see that we agree on at least one thing ;-)

    with all due respect, professor, you ARE being loose with the “facts” beginning with the factually incorrect, terribly misleading, “made-for-TV-soundbite” of “net neutrality.”

    the internet is NOT neutral and it was never designed to be so. the “net”, for example, is designed to handle TCP and UDP packets completely differently which has a profound impact on applications. indeed, because the net is not neutral, “cool” applications like VoIP and IPTV are possible.

    there is absolutely no evidence that a similar bias introduced into the last mile connection to “the internet” would not have a similar positive contribution on applications and content. moreover, there are very good reasons to believe that such a bias would make investment in last mile infrastructure more attractive.

    US broadband sucks, it is too slow, it is too expensive, and it is too unavailable.

    somewhat true: compared to the situation here in Sweden, for example, broadband in the US is too slow, it is too expensive and it is too unavailable.

    but also somewhat false: compared to broadband in the US five years ago, however, it is true that “broadband prices are falling [and] broadband speeds are increasing” and broadband service is available to more people.

    This is not a pro vs. anti-business debate at all. The whole point of the Network Neutrality argument I’ve advanced (for almost 8 years now) is about what conditions produce the greatest growth in applications and content. The aim is to maximize wealth for the economy as a whole, and not just for the network owners.

    *mice choke on our gruel*

    c’mon professor, “applications and content” have to get from point A to point B. it does not happen by magic, but due to massive investments in infrastructure, investments borne by network operators.

    What I “lament” is that the speed of broadband sucks in the US, and the prices are too high.

    …This is a debate about last mile broadband, and the effect certain business models for the last mile will have on competition.

    the idea that preserving the status quo (i.e., “staying the course”) is going to create more competition for the last mile connection is frankly bizarre.

    the problem herr professor, is that in the U.S. there is too little incentive to invest in the last mile connection. it is a problem almost unique to the U.S.

    it seems to be a multi-dimensional problem: partly caused by FCC regulation, partly caused by the mentality of existing network operators, partly caused by the ignorance of consumers, partly caused by the U.S. industry who make the equipment, and partly caused by activists such as yourself with a populist agenda.

    “access tiering” seems to be a reasonable solution to the problem because the “staying the course” certainly hasn’t worked.

    “I’m looking for the economic analysis to show it does.”

    and we’re looking for one that shows it doesn’t.

    here in Stockholm, home of Ericsson, we three blind mice have the choice of three different DSL providers, two different cable providers, and three different 3G providers… we’re connecting to your site over and honest to goodness 24 Mbps DSL connection.

    in such a situation, there is no need for a heavy-handed regulation such as “network neutrailty”.

  • http://www.redowl.ca Jon-o

    The comment about dialup made me almost laugh – ‘sure we had lots and lots of companies offering similar services and trying to attract customers by being better at what they do and by offering better prices, but it was basically a monopoly.’

    It seems there’s a real divide in thought between some that think ‘competition’ implies a completely different technology, network, and incompatible systems, and those that see that real competition happens when customers are completely free to move between different companies, without moving house, buying all sorts of associated products, or dramatically changing the way they use the product/service.

    Two completely different and incompatible products can’t really compete meaningfully with each other!

  • http://del.icio.us/knapjack Jack Johnson
    • Try porridge instead of gruel.
    • One aspect of this debate that I haven’t seen by either side is the barrier to entry in the U.S. Whether or not the variety or choice has improved, it seems broadband is still limited to families that have roughly $300/year in disposable income. To compare broadband competition to, say, cell phone competition or long distance service may seem unjust, but one thing both of those markets have done is address the consumers at the bottom end of the financial spectrum.
  • http://www.mediaaccess.org Harold Feld

    Scott, who is, after all, getting paid to front for the Bells, is no slouch when it comes to believing in competition. TRecently, he has taen me to task for my “strawman” of “perfect competition” as opposed to all the real competition he sees around him. Yes, and the Emperor is wearing the most splendid clothes imaginable, only a fool or an economist fails to see them.

    The old econ standard is that four equal sized firms is the bare minimum for a market to be “moderatley competitive.” That’s the benchmark employed by the Department of Justice Antitrust division and Federal Trade Commission, called the Herfindahl-Hirschman Index. http://www.usdoj.gov/atr/public/testimony/hhi.htm A market with fewer firms is “highly concentrated.”

    Highly concentrated markets behave in dfferent ways from moderately concentrated and genuinely competitive markets. Sweden (which one fellow keeps boasting about) enjoys a high level of competition because of government rules that require it. France and the NEatherlands, now that they have imposed similar open access rules, are experiencing a ddramatic increase in speed and a significant drop in cost as measured by bits/sec.

    A monopoly market is different from a duopoly, is different from a triopoly or oligarchy or highly concentrated market, is different from a moderately concentrated market, is different from a competitive market. Other factors, such as consumer lock-in via high switching costs or network effects, also play a role in whether a market behaves like a competitive market or not.

    Most people learn physics in high school with a simplified model that drills on the basic Newtonian specifics. Most high school physics problems begin “assume no friction.” That works fine to teach the basic principles. What happens if you try to build a rocket using what you remember from high school physics? Yes, it is much cheaper to build rockets if you assume no friction, but they will not get very far.

    Same here. Most folks vaguely remember some basic economics about competition where there is “monopoly” (bad), government ownership (bad), and competition (good). But reality is more complicated.

  • three blind mice

    But reality is more complicated.

    yes it is Harold Feld which is why simplistic solutions such as “network neutrality” are so misguided.

    Sweden (which one fellow keeps boasting about) enjoys a high level of competition because of government rules that require it.

    well, there are actually three of us ;-)

    your statement is partly true. government rules require Telia (the former state-owned telephone monopoly, now Telia/Sonera) to open its copper wires to competing DSL providers.

    but that’s not even half of the story.

    the cable operators (who do not have to provide open access to other ISPs) compete with the DSL providers.

    and the wireless operators (of which there are at least three licensed by the government) compete with both cable and copper for the same broadband “kronor.”

    it is a gross oversimplification to say that the competitive situation in sweden is a result of “government rules that require it.” it is a more a result of different technologies, and different industry segments all going after the same consumer.

    economists fret about things like friction, engineers apply lubrication to reduce it.

    when it comes to broadband, the deleterious effects of friction are greatly reduced by providing the consumer with the lubrication of choice.

  • ACS

    To all

    The perfect example of telecommunications without competition is Australia. There are only two real operators in the Australian broadband market, Optus and Telstra. Telstra owns about 85% of the market as a result of being the old nationalised telecom provider which has subsequently been privatised.

    Telstra owns the entire telecom “backbone” with Optus only engaging in “last mile” connections onto that backbone. The result is that Telstra can raise rents on Optus connections and has effectively priced broadband out of the average persons grasp (This was in part to allow telstra to continue to take advantage of its extensive dial up resources). Accordingly broadband consumption in Australia is very, very low.

    We suffer from what is essentially a (half government owned) monopoly on the majority of internet infrastructure and that is really starting to hurt small business operators and domestic internet traffic. It is worse in rural areas where there is close to zero penetration in the broadband market because of the high installation costs.

    Competition is definitely the way to go. Since Optus stepped up operations the prices of Telstra broadband and cable has come down a little compared to the $1500 ($US 1000) connection fees in 2002/2003. But, competition has to be supplemented with rational policies promising all carriers access to the telecommunications infrastructure. Network Neutrality in application to back bone infrastructure would be suitable and probably a welcome move in Australia. However, we are an extreme example and do not have to advantage of joint ownership of telecommunications ownership that is enjoyed in America.

    Of course, technology will continue to provide but telecommunications (with the thousands of miles of line installation and hundreds of thousands of hours of technical support) is a business that cannot be sustained with a “one size fits all approach”. The choice to offer a premium service should be engaged if only to improve high quality services. On the other hand there should be safety net so that disadvantaged persons (IE rural persons, small business and domestic connections) are not further left behind on the information superhighway.

    In conclusion, I think what we are looking at here is a new form of social welfare in relation to internet connection and it would be negligent to leave our moral compasses at home when considering these issue.

  • Link Hoewing

    Larry, I read your note about broadband competition. I think we need to focus on several questions that are at the core of this issue. Is the U. S. market performing well for the consumer and for the economy? Are the trends in terms of more speed, more capacity, more options going in the right direction? Are more Americans getting choices in the market place and are American companies investing in better technology and better networks?

    I think the answer to all of these questions is yes, although I would be the first to say we have some gaps in broadband deployment – especially in rural areas – and performance I would like to see improve.

    First, consider where we are today. You say we are “worse off today than we were four years ago” but consider what is happening. Four years ago (see http://www.pff.org/issues-pubs/other/GlobalPerspectivesontheDigitalEconomy.pdf – slide six in the presentation), American broadband providers were at a low point in their capital investment in networks. The Internet “bubble” had exploded and investment tanked.

    In 2003, the FCC (see http://hraunfoss.fcc.gov/edocs_public/attachmatch/FCC-03-36A1.pdf) adopted its broadband policy and exempted packet switching and broadband data networks from its “telephone” unbundling rules. Not long thereafter, Verizon (see http://eweb.verizon.com/npage/archive/stories_03jul/20031117.shtml) announced its plans to begin rolling out FTTP or fiber to the home. Other telcos began to build out fiber closer to neighborhoods and homes as well.

    Since that time, Verizon has been building out its fiber network in 18 states. We now offer speeds up to 50 megs (see http://gigaom.com/2006/07/18/50-megs-for-80-from-verizon-fios/) in some areas and have a standard 30 meg offering over our fiber network that is priced where DSL running at a meg or so was just a few short years ago.

    You point to price as a problem with broadband today in the U. S. Yet, the U. S. has experienced very fast uptake in broadband connections over the last couple of years and now has more homes connected to broadband than any other country (see http://www.oecd.org/document/9/0,2340,en_2649_34223_37529673_1_1_1_1,00.html). Our offerings must be affordable to have so many Americans connected to broadband. The OECD report is actually dated because more recent data shows that nearly 70 percent of Internet connected homes in the U. S. are now using broadband as opposed to dialup (see http://www.websiteoptimization.com/bw/0604/), which means that well over 40 million American homes are connected to broadband.

    Yes, many of those homes have lower speeds than are commonly found in places like Korea but as I noted, U. S. speeds are increasing quite rapidly due to new technologies like fiber and competition from cable and other providers (Cablevision has announced a 30 meg speed in response to Verizon’s offerings – see http://gigaom.com/2006/06/21/cablevision-broadband-now-faster-than-lie/). Just over a year ago (see http://www.ftthcouncil.org/documents/352861.ppt slide seven in the presentation), only 100,000 or so homes had fiber connections in the U. S. while today, over 1 million do, substantial growth.

    I also do not believe that the price comparisons I’ve seen are fair or accurate, and frankly I think what people are actually doing in the market – in other words whether they are moving to connect to broadband in large numbers – is a more important indicator. However, Verizon’s FiOS 15 meg offering is priced just about where entry level (one megabit per second or so) DSL was a couple of years ago ($45) and this puts our offering right in the middle of the pack (see http://www.itu.int/osg/spu/presentations/2006/shaw-emerging-policy-regulatory-challenges-ngns-6-july-2006.pdf slide 14 in this presentation) on a U. S. dollar per 100 kbit/s basis according to the ITU data – above flagship offerings in Canada, Finland and even much touted France. But again to me the key metric is affordability and based on the pace of broadband uptake in the U. S., I can’t see how that is an issue.

    So, while things are not perfect with the U.S. broadband deployment, the trends are going in the right direction when it comes to speed, availability, competition and affordability. And if we can keep these trends going, I believe we will continue to advance and outpace most other countries when it comes to landline broadband competition and capabilities. In Europe (see http://lw.pennnet.com/articles/article_display.cfm?article_id=271114), they are not seeing the widespread landline broadband competition we have and this is a major reason why major incumbent providers are not investing in fiber technology. The trends I see are the kind the FCC was trying to stimulate with its policy changes and the policies are working. The more high capacity fiber that gets deployed – and the more responses there are in the market from cable and other telcos to upgrade their speeds – the more we will move ahead as a nation. The beneficiaries are the U. S. consumer and our economy.

  • Link Hoewing

    I forgot to mention in my post that if anyone wants to visit our new policy blog at Verizon to respond to my comments, you can do so on our blog here http://poliblog.verizon.com/PoliBlog/Blogs/poliblog.aspx. Thanks.

  • Link Hoewing

    I note that I made a mistake in my post. I said the following at one point: “We now offer speeds up to 50 megs in some areas and have a standard 30 meg offering over our fiber network that is priced where DSL running at a meg or so was just a few short years ago.” I meant to say “15 meg offering over our fiber” not “30 meg”. Sorry about that.

  • http://www.jupiterresearch.com Ian Fogg

    This piece is also relevant to the debate of how the US compares with Europe:
    (free registration required)

  • Ray Molawole