Patterns of broadband adoption
Two surveys of broadband adoption were released last week, and they both tell us we have a long way to go in closing the digital divide.
The FCC’s excellent survey, Broadband Adoption and Use in America, conducted at the end of 2009, shows that 35% of Americans do not use broadband at home. Of that group:
“➤ Gender: 57 percent of non-adopters are women versus 49 percent of home broadband adopters.
➤ People with disabilities: 39 percent of non-adopters have a disability, compared with 15 percent of adopters.
➤ College graduates: Just 11 percent of non-adopters have college degrees versus 37 percent of broadband users.
➤ Age: 32 percent of non-adopters are age 65 or older versus 9 percent of adopters.
➤ Nearly two-thirds (65%) of non-adopters who are senior citizens are women.
➤ Income: 43 percent of non-adopters live in households with annual incomes of $20,000 or less, compared with 17 percent of home broadband users.
➤ Rural: 24 percent of non-adopters live in rural areas versus 13 percent of broadband adopters.”
56% of non-users cited at least three reasons why they weren’t adopters. Here are some of the top reasons:
Monthly cost is too expensive 47%
I am not comfortable using a computer 46%
I am worried about all the bad things that can happen if I use the Internet 45%
The activation and installation fee to get service is too much 42%
I cannot afford a computer 40%
There is nothing on the Internet I want to see or use 35%
The Internet is just a waste of time 33%
I can access the Internet all I need to at work 14%
It’s not available where I live 13%
In general, the three most important reasons for non-adoption were related to cost (36%), “digital literacy” or comfort using computers and the Internet (22%), and relevance (19%).
Lower percentages of African-Americans - particularly older African-Americans - use high-speed access at home compared to all Americans (59% v. 66%). Just 49% percent of Hispanics have adopted broadband, again with older Hispanics at lower adoption rates. Cost is the big reason for both of these groups.
The Joint Center for Political and Economic Studies also released a report this past week. Called National Minority Broadband Adoption: Comparative Trends in Adoption, Aceptance and Use, the report also noted lower adoption numbers for low-income, older, and less-educated African-Americans and Hispanics.
Higher-income minorities are the fastest-growing groups of broadband adopters.
As compared to whites, minorities are more likely to use the Internet at community anchor institutions (libraries and schools) and more likely to use it to search for jobs.
According to the Joint Center, “For non-adopting minority respondents, a general lack of interest, followed by lack of accessibility and then high cost, are the primary barriers to acceptance and use.”
For all groups, comfort with computers and networks seems to grow with time online. Lack of interest may be based on lack of knowledge - and relevance may only be demonstrated by use.
Both of these reports demonstrate that there are deep and continuing demographic/cost reasons why adoption numbers remain stubbornly low in this country.
Big week
1. FCC says a major reason for low adoption of broadband in the US is price.
2. FCC also reminds us that there’s a looming spectrum crisis.
Let’s assume mobile phones providing high-speed Internet access require additional spectrum. But they’re also going to need much more deeply-available fiber - mobile communications need to jump quickly to a wire in order to provide true high-speed access. More infrastructure is needed.
Berkman reminds us that having much more deeply-available, open access fiber will also lower the price of high-speed Internet access connections. Competition is generally agreed to have a favorable impact on price. Lower price, greater adoption.
All of this seems to point to a serious focus on the provision of Internet access, and on the pro-competitive conditions under which this access is provided.
Meanwhile:
3. General-purpose transport providers are joining together to fend off any potential regulatory focus on any of this.
Videocracy
I spent part of today watching Videocracy. It’s a nightmarish film about the culture surrounding Silvio Berlusconi. Control over 90% of the medium - television - on which 80% of Italians rely for information adds up to significant cultural saturation.
Berlusconi himself, with his easy smile and his fascination with spectacle, isn’t the focus. Instead, the dead-voiced narrator describes the lives of others who are caught up in Berlusconi’s picture of the world: a mechanic desperate for televised fame, a pudgy TV producer in an endless celebrity-rich party, an extortionist trying to stay at the center of the story. There’s a sleek smugness in the one who has made it to the top - the producer - and a crazed absence in the other two. Thousands of young women crowd around the camera, longing to be showgirls. Women sing “Thank God Silvio exists!” in a campaign commercial.
It’s a deliberate, dreamlike, appalling film. You really ought to see it.
West Virginia Statewide Broadband
The recent Berkman broadband report found that the $7.2 billion in stimulus funding allocated in the U.S. for broadband was in line with what other countries are doing - indeed, “several countries [including South Korea, Japan, Sweden] have invested over the long term as a strategic choice rather than as a stimulus measure.” The U.S. is spending more in the stimulus mode than other nations, but less than the most publicly-funded nations - who are (by far) the leaders in fiber deployment.
Berkman (at 232): To the extent that one sees the long-term trajectory of the fixed element of next generation networks to be in fiber closer to - and ultimately at - the home, we can perhaps say that substantial government investments seem to be associated with approaching that goal more rapidly.
This week, the Department of Commerce announced a whole slew of broadband grants using stimulus funding. There’s a particularly exciting $126M grant to West Virginia (factsheet) included. WV is a large black spot on the map of broadband availability in the US, and this grant will make a big difference. Listen to what’s involved:
The West Virginia Statewide Broadband Infrastructure Project plans to bring high-speed Internet access to this vastly underserved region by expanding the state’s existing microwave public safety network and adding about 2,400 miles of fiber. The expanded statewide network expects to directly connect more than 1,000 anchor institutions, including public safety agencies, public libraries, schools, government offices, and other critical community facilities at speeds of up to 45 Mbps. As a result of this project, every K–12 school in the state will have a high-speed Internet connection. In addition, access to healthcare, distance learning opportunities, and broadband and video applications for emergency first responders will be greatly expanded. The project intends to spur affordable broadband service impacting more than 700,000 households, 110,000 businesses, and 1,500 anchor institutions, by allowing local Internet service providers to connect to the project’s open network.
Almost 500 schools, 200 libraries, a whole bunch of 911 answering centers, 200 telemedicine sites, and nearly 800 law enforcement offices/fire departments will benefit from this grant. That’s substantial and transformative for West Virginia.
E-Rate sense
The US spends $2.25B annually on its E-rate program, supporting schools and libraries that provide Internet access and other related services. E-rate is an important part of US communications policy - it hasn’t always been smooth sailing, but the program has led to substantial internet adoption by these crucial anchor institutions.
Today the FCC said it would waive rules that require schools to certify that they will use E-rate funded services solely for “educational purposes,” defined as activities that are integral, immediate, and proximate to the education of students. These rules have forced schools to carry out elaborate calculations focusing on their provision of internet access - with the consequence that school Internet access facilities that rely on E-rate funding stay closed during non-school hours.
Giving schools the option to open their E-rate funded facilities to members of the public during non-operating hours makes a lot of sense. As Chairman Genachowski said today,
These connections will be available to adults taking evening digital literacy courses, to unemployed workers looking for jobs posted online, to citizens using e-government services, and for other uses that local schools believe will help their communities.
The “national purposes” presentation at today’s FCC meeting (foreshadowing next month’s National Broadband Plan) focused helpfully on education. Broadband access could help personalize instruction so that students learn more, but there’s a big gap in high-speed access availability to meet the demands of schools and students. Upgrading E-rate to set higher goals for school and library connectivity and to support more internal connections makes sense. So does allowing off-hours community use of school Internet facilities. Very sensible.
American exception: telecommunications policy
During 2007-2008, The New York Times ran a terrific series of articles by Adam Liptak about legal stances that the US takes alone: convicting juveniles as adults and sentencing them to life, making accomplices as liable as the killer for murders committed during felonies, allowing bail bondsmen to do business, having less than 5 percent of the world’s population but almost a quarter of its prisoners, electing judges, hiring partisan expert witnesses..
Today’s release of the Berkman Center’s thoughtful and exhaustive report, Next Generation Connectivity: A review of broadband Internet transition and policy from around the world, adds high-speed Internet access policy to the list of American exceptions.
Some of the report’s highlights:
- Across many sets of rankings published by a variety of actors, the US is a weak performer on prices for high-speed access. Prices in this country are high and speeds are relatively slow. US performance has declined relative to other countries.
- There is a very broad consensus in developed countries outside the US that open access policies requiring shared infrastructure are both necessary (to competition, ubiquity, lower prices, higher speeds) and work. These policies prompt competition by lowering barriers to entry, thus enabling lower prices and higher speeds.
- Greater fiber penetration is a goal of many countries, because of its vastly higher capacity and upgradeability.
- High-speed Internet access is related to economic growth.
- Incumbents all over the world always complain about open access, but professional, persistent regulators succeed.
- There are many different ways to implement open access.
- The theory that unbundling deters investment is not proven by either empirical or theoretical literature. Nor have the theories that attempt to explain why unbundling works been proven. But countries that have effectively pursued open access have done better than the US has over the last few years - so it’s very valuable to compare the case studies in those countries to our situation.
This report is an enormous contribution. It shows that the US has much to learn from the rest of the world.
The Canoe tipping point
Google has successfully created a nationwide (worldwide) fine-grained, targeted ad market by using queries to its search engine. The cable industry would like to be able to use its cable and broadband subscriber data to create a nationwide, fine-grained, targeted ad market. This race has substantial implications for the future of online video - and online activity generally. Large cable operators have a major advantage: they are tied to TV sets and set-top boxes, which are closer to consumers’ video consumption (right now, at least) than PCs.
Right now, most people consume video over television sets, and most people in the US subscribe to cable. The TV advertising market is still doing pretty well - about $70B in 2010. TV is still the dominant mass medium. But cable operators don’t capture very much of that total $70B - maybe only $5B goes to them.
With (1) an intermediary providing anonymized but complete data about what all cable subscribers are up to at all times (including their broadband usage), (2) standardized technology for ad insertion for all major cable systems, and (3) a “return path” (a way for users to interact with advertisers and others), the cable industry could achieve the Holy Grail: household-by-household targeted advertising via must-have cable set-top boxes.
Instead of a fragmented market in which nationwide advertisers have to sell to multiple cable operators and programmers (and TV networks) without useful metrics, you’d have a dream interface for cable advertising - nationwide, interactive, one-stop shopping for advertisers, using actual data rather than relying on Nielsen-like guesstimates. Clutching our remotes, we consumers will click around giant TV screens, buying products shown on cable network shows, viewing ads targeted to our household, and responding to well-organized stimuli designed to appeal just to us.
This dream is what the six largest cable operators, led by Comcast, have been working towards since 2008. It’s called Canoe Ventures. So far, Canoe hasn’t quite delivered. It had to drop 2009 plans to provide targeted ads because it was held back by clunky legacy cable ad-insertion hardware. But it’s planning to offer RFIs (allowing viewers to ask for information) later this spring, and Comcast is already piloting this technology in Chicago, SF, and Detroit. About 12 million Comcast households are technically ready for interactive ads.
Here’s the CEO of Canoe, David Verklin, in a June 2008 article in Multichannel News: “At some point, Canoe will combine TV viewing metrics with Internet surfing data to provide advertisers an integrated view across both platforms.” The success of Canoe is central to the future success of cable.
Canoe has to move fast. Online ad spending is climbing rapidly. Internet video — which hasn’t yet taken off - could destroy cable. But if Comcast and its colleagues find a way to corner the mainstream advertising market by making ads much more effective through their set-top boxes, online ad-supported TV and other video won’t work.
It’s called Canoe because all the major cable operators are in the same boat. They’re trying to row together.
Here’s a quote from 2009 from a programmer, Discovery Communications CEO David Zaslav:
In the industry, there are a lot of things we debate, but there is no debate about Canoe, Discovery Communications CEO David Zaslav said. It’s hard to raise your hand and say that having a platform you can customize isn’t anything but a positive.
Zaslav also commented on the potential threat of online video to operators and programmers. Discovery, he added, has opted to put mostly short-form clips of its shows on the Internet for free, a move to drive traffic to the television. But he said that while operators and programmers are in the same boat, they can’t ignore that a shift in viewing habits is beginning.
We have to put our content on platforms people are consuming, Zaslav said, adding that part of the problem was created by programmers who rushed to put all of their content on the Internet for free. He said that while that model obviously doesn’t work for everyone, all parties should be wary of what they ultimately decide.
We have to be careful we don’t train people to view on platforms that are going to put us out of business, Zaslav said.
More on monopoly maintenance
Thanks for the comments and emails yesterday.
1. According to the DC Circuit, we can infer monopoly power from a firm’s possession of a dominant share of a relevant market that is protected by entry barriers.
- Here, we’ve got a dominant player in the local wired high-speed Internet access market. Nothing is really substitutable right now for local wired access - wireless isn’t fast enough, and nothing’s quite like Internet access. In formal terms, there are no substitutes that can reasonably constrain the pricing of wired high-speed Internet access in the reasonably foreseeable future.
- It’s a natural monopoly, because in the last mile the costs of providing services to each house decline dramatically given greater scale. No one really wants more than one wire to a home.
- Entry barriers could include access to rights-of-way granted by municipalities.
- Rates for cable service are said to have been going up - what about rates for broadband service provided by this cable company? Hard to tell with all the bundling, but let’s try to figure this out.
2. The issues for high-tech network products (tipping, lock-in) are arguably much greater for physical infrastructure than for applications. Applications come and go, and users can always leave (”with a click of the mouse,” as the saying goes). But there are very few choices of physical gateways to the Internet these days in this country. Transport really is different. Plus, high-speed Internet access consumers are confronted with sticky bundles of services that can’t be taken apart and are difficult to switch away from. Incompatible hardware, high switching costs - that’s where tipping and lock-in can really take effect. Meanwhile, search engines, online VoIP providers (independent of the network providers), and social networking gambits burst on the scene and disappear later. As a matter of both kind and degree, physical infrastructure is different from the application layer. (Watch out for intentional munging using the phrase “the Internet ecosystem.” These things - transport and applications - really are different.)
3. What procompetitive justification does a natural monopoly provider of high-speed Internet access service have for merging with must-have content? What effiencies are created? It can’t be that the answer is “saving that great American icon, NBC.” That’s patriotic (and believe me, I feel the tug) but hardly efficient. Even if there are efficiences, aren’t they outweighed by the anticompetitive harm of the conduct? That harm includes foreclosing competition from both other cable operators and Internet delivery of network content.
4. All you have to show is that this anticompetitive conduct “reasonably appear[s] capable of making a significant contribution to . . . maintaining monopoly power.” We can infer causation “when exclusionary conduct is aimed at producers of nascent competitive technologies as well as when it is aimed at producers of established substitutes.” Here, both the nascent (Internet video) and the established (smaller cable companies) are being aimed at.
Keep those notes coming.
Monopoly maintenance
Maybe it’s time to look back at the D.C. Circuit’s opinion in U.S. v. Microsoft. The argument takes a few steps to outline - here’s a first try:
1. High-tech markets are different from traditional manufacturing markets. Products that get an early lead in high-tech markets are likely to stay ahead, because consumers want the goods and services that others have already bought or may buy in the future. High-tech markets may tip easily, locking in consumers with high switching costs.
2. If you’re sensing anticompetitive behavior in a high-tech market, you’ll need to act quickly. Otherwise it may be too late. No one will invest in the path of a dominant actor.
3. Let’s say you’ve got a dominant player in local broadband markets who also provides cable service. That player can use its dominant position in broadband to enter the content marketplace.
4. Even if the broadband actor doesn’t control content - content is hard, but not impossible, to dominate - it can do many things in the content market to avoid commoditization of its broadband service.
- it can raise prices for must-have content for its rivals, who have to provide bundled services in order to survive
- it can seamlessly technically intertwine must-have content with its broadband service, giving it an advantage that no one can match, driving adoption of its bundled services
- it can favor its content over broadband content
- through its purchasing power, it can influence content providers to constrain how they distribute programming over the Internet
5. The DC Circuit said something similar in Microsoft, finding that the company illegally maintained its monopoly in the operating systems market through its actions in the browser marketplace.
6. It’s not clear what efficiencies are created by combining a dominant broadband provider and a must-have content company.
Fire away - all comments welcome.
Disruptive nudge: Google fiber
The only news to pay attention to today in my world is Google’s announcement that they’ll be putting money into experimenting with open-access fiber-to-the-home 1Gps networks at some point soon. They’ll open up an RFI - they want to hear from mayors.
This is a truly significant announcement.
- We’ll learn so much about how much it really costs to bring high speeds to communities. We won’t have to rely on the carriers’ doomsaying about these expenses (”Hundreds of billions of dollars!”).
- We’ll learn what applications people want to use. Right now the network providers can say that the market isn’t clamoring for high speeds - but that may be because high speeds aren’t available.
- We’ll learn how easy it is to have competitive providers attach to these fiber networks. Until just a few years ago, all of our general purpose networks were required to allow this kind of attachment. Then the US abandoned this approach at the behest of our incumbents while other countries moved down this path with great success.
- We’ll see this model supported by a great brand. Several municipalities are doing this, but the Google move will get more attention.
- We’ll stop being content with minimal baseline speeds for America.
It’s a tremendous move. Even if Google ends up serving just 50,000 people, the game in the US will have been forever disruptively nudged.