Mistakes, Mistruths and Outright Lies in the Assessment of Broadband Competition

           Readers of the May 13, 2015 edition of the Wall Street Journal got a triple dose of snark and questionable journalism.  On back to back pages, this major publication informed us that U.S. “broadband is a competitive market and becoming more so as fixed and wireless converge.”  Holman W. Jenkins, Jr. suggests that we ignore any rebuttal or contarary reports from the “know-nothings in Washington.”  See http://www.wsj.com/articles/why-aol-matters-again-1431471920.

            On the next page, the editorial writers of the Journal contradict Mr. Jenkins on Verizon’s motivation for wanting to acquire AOL.  Instead of pursuing new profit centers through vertical integration, such as advertising platforms, Verizon has to acquire AOL due to competitive necessity: “A company with a stranglehold on the connection to the customer wouldn’t need to buy AOL.”  See http://www.wsj.com/articles/the-aol-telltale-1431472741.  The editors see broadband competition and the Internet as “hypercompetitive.”

            The Wall Street Journal has led a campaign to convince legislators, judges, consumers and others that the broadband Internet access marketplace operates with such robust and  sustainable competition that any government oversight is inappropriate, if not illegal. 

            This conclusion is simply not true unless one intentionally ignores basic economics, antitrust law and common sense.  To conclude that the U.S. broadband marketplace is competitive one must include any source of access to the Internet, regardless of bandwidth, transmission speed (bitrate) and cost.  Instead of many instances where consumers, such as myself, count one and only one available broadband supplier, broadband competition true believers see seven or more suppliers.

            True believers see what economists term cross-elasticity where it surely does not exist.  They treat as “options” Internet access technologies that consumers do not consider equivalents. So for true believers, it is reasonable to include Digital Subscriber Line and perhaps even conventional dial up access, even though 1.5 Megabits per second service and surely 5.6 kilobit per second service does not cut it for the kinds of services broadband subscribers expect to access via their links.  DSL might barely provide a single, tolerable link to Netflix, but not if two members of a single household seek access at the same time.

            True believers in broadband competition readily add four or more terrestrial wireless carriers and at least one satellite option to the invemtory.  Yes 4G wireless can provide broadband access at sufficient high speeds, but a competitive analysis requires consideration of cost.  Many 4G subscribers gladly pay for wireless data plans, but the willingness to pay ends when a free or lower cost option is available. With data plans limiting subscribers to a miserly 1 or 2 Gigabytes per month, subscribers understandably migrate to their wired broadband service accessible with a wireless Wi-Fi router. Wireline broadband offers a monthly data allowance of 250 Gigabytes or more.  

            A back of the envelop calculation shows a wireless broadband rate of approximately $20 a Gigabyte and even more for a satellite option, factoring in equipment costs.  Wireline access costing as low as 12 cents a Gigabyte, based on monthly consumption of the full allotment.  The statistical compilation gets tricky here based on one’s agenda.  Sponsored researchers can show that Americans have the lowest cellphone rates in the world as least for voice and texting, by using 1000s of minutes and 1000s of texts per month.  So in fairness the 12 cent rate for wired broadband access could rise to about a $1 per Gigabyte if a broadband subscriber used far less than the total amount available.

            Wall Street Journal editorial writers and columnists ignore the reality of what consumers consider truly competitive broadband options.  Few consumers think “Two Buck Chuck” wine from Trader Joe’s competes with one hundred dollar Grand Cru even though both are wine products. Some might even consume both on different occasions, but doing so does not make the two product competitive alternatives.

            So at the end of reading the two pages, this loyal subscriber to the Journal wonders did they make a simple mistake or two or three, offer a little misinformation to make a bigger point, or lie through their teeth?

Lies, Damn Lies and Broadband Statistics

The FCC and NTIA have launched a broadband map that purports to give quite granular and current data about broadband accessibility.  Don't count on realistic statistics.  See Broadband Map

The casual reader won't catch the use of advertised, maximum speeds.  When, if ever, will the FCC and NTIA start to plug in real, measured speeds? 

The casual reader also may not quibble about the reported, advertised speeds.  When the site reports 50-100 megabits per second, as it does for my location, would not a reader infer a speed somewhere probably midway between the two poles?  Comcast offers a $99.99 plus, plus Extreme 50 Plan for downlink speeds "up to" 50 megabits per second.  So why not bump that platinum plan up to the NEXT rate band?  And let's forget about how many people actually subscribe to this level of service, if really available.

At first glance, the FCC and NTIA, are overstating reality.  This reminds me that there are lies, damn lies and broadband statistics.

Broadband Penetration in the U.S.: Saturated, Recession-Affected, or Pricing Out Many?

In addition to providing a better sense of what specific broadband service options consumers have in more narrowly drawn geographic areas, the FCC’s most recent statistics on broadband show a significant decline in new subscriptions. See INTERNET ACCESS SERVICES: STATUS AS OF JUNE 30, 2009 (September 2010); available at: June 2009 FCC BB Stats.

The Commission also reports that as of June 2009 there were 61 reportable residential fixed-location connections per 100 households, with 56 connections per 100 households operating at advertised whose speeds in excess of 768 kbps downstream and only 27 connections per 100 households operating at advertised speeds near the broadband availability target--actual download speeds of at least 4 Mbps and actual upload speeds of at least 1 Mbps--recommended in the National Broadband Plan.

Does a downturn in new broadband statistics point to market saturation? It sure seems as though major broadband carriers are content with their subscription numbers. For example, Comcast recently raised by $2 both its service tiers. In light of the comparatively high rates charged in the United States, $30-60 a month, a significant portion of Americans do not appear willing to pay. Alternatively, current economic conditions might have forced prospective users to hold back.

Lies, Damn Lies and Statistics at the Federal Communications Commission

The Federal Communications Commission recently discovered that 14 to 24 million Americans, located in 1,024 out of the nation’s 3,230 counties, do not have access to any broadband service at any price. This finding greatly contrasts with the Commission’s numerous previous statements that an unregulated and robustly competitive marketplace has provided universally accessible broadband at affordable rates just about everywhere. In reality the FCC could conclude that “broadband is being reasonably and timely deployed to all Americans” only by using false data.

It should come as no surprise that the FCC could so miss the mark on actual broadband access. The agency is awash in partisanship, pseudo science, fuzzy math, creative interpretation of economic principles and legal concepts, selective interpretation of the facts, innovative collection of statistics, and flawed thinking. These defects support results-driven decision making where FCC managers first reach a decision and subsequently support that outcome by framing the policy issues, “finding” facts and compiling data in ways that rationalize the preordained conclusion.

The FCC lacks the resources or resolve to compile a record independent of what parties with a financial stake file when the Commission seeks public comments. This means that the FCC does not have an unbiased, empirical record that would meet a threshold standard of fairness and reliability assessed by independent third parties, a process known as peer review. Because the FCC relies on data compiled by stakeholders, the Commission typically lacks the ability to differentiate credible research from “cooked books.” By relying on data compiled by the companies it regulates, the Commission regularly agrees to treat the information as proprietary, making it impossible for third parties to corroborate or refute the evidence used by the FCC to support its decisions.

In the case of broadband the FCC’s commitment to confidentiality has gone so far as to deem as “trade secrets” data about whether a carrier does or does not operate in a specific locality. Trade secrets typically refer to essential business information such as food and beverage recipes, but the FCC has managed to equate information about broadband accessibility with a company’s most essential assets. Bear in mind that the Commission must act on a congressional mandate to identify and remedy broadband access scarcity.

Notwithstanding a statutory obligation to track broadband access closely the FCC purposely overstated the scope of market competition and how well carriers had made service available. The FCC defined broadband in 1999 as a bit transmission speed of at least 200 kilobits per second in one direction. The FCC retained that now woefully inadequate bit rate until this year when it acknowledged that many Internet services require higher speeds. The Commission also used zip codes as the most focused geographical measure for broadband market penetration until this year. The Commission could reach its conclusion of 99+ percent market penetration by claiming “mission accomplished” for the entire zip code if at least one subscription opportunity existed somewhere within the zip code.

What statistics the FCC complies and how the Commission interprets the data has a substantial impact on how the agency interprets its regulatory mission. If the FCC wants to deregulate and abandon existing public interest safeguards, the Commission can claim evidence proves a robustly competitive marketplace can self-regulate. Until this year the FCC considered the wireless marketplace so competitive that the Commission could deem precompetitive numerous horizontal mergers where one competitor buys out another and acquires additional market share. The Commission’s most recent analysis of the wireless marketplace makes only passing reference to the fact that Verizon and AT&T national carriers have over a 60% market share, four national carriers control over 90%, and the rate of market concentration has grown in light of FCC-approved acquisitions so much so that it now well exceeds the Justice Department’s threshold for a “highly concentrated” market. The Commission also reports that U.S. wireless carriers enjoy healthy returns led by Verizon with an enviable 46.3% margin for the second quarter of 2009.

If the FCC wants to expand its regulatory wingspan, the Commission can claim evidence supports the need to curb market power. A former FCC Chairman, normally adverse to regulatory expansion, nevertheless wanted to further regulate cable television operators based on his perception that the industry had become too dominant. Using data, not compiled by FCC staff and highly questionable in light of market conditions favoring more competition from satellite and telephone companies, this Chairman believed that cable market penetration had reached a congressionally-drawn threshold. Neither the Chairman nor his staff could generate empirical data to support this conclusion.

The FCC can rely on poor fact finding only if reviewing courts accept such practices as worthy of judicial deference to the agency’s expertise. Some courts appear not to second guess the Commission, but others readily find flaws. Examples of the latter include a court’s refusal to allow the FCC to count as equals any media outlet in a market, regardless of significance and market share.

As information, communications and entertainment become an increasingly significant component in the economy, we cannot afford to have the FCC ignore instances where market self-regulation does not serve the national interest. The FCC has undertaken some recent efforts to improve its statistical compilations, but longstanding institutional flaws remain

Identifying Areas in the U.S. Lacking Any Broadband Options

Despite previous proclamations of near ubiquitous broadband access in the United States, using smaller and more numerous counties instead of zip codes and considering broadband to require far greater than the previous 200 kilo bits per second floor, the FCC now acknowledges that significant numbers of Americans residing in many largely rural areas with low incomes lack any access at all. [1] The Commission now acknowledges “that broadband deployment to all Americans is not reasonable and timely. This conclusion departs from previous broadband deployment reports, which held that even though certain groups of Americans were not receiving timely access to broadband, broadband deployment ‘overall’ was reasonable and timely.” [2]


The Sixth Broadband Deployment Report confirms that a sizeable number of Americans have no broadband access whatsoever, or have access that do not meet the National Broadband Plan goal of affordable service with download speeds of at least 4 megabits per second (“Mbps”) and upload speeds of at least 1 Mbps. [3] The FCC recognized the prior 200 kilobit per second rate, in either direction, “simply is not enough bandwidth to enable a user, using current technology, ‘to originate and receive high-quality voice, data, graphics, and video telecommunications,’ as section 706 [of the Telecommunication Act of 1996] requires of such services.” [4]

Using the higher bit rate threshold the FCC estimates that 1,024 out of 3,230 counties in the United States and its territories are unserved by broadband, [5] and between approximately 14 to 24 million Americans do not have access to broadband today. [6] The Commission makes a number of candid acknowledgements:

The . . . [unserved] group appears to be disproportionately lower-income Americans and Americans who live in rural areas. The goal of the statute, and the standard against which we measure our progress, is universal broadband availability. We have not achieved this goal today, nor does it appear that we will achieve success without changes to present policies. The evidence further indicates that market forces alone are unlikely to ensure that the unserved minority of Americans will be able to obtain the benefits of broadband anytime in the near future. Therefore, if we remain on our current course, a large number of Americans likely will remain excluded from the significant benefits of broadband that most other Americans can access today. Given the ever-growing importance of broadband to our society, we are unable to conclude that broadband is being reasonably and timely deployed to all Americans in this situation. [7]

As evidenced by the ambitious goals in the National Broadband Plan, the Commission aspires to do a better job of promoting affordable and ubiquitous access going forward.

[1] Inquiry Concerning the Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, and Possible Steps to Accelerate Such Deployment Pursuant to Section 706 of the Telecommunications Act of 1996, as Amended by the Broadband Data Improvement Act, GN Docket No. 09-137, Sixth Broadband Deployment Report, (rel. July 20, 2010); available at: http://hraunfoss.fcc.gov/edocs_public/attachmatch/FCC-10-129A1.pdf[hereinafter cited as Sixth Broadband Deployment Report].


[2] Id. at ¶2.

[3] See FCC, OMNIBUS BROADBAND INITIATIVE (OBI), CONNECTING AMERICA: THE NATIONAL BROADBAND PLAN, GN Docket No. 09-51 (2010) (NATIONAL BROADBAND PLAN); Inquiry Concerning the Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, and Possible Steps to Accelerate Such Deployment Pursuant to Section 706 of the Telecommunications Act of 1996, as Amended by the Broadband Data Improvement Act; A National Broadband Plan for Our Future, GN Docket Nos. 09-51, 09-137, 2010 W.L. 972375 (rel. March 16, 2010). See also, National Broadband Plan, World Wide Web Site, http://www.broadband.gov/plan/.

[4] Sixth Broadband Deployment Report at ¶10.

[5] Id. at ¶22.

[6] Id. at ¶28. The Commission previously reported that about 80 million Americans either do not have access, or do not subscriber to an available broadband service.

[7] Id.

Law Review Article on U.S. Broadband Statistics

Lies, Damn Lies and Statistics: Developing a Clearer Assessment of Market Penetration and Broadband Competition in the United States is available at:
http://www.vjolt.net/vol14/issue2/v14i2_100%20-%20Frieden.pdf.

Here's the abstract:

This paper examines United States broadband penetration and pricing statistics with a critical eye, in light of other contradictory compilations by organizations other than the Federal Communications Commission and the National Telecommunications and Information Administration. The paper also compares and contrasts the FCC’s identification of broadband options in the author’s home zip code with what actual options the author could identify.

The paper concludes that the U.S. government has overstated broadband penetration and affordability by using an overly generous and unrealistic definition of what qualifies as broadband service, by using zip codes as the primary geographic unit of measure, and by misinterpreting available statistics.

The paper concludes that credible calculations, using better-calibrated measures, show a mixed outcome based on a more granular geographical and cost focus. The paper provides suggestions on how the FCC could stimulate next generation network deployment.

Work in Progress-Lies, Damn Lies and Statistics: Developing a Clearer Assessmentof Market Penetration and Broadband Competition in the United States

Set out below is the abstract for a current work in progress on the dodgy world of broadband market penetration data collection. The work will appear in an upcoming edition of the Virginia Journal of Law and Technology. A current draft is available at: http://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=102928

Depending on the source one can conclude that United States consumers enjoy access to a robustly competitive and nearly ubiquitous marketplace for inexpensive broadband Internet access, or they suffer the consequences of a tightly concentrated industry offering inferior service at high rates. On one hand, the Federal Communications Commission (“FCC”), the National Telecommunications and Information Administration (“NTIA”) and some sponsored researchers offer a quite sanguine outlook, possibly influenced by their appreciation for the political and public relations dividends in compiling positive results.

On the other hand, other statistical compilations and interpretations show the U.S. behind in terms of market penetration and price, even trailing some nations that have similarly unfavorable geographical and demographic characteristics. In the light of the extraordinary global success achieved by domestic ventures in information and communications technology (“ICT”), it would appear counterintuitive for some current broadband statistics to show the United States lagging other nations in terms of favorable access to next generation networks.

The FCC has used evidence of robust market penetration and competition in broadband markets to support an aggressive deregulatory campaign. Advocates for even more deregulation regularly cite the Commission’s statistics as evidence that the unfettered marketplace can achieve broadband access and affordability goals. Both the Commission and many stakeholders assume the frequently cited statistics present a true picture of the marketplace. A recent NTIA document concludes that the United States has achieved the goal of “universal, affordable access for broadband technology by the year 2007” articulated by President Bush in 2004.

This paper will examine the United States broadband penetration and pricing statistics with a critical eye, in light of other contradictory compilations by credible organizations including the International Telecommunication Union and the Organization for Economic Cooperation and Development. Additionally the paper will compare and contrast the FCC’s identification of broadband options in the author’s home zip code with what actual options the author could identify.

The paper concludes that the FCC and NTIA have overstated broadband penetration and affordability by using an overly generous and unrealistic definition of what qualifies as broadband service, by using zip codes as the primary geographic unit of measure and by misinterpreting available statistics. Additionally the FCC includes as competition services lacking any true cross-elasticity with other services based on substantial price differences.
The paper concludes that credible calculations, using better calibrated measures, show a mixed outcome based on different geographical focus. Some U.S. residents, particularly in urban locales, enjoy comparatively excellent broadband service, while rural residents may have ample access options, albeit at comparatively high prices in light of limited price competition. The paper concludes that the absence of robust price competition among many facilities-based broadband operators in many areas of the nation challenges many of the assumptions built into recent FCC policy initiatives that seek to abandon consumer safeguards. The paper also concludes that a statutory mandate to promote universal access to advanced telecommunications capability requires the FCC to collect and disseminate credible statistics on next generation network deployment.

Broadband Statistics and the Lack of Transparency

Above is a prepresentative page constituting the sum total of the FCC's broadband statistics compilation. Might the FCC have both the incentive and the ability to overstate penetration? I'll answer that question with a big yes! The FCC currently uses a 200 kilobits per second threshold and counts an entire zip code as served should one e-rate or other subscriber exist. Even with a 768 kbps thershold and greater geographical granularity the overstate remains. First the FCC does not use actual, measured throughput, but instead relies on carrier reports. Carriers sharing the FCC's incentive to overstate success can claim to meet the 200 or 768 kbps threshold based on a theoretical possibility. So the FCC can reach double digits even in rural areas based on the bogus assumption that 2G terrestrial wireless and satellite broadband exceed a theoretical throughput floor.

Additionally the FCC makes no price comparison, so a triple digit monthly subscription rate offers no disincentive at least for counting options.

The FCC proves the adage that there are lies, damn lies and statistics designed to prove a mission accomplished.

U.S. Wireless: Cutting Edge or Inferior?

The FCC’s 12th Annual Report to Congress on the Commercial Mobile Radio Service offers an unconditionally upbeat assessment of the wireless marketplace in the United States. See http://wireless.fcc.gov/index.htm?job=cmrs_reports#d36e145. From start to finish, the Report contains summary conclusions that leave no doubt that “U.S. consumers continue to reap significant benefits – including low prices, new technologies, improved service quality, and choice among providers” (¶1) and they “continue to benefit from effective competition in the CMRS marketplace.” (¶290).

Has the FCC made a fair-minded and truthful assessment, or made yet another deliberate overstatement?

If the wireless marketplace has unquestionable characteristics such as robust facilities-based competition, then it follows that U.S. wireless consumers should benefit from best in class innovation and other dividends. Economists have convinced me that competitive necessity forces competitors to spend sleepless afternoons sharpening their pencils and marking down prices even as they work overtime thinking about how to capture market share by enhancing the value proposition of their service.

So if I’m supposed to join in a wireless lovefest here how could the Economist (July 12th edition), not known for false reporting, come up with a far less sanguine assessment in the context of Apple’s introducing a 3G iPhone:

PITY us poor mobile-phone users in America. While the rest of the world enjoys network speeds that let people watch television on the move, surf the mobile web in its living glory, download videos in a trice, or exchange video messages with one another, we celebrate Apple’s launch of its iPhone 3G today as if were some great leap for mankind. (available at: http://www.economist.com/research/articlesbysubject/displaystory.cfm?subjectid=7933610&story_id=11700916)

The Economist article mentions that the iPhone offers bitrate speeds in the 400-700 kilobits per second (“kbps”) range, throughput that hardly constitutes broadband except in the United States where the FCC still uses a 200 kbps threshold. Wall Street Journal columnist Walt Mossburg calculated the 3G iPhone bitrates at not terribly blazing 200-500 kbps, still an improvement over 70-150 kbps on AT&T’s old EDGE network.. See http://ptech.allthingsd.com/20080708/newer-faster-cheaper-iphone-3g/

I agree that wireless throughput offers only one benchmark for a reality-based assessment U.S. wireless performance. But for this criterion even the next best thing, using the best network available, does not come close to showing global best practices, or event true 3G performance.

I have no doubt that the FCC’s bogus broadband assessments include terrestrial wireless carriers in the Commission’s numerical count of broadband providers. But credible assessments show that U.S. 2.5G networks do not meet even the 200 kbps broadband threshold and current 3G networks do not match DSL speeds.

So much for the FCC’s undisciplined shout out to the wireless industry.

Slightly Less Deceptive FCC Broadband Statistics Forthcoming

Recognizing the need for better calibrated broadband statistics, which more closely tracks actual choices available to consumers, the FCC has made improvements designed to increase the precision and quality of broadband subscribership data collected. Rather than generally report on market penetration by any broadband service that offers 200 kilobits per second in one direction, the FCC has expanded the number of broadband reporting speed tiers to capture more precise information about upload and download broadband speeds. The Commission also will require broadband providers to report numbers of broadband subscribers by census tract, broken down by speed tier and technology type, instead of the much geographical region represented by a zip code. Additionally the Commission expects to improve the accuracy of information it gathers about mobile wireless broadband deployment.

Now the not so good news: achieving better broadband penetration in the United States increasingly is a matter of cost, not availability. One would think that with all the in-house and sponsored researcher economic help available the FCC would happen across the concept of CROSS-ELASTICITY, i.e., apples to apples comparison of similarly priced options. Put simply in the current broadband marketplace there are some options that offer comparatively slower bit rates at higher prices, e.g., wireless satellite and terrestrial options. Consumers opt for these services when they so value mobility that they are willing to make bitrate and price tradeoffs, or when they have no better options.

The FCC will still count broadband options regardless of price, so the bad news remains that the Commission’s statistics will overstate what options exist within the same price points.

Network Neutrality Unneeded in a Competitive Broadband Marketplace

In the Feb 25th edition of the Wall Street Journal former hedge fund manager Andy Kessler strongly suggests that competition would solve any calamity that network neutrality rules would (hamhandedly) attempt to remedy. See http://online.wsj.com/article/SB120390160543089503.html?mod=todays_us_opinion.
Mr. Kessler suggests the need for legislation to promote broadband competition, a Bandwidth Competition Act of 2008. Regretably, he offers few specific recommendations.
I heartily agree that in a perfect world facilities-based competition would enable consumers to “vote with their dollars” and punish any ISP foolish enough to handicap, delay, drop or otherwise meddle with their traffic. If I had a number of competitors from which to choose perhaps I could find a rate plans that provides a better fit than the current “one size fits all” model that motivates Comcast and other ISPs to rein in heavy users.
If only we had such competition. If only we had a legislature that would pay enough attention to the problem to direct the FCC and a Federal-State Joint Board to revamp efforts to jump start broadband deployment through a variety of strategies, including financial cross-subsidies.
Sponsored researchers will tell you that sufficient competition already exists thereby obviating both the need for network neutrality rules and legislation, unless of course the legislation steers money their clients’ way. According to the FCC’s latest statistics my rural zip code now has 11 broadband options, up from 9 last time. Accepting this number for the sake of discussion I still can conclude that robust facilities-based competition does not exist in my neck of the woods. Put another way access to $100 a month satellite semi-broadband and $60 a month cable broadband will not jump start broadband adoption. Nor will it create the critical mass we both agree would—if it existed—that would provide consumers and suppliers to come to terms on price and quality of service tiers.
Legislators and the FCC have to consider network neutrality rules in the absence of competition and diverging price points for broadband access.

Boring into Broadband Penetration Statistics

In preparation for a conference on network neutrality, I am taking a closer look at broadband penetration statistics in the U.S. and in other countries. I conclude that broadband policy should address both accessibility and affordability.

The U.S. has achieved a mixed record in broadband penetration not accessibility. In other words while some potential subscribers can access broadband at "best practices" rates, others have quite high charges to consider. Normalizing rates on a per 100 kilobit rate provides a good measure of affordability.

Promoting broadband in the U.S. going forward will have less to do with achieving geographic penetration and more with promoting lower rates. Devising a workable plan for subsidizing access is a daunting task. U.S. long distance telephone service callers contributed over $7 billion for promoting mostly narrowband, basic voice service affordability last year. The current universal service funding mechanism is expensive, flawed, prone to abuse and lacking a broadband component outside schools, libraries, and medical facilities.

Here's a link to my presentation entitled " Internet Access as Essential Infrastructure: Public Utility, Private Utility or Neither?":http://www.personal.psu.edu/faculty/r/m/rmf5/USF%20Network%20Neutrality%20Conference.ppt.