InfoTech & Public Policy

In class a few weeks ago, reference was made in passing to the “digital divide.”  In 2006, what does that mean?

The digital divide can be subdivided many ways.  Initially it referred to access to computers and the Internet.  Globally, that’s still a major concern, where even $100 laptops don’t help the 2.5 billion people who live on less than two dollars a day.[1]  But here at home the issue has been redefined as bandwith access. Others apply it to technology literacy.  With such varied definitions, it’s worth considering which issues are worthy of the outcry the term has generated. 

First, consider access to computers and the Internet.  As Robert Samuelson pointed out way back in 2002, the argument that society was segregating into technology “haves” and “have nots” that would result in greater economic inequality was wildly exaggerated.[2]  From the late nineties through the early twenty-first century computer prices fell dramatically, and Internet access spread across demographic groups.  This was largely a result of private enterprise.  Government intervention was most effective in connecting schools and libraries to the Internet, and by 2000 alone there was approximately one computer for every four students in public schools.

Soon after, the “digital divide” was redefined to refer to access to broadband Internet service. By this formulation, rural communities and minorities are being excluded from the information economy.  Narrowing those gaps though government spending is good politics, but is it good policy?

It’s hard to see how in the case of rural communities.  The telephone took 67 years to reach 75% of households.[3]  In contrast, 73% of Americans are already online, an increase of seven percentage points from January 2005.  Many rural Americans are currently served by dial-up access, and more get high speed every day.  42% of Americans have broadband connections at home, up a significant 13% from last year.  Note the broadband statistic doesn’t take into consideration those with high speed access at work or those that don’t need or desire broadband access. 

While Members of Congress trip over each other to fund rural assistance programs, the private sector is developing cost-effective means to access the Internet in rural areas.  Consider the development of wireless Internet access.  At least one major study found that WiFi access is financially viable in rural areas.  In short, Congress should be more selective in deciding when to intervene in the market.

The digital divide is also closing for minorities.  Since 1998, Internet usage has almost doubled among African Americans to 61%.  Over the same time period, usage by whites is up 32% to 74%, and up the most by English speaking Hispanic-Americans, from 40% to 80%.[4]  Unsurprisingly, the debate is now shifting again, this time to “quality of access.” 

Among minorities and low-income individuals, the case for government intervention is stronger, and much of the action is taking place at the municipal level.  But as with the original access issue, some policies are better than others. 

For example, a program in LaGrange, Georgia that offered free internet access to every citizen achieved a penetration rate of only 40%, and has since been terminated.  Milwaukee, Minneapolis, Philadelphia, San Francisco and many other cities are racing to build citywide wireless networks, with “bridging the digital divide” as a key selling point.  For such initiatives the jury is still out, but financing and pricing will be a key determinant of success.  For its part, Baltimore is creating incentives for low-income housing developers to install Internet hardware in the units they build.[5] 

Of course, even defining broadband access is controversial.  Many commentators have pointed out that the FCC overstates broadband penetration, by for example defining broadband as speed in excess of 200 kbps.  Also, access is always relative in today’s economy.  The US is ranked a disappointing 16^th  in the world for number of broadband subscriptions per 100 inhabitants.  

In conclusion, the problem with sketching the digital divide debate in such broad terms is that it detracts attention from the issues that really do matter.  The federal government should focus its efforts on providing what the private sector can’t:  an educated workforce that can compete globally.  Guaranteeing high speed Internet access to every last rancher in Wyoming is not nearly as important as making sure the residents of LaGrange have the ability to take advantage of the Internet to improve their well-being. 

“Another blow for TV advertising” screams the headline.  According to a recent article on e-marketer.com, Network TV advertising spending in the U.S. declined by 1.5% in 2005.  But look a little closer at the numbers from Nielson Media Research, and you’ll find that Cable TV advertising dollars actually increased 11%, and Spanish-language TV ad spending went up by a whopping 16.9%.  Now a company that makes money by “helping businesses make sense of the e-business numbers and trends” might be expected to generate self-serving headlines.  But the point remains the same- the death of television advertising, like the global oil crisis, Linux and Vince Carter, is overhyped. 

Topping the suspect list of television advertising’s supposed slayers is digital video recorders like those marketed by TiVo, which allow consumers to record shows and fast forward through commercials.  Now that we finally have some data, it turns out TiVo’s more of a paper cut than subdural hematoma.  In the last week of 2005, Nielson began measuring time shifted viewing.  They found that only 12% of households have DVRs.   More importantly, they found that 70%, that’s 7-0 %, of viewing is live.  Since you can only fast forward during DVR playback, the vast majority of DVR viewers are still exposed to ads- even assuming that 100% of viewers watching recorded shows skip 100% of commercials.  Even if DVR growth eventually meets expectations, this data suggests TV advertising will make out just fine.  Also keep in mind that early adopters are more likely to take advantage of DVR’s time-shifting features- so that 70% number may actually increase as less tech-savvy viewers buy in.

Another defendant is Slingbox, the place-shifting technology that enables viewers to watch their television on portable devices like laptops.  The main threat from Slingbox seems to be that it jeopardizes a decades-old business model that depends upon local-market viewing and retransmission-consent agreements.   That concern is overblown.  Very few Slingbox viewers in California will be shipping their unit off to a distant relative in New York to get Oprah on their lunch break.  Instead, they’ll be using it to watch the TV they’ve already paid for in their bedroom or at the airport.  If sharing cable subscriptions between friends becomes an issue, Slingbox accounts could be limited to those living in the house in which the cable service is provided.  We’ve outlawed cable splitting, and it is intellectually consistent to limit Slingbox sharing to household members. 

This is not to say that television advertising isn’t changing.  It’s simply to point out that many have a stake in heralding its decline and that the changes aren’t coming as fast as promised.  Smart executives will work with TiVo and Slingbox, not against them.  In the meantime, the major networks have many resources at their disposal to make up that 1.5%.  From product placement to split screens to video iPods to streaming video to cross-platform advertising to interactive television (incidentally being pioneered by TiVo through ad-tagging), the only limit to greater profits is the bounds of networks’ creativity.  And isn’t creativity what advertising’s all about?

The Iraq war causes it.  So does tobacco ($92B),[1]  vision impairment ($10B),[2] bank holidays ($1.7B),[3] workplace interruptions ($588B), lethargy ($2.8B),[4] coffee breaks ($1B),[5] workers over 50 (millions of dollars),[6] Star Wars: Attack of the Clones ($319M),[7]  major league baseball playoffs ($225M in a week) [8] and March Madness ($3.8B).[9]  What we’re talking about, of course, is lost productivity. 

The fact that most lost productivity figures are cooked by self-serving consultancies and interest groups is well documented.  For example, one outplacement consulting firm was responsible for the last 3 estimates.  I won’t mention them by name, because publicity is their motivation for issuing the numbers.  Suffice to say their math has more holes than a colander.

Such rampant manipulation undermines valid lost productivity concerns.  But the reason it’s so easy to color the numbers is that there’s lots of room for shading. 

Calculating productivity sounds pretty straightforward- divide GDP by the number of hours worked.  Of course, it’s not.  Even putting aside all the problems in getting a precise measure of the total value of all final goods and services in an economy, there is still a lot of room for error in calculating the denominator.  The federal government’s estimates assume worker hours barely changed over the 1990s.  How many people do you know who would agree?

Information technology (IT) is often cited as a source of America’s “productivity miracle.”  And who’s to doubt that e-mailing, the Internet, and spreadsheets have made everyone more efficient? 

Writing in The New Yorker six years ago, John Cassidy made a strong case that the productivity gains from IT have been oversold.[10]  For example, he cites evidence that just three industries account for 4/5 of all IT business spending, and those industries have actually seen slow productivity growth for years.  Also referenced is the classic I.B.M. study that found that letters written on a word processor are modified, on average, 5 times more often than handwritten or typed letters, with no noticeable difference in quality.  In the same vein, a recent literature review found the benefits of telecommuting to be vastly overstated.   

This is not to say computers don’t make people more productive, or that their viruses don’t harm the economy.  It is simply to point out that there’s lots of room to measure how much and in what ways, and that the methods used will have a major impact on how one decides if the market is efficiently addressing the damages caused by lax computer security.

For good measure, here’s a statistic on productivity lost to Spyware:  $793 million annually in the UK alone.[11]  It comes from Internet security firm Webroot.  Clearly one of the few certainties in this debate is the need for more independent research.

This week’s Business Week has a cover story on the economics of Second Life.  The first slide show is worth checking out. 

The debate on network neutrality comes down to this:  is their significant broadband competition?  If access tiering becomes an issue, a competitive market will respond to demand for neutral platforms.  Significant choice in high-speed service will allow customers unhappy with access tiering to change providers, and network neutrality’s fate will be decided by the market.  Problem solved.

But the market can only respond efficiently if it is competitive.  Is it?  As expected, it depends on who you ask.  Testimony at the recent Senate Commerce Committee hearing on net neutrality was split.[1]  According to the US Telecom Association, “There is vigorous competition…”  The National Cable and Telecommunications Association agrees, citing a “broadband marketplace characterized by robust competition.”  In contrast, Google cites 2004 FCC figures that show “…only 53 percent of Americans have a choice between cable modem service and DSL service.  Of the remaining customers, 28 percent have only one choice, and 19 percent have no choice at all.”  Lawrence Lessig concurs, arguing “an effective duopoly controls access to high speed Internet.”

So who’s right?  I can’t figure out how Google and Lessig came up with their numbers from the 2004 FCC data, and the Associations don’t provide any quantitative data at all.  But since the hearing, the FCC has released new statistics on high-speed Internet access.  Data for 2005 shows that 25% of Americans have a choice between providers of asymmetric DSL and/or cable modem service.  26% have no choice and 15% have no service at all.  Only approximately 34% of Americans have more than one choice. 

Two things to note.  First, Google frames it as a choice between cable and DSL, and these statistics only provide and/or info.  Second, the access number surpasses 50% when all technologies are considered.  Google and Lessig may have good reasons beyond spin for not including wireless, satellite, power line and other high-speed lines, but I don’t know what they are.

Either way, it’s tough to argue that consumers have real choice in broadband access.  For that reason, safeguards should be implemented to protect network neutrality until consumers have more choices. 

In the meantime, there are a variety of options for cable and telephone companies to generate the revenues necessary to build bigger pipes.  Fees could be increased for all companies to connect to the Internet. Differential pricing already exists for connection speeds, and providers could also raise fees for the fastest access levels- generating new income from those who are most likely to use lots of bandwidth and benefit from the upgraded networks.  Some will cry foul and argue in favor of shifting those costs to online companies through access tiering.  But that doesn’t relieve consumers’ pain, because ultimately the access fees will get passed on to you and I. 

It remains to be seen if access tiering will harm Internet growth.  However, one thing’s for certain under the scenario described by Ashley- as competing nations roll out higher-bandwidth, neutral platforms, America’s best case for Internet innovation is just keeping up. 

Are governments prepared to deal with podcasting?  A quick Internet search suggests not.  The British government’s March 2006 white paper on the future of the BBC mentions podcasting once.  That’s one more time than the FCC, whose website search engine generates no hits for podcast or its variants.  For comparison, regulation yields 30,862 hits, and even VoIP (Voice-over-Internet Protocol) can be found 334 times.  But maybe we’re asking too much.  After all, in late 2004 Googling podcast still generated the result “Did you mean: broadcast?”

Depending on what you read, podcasting is either the death knell of television, the savior of radio, or both.  It certainly has its advantages.  Podcasts can be listened to at any time, any where.  They aren’t limited by spectrum.  They can be updated automatically through a subscription feed.  And to date, they are unregulated by the FCC. 

But should podcasts be regulated, and if so, how?  For China, the answers are clearly yes, and by all means possible.  That means beyond the Chinese government’s extensive policing and monitoring of the Internet, US companies will be asked to censor their products further.  Google.cn restricts e-mail messaging and the ability to create blogs.[1]  Expect podcasting to follow very soon, if it hasn’t already. 

For democratic countries, the answers are less clear, though there is good reason to believe governments will try.  After allowing Internet communication technologies to develop on their own for a couple years, regulatory agencies are starting to flex their muscle.  Last year the FCC gave VoIP service providers 18 months to adapt their products for government surveillance.[2]  In Canada, the Radio-television and Telecommunications Commission is requiring satellite radio services to provide 10 per cent Canadian content in their broadcasts to receive a license.[3]  Finally, the Federal Election Commission (FEC) recently voted to allow almost all political activity on the Internet.[4]  Many bloggers had been concerned that the FEC would consider their writings to be part of political campaigns, and thus subject to regulation.  

Podcasting raises a question central to regulatory agencies’ work:  To what extent should they act to anticipate new developments?  Act too slowly and they risk being out of touch.  Act too quickly and they risk stifling innovation.  

Clearly a balanced approach is necessary.  For the FEC, that means they should have examined podcasting during their investigation of the political regulation of blogs.  For the FCC, that means supporting self-regulation for now.  The government should allow more time to see how the technology develops and is adopted.  Most podcasts are downloaded from a few major sources, such as iTunes, Odeo.com and Podcast.net., and those sites have an interest in self-policing their content, though that’s not to say that they do.  

Another consideration for the FCC is indecency.  The commission decided against applying indecency rules to Satellite Radio because the subscription-based services do not provide indiscriminate access to children.  Does the same logic apply to podcasts?  It certainly does in the case of fee-based podcasts.  But what of the millions of free podcasts being downloaded every day?  

Podcasting clearly raises many challenging questions for governments.  Now is the time to start answering them. 

Christian and I have done some research. It complements JDF’s research. Sorry, my hyperlinks aren’t working on this computer.

Here is an excellent resource for acquiring more information regarding our proceeding:
https://gullfoss2.fcc.gov/prod/ecfs/comsrch_v2.cgi
enter proceeding: RM-11277

The 81 records filed to date for our proceeding can be found there. Most are brief comments, but a couple are longer. We may wish to divide them up to read them all.

Note the following information from the Wireless Association (CTIA) brief (02/02/06)

The fact that DATA BROKERS APPARENTLY HAVE BEEN ABLE TO BREAK AND ENTER CARRIER CUSTOMER SERVICE OPERATIONS TO OBTAIN CALL RECORDS has given our industry a black eye…

Overwhelmingly, THE VAST MAJORITY OF CELL PHONE RECORDS ARE BEING FRAUDULENTLY OBTAINED THROUGH THE USE OF “PRETEXTING,” which is nothing more than lying to obtain something you aren’t entitled to procure lawfully. Allow me to explain how these data thieves operate. For the sake of illustration, if someone — and in most cases it appears to be a private investigator – wants to acquire my call records, the private investigator will go to a website that publicly offers to obtain such records such as locatecell.com. The person trying to obtain my call records will provide the website in most cases with nothing more than my name and phone number. At that point, the website or a subcontractor of the website will pose as Steve Largent and call a carrier’s customer service department to get the records. Customer Service Representatives (CSR) are trained to require more than just a name and phone number, but the thieves are well trained too and often badger, threaten or plead with the CSR to acquire the records as if they are the actual customer.

THE CTIA’S SUBMISSION OF 11/04/05 PROVIDES AN OVERVIEW OF CUSTOMER SERVICE PROTECTIONS EMPLOYED BY PROVIDERS.

NOTES ON OUR SUBMISSION

We assume we will file electronically by filling in the cover page and attaching a document.
http://www.fcc.gov/cgb/ecfs/welcome.html

Note the following from the electronic filing user manual

This process contains three phases: (1) Completing a cover sheet, and (2) Attaching documents or submitting typed comments, and (3) Receiving a Confirmation.

Please do not attach documents that are password protected, have redlined markings, macros enabled, formatted as read-only, or have automated links to source documents that are not included in your filing.

PDF documents: Although ECFS prefers documents to be submitted in PDF, you must remove all “Document Security” used to encrypt a document before attaching the file.