The Cost of Connectivity 2012
Policy Paper
July 19, 2012
Below, you will find the full text of the Open Technology Institute's report, "The Cost of Connectivity," which compares the prices of high-speed internet in 22 cities worldwide. To download a PDF of the report and its appendices, click here. To download an Excel file containing the full data set (also embedded at the bottom of the page) click here.
Report Summary
In this study, we compare high-speed Internet offerings in 22 cities around the world by price, download and upload speed, and bundled services. We have included some of the most relevant findings from our research in the report that follows, as well as a discussion of policy recommendations for the U.S. The report includes:
A comparison of "triple play" offerings that bundle Internet, phone, and television services;
A survey of the best available Internet plan for approximately $35 USD in each city;
A comparison of the fastest Internet available in each city.
The results indicate that U.S. consumers in major cities tend to pay higher prices for slower speeds compared to consumers abroad. For example, when comparing triple play packages in the 22 cities surveyed, consumers in Paris can purchase a 100 Mbps bundle of television, telephone, and high-speed Internet service for the equivalent of approximately $35 (adjusted for PPP). By contrast, in Lafayette, LA, the top American city, the cheapest available package costs around $65 and includes just a 6 Mbps Internet connection. A comparison of Internet plans available for around $35 shows similar results. Residents of Hong Kong have access to Internet service with symmetrical download and upload speeds of 500 Mbps while residents of New York City and Washington, D.C. will pay the equivalent price for Internet service with maximum download speeds that are 20 times slower (up to 25 Mbps and upload speeds of up to 2 Mbps).
The results add weight to a growing body of evidence that suggests that the U.S. is lagging behind many of its international counterparts, most of whom have much higher levels of competition and, in turn, offer lower prices and faster Internet service. It suggests that policymakers need to re-evaluate our current policy approaches to increase competition and encourage more affordable high-speed Internet service in the U.S.
Introduction
Around the world, high-speed Internet access continues to increase in importance as a vital means of communication. Consumer demand for faster residential and mobile Internet access is growing at a rapid pace in both the United States and abroad. At the same time, the United States’ international ranking in broadband speed, pricing and availability has been slipping in recent years, according to the yearly data gathered bythe Organization for Economic Co-operation and Development (OECD) on global broadband markets.[i] Other examples of the U.S.’s decline can be found in Akamai's quarterly "State of the Internet" report, which analyzes connection speeds, Internet penetration, broadband adoption, and mobile usage, among other factors. In the fourth quarter Akamai report for 2011, the U.S.had fallen to 13th internationally in average measured connection speed, behind South Korea, Japan, Hong Kong, the Netherlands, Latvia, Switzerland, Ireland, the Czech Republic, and Romania, among others.[ii]
Public interest and consumer advocates have cited the U.S.’s recent decline in global competitiveness as evidence of the need for policy reforms that will spur greater competition and investment from the nation’s Internet providers.[iii] Meanwhile, opponents of this view have dismissed these international rankings as deeply flawed and akin to comparing apples and oranges. They cite the United States' significantly lower population density and larger geographic size relative to other nations at the top of the rankings,which tend to have smaller land areas and greater population densities.[iv] Both of these general qualities can make deployment of high-speed Internet infrastructure more cost-effective.
In this report, the Open Technology Institute examines how residential high-speed Internet services in U.S. cities compare to offerings in other cities around the world. The report focuses on the following cities:
- Amsterdam, Netherlands
- Berlin, Germany
- Bristol, Virginia, United States
- Bucharest, Romania
- Chattanooga, Tennessee, United States
- Copenhagen, Denmark
- Dublin, Ireland
- Hong Kong, China
- Lafayette, Louisiana, United States
- London, United Kingdom
- Los Angeles, California, United States
- Mexico City, Mexico
- New York, New York, United States
- Paris, France
- Prague, Czech Republic
- Riga, Latvia
- San Francisco, California, United States
- Seoul, South Korea
- Tokyo, Japan
- Toronto, Canada
- Washington, DC, United States
- Zurich, Switzerland
We selected a number of cities that ranked highly in surveys of Internet speeds, including many at the top of the list in the Akamai report, as well as major cities with relatively similar population sizes and densities. We also chose to include three smaller U.S. cities which have municipal broadband networks: Bristol, VA; Chattanooga, TN; and Lafayette, LA. These cities currently offer some of the fastest Internet connections in the U.S. because the local communities have invested in and built their own communications infrastructure.[v]
Currently, there is very little publicly available data on worldwide prices for high-speed Internet access. The OECD collects certain pricing information about its 34 member countries, but its data does not provide the same level of depth in terms of variables recorded or number of competitors surveyed as this report.[vi] Even in the U.S., actual data on the cost of Internet connectivity is largely unavailable to the public and policymakers. The National Broadband Map maintained by the National Telecommunications Information Administration (NTIA) only includes data on Internet speeds and availability, and does not list any pricing information. The Federal Communication Commission (FCC) also does not collect or disclose any information related to the cost of residential or business Internet connectivity for the nation’s service providers, even though this information is vital for determining the state of competition and whether the market is working for consumers. The pricing information that does exist is compiled by industry and marketing analysts and is only available for tens of thousands of dollars.
This report compares the prices and service plans available to consumers in different cities. We focused on collecting data at the city level to avoid the complications that arise when averaging out and comparing connectivity at the national level. Our researchers simulated the consumer experience by visiting the websites of providers in a given city. While collecting data about international cities, we found translating foreign languages and websites to be a challenge in some cases. However, it was often domestic Internet service providers (ISPs) from whom it was the most difficult to obtain pricing information. Details such as download speeds, data caps, and installation fees were not always clearly displayed. In addition, U.S. providers often required a consumer’s phone number or address before disclosing available plans. As a result, the research represents a best effort attempt to collect accurate pricing and plan data.
Service Level Criteria and Metrics
In 2010 the Federal Communications Commission (FCC) updated its definition of broadband Internet service to a minimum speed of 4 megabits per second (Mbps) download and 1 Mbps upload. We used this metric as a threshold for our research and did not record any high-speed services where advertised download speeds were slower than 4 Mbps. For each plan, we collected data on the following metrics:
- Download and upload speeds
- Monthly costs
- Data caps & penalties (overage fees or slowed speeds)
- Activation and installation fees
- Modem and equipment rentals
- Contract lengths
All foreign currencies have been adjusted to U.S. dollars via the World Bank’s purchasing power parity (PPP) metrics.[vii] Unlike direct exchange rates, which are often volatile and do not account for global income disparities, PPP conversion rates adjust for differences in costs of living, price levels, and other factors that affect a consumer’s purchasing power. This allows us to make more effective comparisons between the cities that we surveyed.[viii]
In the report that follows, we have included some of the most relevant findings from our research as well as a discussion of the policy implications for the U.S. The sections in the report include:
- A comparison of "triple play" offerings that bundle Internet, phone, and television services;
- A survey of the best available Internet plan for approximately $35 USD in each city;
- A comparison of fastest speed available in each city.
All speeds in this report are based upon advertised speeds and listed in Mbps. Note that when certain information was not available, we have indicated this by marking “--“ in the table. Research was conducted in May and June of 2012. For a comprehensive set of pricing information collected, please consult Appendix B.
Section 1: Triple Play Rankings
The term “triple play”generally refers to a bundle of services that includes high-speed Internet, telephone and television in a single monthly rate. Although it varies by country, consumers usually get discounts by subscribing to bundled services rather than paying individually for all three. Comparing triple play is a useful metric for most consumers as a substantial number of individuals purchase their high-speed internet in conjunction with television and phone packages. Triple play offerings are quite popular not only in the U.S., but also increasingly in other countries.[ix]
In this section, we rank the pricing of all triple play packages across the cities we surveyed. Although there are a variety of premium channels and Voice over IP add-on services available, we have highlighted the conventional triple play packages for the purpose of direct comparisons. The plans listed in Table 1 will get a consumer a standard cable subscription with high-definition (HD) programming, basic phone service and the lowest tier Internet speed available. The table ranks triple play offerings available with the above criteria in each city. Prices listed are for the first 12 months of a contract and do not include introductory offers.
Table 1. Triple Play Rankings (Internet, Phone, and TV Bundles)
Rank
City
ISP
Download Speed
Upload Speed
Price (USD/PPP)
Network Technology
Data Cap (GB)
1
Riga
Balti-Com
20
5
25.43
Cable
None
2
Seoul
KT
50
50
32.74
Fiber
None
3
Paris
Free
100
50
34.47
Fiber
None
4
Seoul
LGplus
100
100
35.78†
Fiber
None
5
Paris
SFR
25
--
36.67†
ADSL
None
6
Paris
Bouygues Telecom
20
0.512
36.67
ADSL
None
7
Paris
Orange
20
--
37.53
ADSL
None
8
Copenhagen
Telia
20
1
40.36†
DSL
None
9
London
Virgin
30
--
43.03†
Fiber
None
10
Paris
Darty
100
5
43.56†
Fiber
None
11
Amsterdam
T-Mobile
20
1
44.64†
DSL
--
12
Berlin
TeleColumbus
16
1
44.93
Cable
None
13
London
BT
16
--
45.45
DSL
None
14
Riga
Lattelecom
20
20
45.66
Fiber
None
15
Hong Kong
City Telecom Limited
500
20
46.72
Fiber
None
…
…
…
…
…
…
…
…
30
Lafayette
AT&T
6
1
65.60
DSL
250
32
Chattanooga
AT&T
6
1
68.44
DSL
250
33
San Francisco
AT&T
6
1
73.94
DSL
250
34
Los Angeles
Verizon
7
--
79.99†
DSL
None
38
San Francisco
Comcast
20
4
89.97
Cable
300
39(t)
New York
RCN
25
2
89.99†
Cable
None
39(t)
Washington DC
RCN
25
2
89.99†
Cable
None
41
Washington DC
Verizon
15
5
89.99†
DSL
None
43
Los Angeles
TimeWarner
10
1
89.99
Cable
None
46
Bristol
BVU
16
3
96.40
Fiber
None
47
Washington DC
Comcast
20
--
99.00
Cable
300
48
Bristol
Charter
15
3
99.97†
Cable
100
49
Chattanooga
Comcast
20
4
99.99
Cable
300
50
Lafayette
LUS
15
15
103.89
Fiber
1024
51
Bristol
CenturyLink
10
0.7
107.26†
DSL
250
52
New York
Time Warner
15
0.78
112.69
Cable
None
53
San Francisco
Astound
10
1
124.00
Fiber
100
55
Chattanooga
EPB
30
30
129.51
Fiber
None**
56
Lafayette
Cox
15
1.5
133.96
Cable
200
57
New York
Verizon
10
2
154.98†
Fiber
None
This table ranks the top 15 triple play offers we found, and then indicates where the U.S. carriers that offer triple play services fall in the list.
‘--‘ indicates that data could not be found.
CenturyLink, Verizon, and AT&T partner with DirecTV to provide television services for these plans.
*Reserves the right to enforce a bandwidth cap but does not define a specific limit.
†Prices include some or all monthly equipment rental charges.
Section 2: What Can You Get For $35?
Affordability is often a barrier to subscribing to high-speed Internet service. According to a 2010 study conducted by the NTIA, individuals who currently do not have high speed Internet service at home are more likely to indicate price, rather than availability of services in their area, as the reason for not subscribing.[x] In this section we compare the best “bang for your buck” possible in our selected cities, ranking the top Internet service plans that we found priced around the equivalent of $35 in each city.
Table 2 ranks wireline or landline offerings such as cable broadband or fiber-optic broadband in terms of the highest speed available around the $35 price point. We display standalone Internet packages unless providers only offer Internet bundled with other services. Prices shown represent the best effort to display the regular prices for the first 12 months of a contract and do not include temporary introductory deals on the first three or six months of service, as those rates often increase substantially after expiration and do not reflect the long term cost for consumers. We do not include wireless or mobile broadband offerings in this table because the advertised speeds tend to be far lower than the hardline packages.
Table 2. What's the Best High-Speed Connection You Can Get for $35?
Rank
City
ISP
Download Speed
Upload Speed
Price (USD/PPP)
Network Technology
Data Cap (GB)
1
Hong Kong
3 in HK
500
500
37.34
Fiber
None
2(t)
Tokyo
Yahoo BB
200
200
26.85
Fiber
None
2(t)
San Francisco
Webpass
200
200
37.50
Fiber
--
4(t)
Riga
Balti-Com
100
100
31.14
Cable
None
4(t)
Seoul
LGplus
100
100
31.90
Fiber
None
6
Paris
Free
100
50
34.47
Fiber
None
7
Bucharest
Romtelcom
100
32
28.52
Fiber
None
8
Berlin
Kabel Deutschland
100
6
24.57
Cable
None
9
London
Virgin
60
--
28.03
Fiber
None
10
Amsterdam
Tele2
50
5
33.93
DSL
None
11
Copenhagen
ComX
40
40
37.94
Fiber
None
12
Prague
UPC
30
1
35.12
Fiber
None
13
Toronto
Acanac
28
1
32.78
DSL
None
14
Zürich
UPC
25
2.5
32.89
Cable
--
15
New York
RCN
25
2
34.99
Cable
None
16
Washington DC
RCN
25
2
39.99
Cable
None
17
Dublin
Smart Telecom
24
1
34.83
Fiber
350
18
Lafayette
LUS
15
15
34.95
Fiber
1024
19
Bristol
Charter
15
3
34.99
Cable
100
20
Los Angeles
Time Warner
10
1
29.99
Cable
None
21
Chattanooga
AT&T
6
1
24.95
DSL
250
22
Mexico City
Cablevision
4
--
37.75
Cable
--
‘--‘ indicates that data could not be found.
*Offers included additional bundled services.
The San Francisco story: Local ISPs
Broadband connectivity in San Francisco lags behind global cities like Seoul and Hong Kong, but access in the Bay Area remains faster, more affordable, and far more flexible in comparison to other large U.S. cities like New York, Washington DC, and Los Angeles.
Why is this the case? One key factor is the presence of local ISPs. Compared to the other U.S. cities surveyed in this report, the Bay Area broadband market is the most diverse, boasting several popular local ISPs that deliver high-speed wireless and fiber access to residences and businesses. In addition to offering faster speeds than Comcast and Verizon, these ISPs tend to have lower prices and do not bind their customers to lengthy contracts. These ISPs have injected a competitive, innovative element to San Francisco’s broadband market.
Sonic.net, for example, has grown into Northern California’s largest independent ISP since it was founded in 1994. The company strings fiber over power lines, providing high speed access to over 40,000 customers. At the moment, Sonic offerscusomers in San Francisco20 mbpsdownload speeds as well as unlimited home phone for a flat month-to-month fee of $39.99. Sonic.net even offers a 1000 Mbps connection for $69.95 a month in parts of Marin County.Other local ISPs include Monkeybrains and Webpass, which provide “last mile” connectivity between fiber lines and residences via wireless devices.
Section 3: Speed Leaders
The speed of your Internet connection determines your ability to view web pages, download and upload content, and use applications and services like Voice over IP and two-way videoconferencing. As network technology has improved over the years, the general trend has been an increase in broadband connections speeds.[xi] At the same time, however, the amount of bandwidth required by the average consumer increases as streaming video and sharing user-generated content becomes commonplace. Speed, therefore, remains an important metric for consumers evaluating their broadband options and considering how they plan to use the Internet.
In this section, we compare “speed leaders” and rank offerings in each city based on the fastest advertised Internet speeds found in our research. We chose basic Internet packages, without bundled services, unless the Internet service provider only offered bundled services. The offerings are ranked by download speeds,with higher upload speeds used to differentiate between plans that are tied. Advertised download speeds continue to be a major selling point for consumer Internet service and often are substantially higher than upload speeds. However, upload speeds are rising in importance as users create and share more content and data. Higher upload speeds allow for more effective use of online video conferencing and the transfer of files to cloud-based storage options. While more ISPs are offering symmetrical service, where the download and upload speeds are equal, most providers continue the standard practice of engineering their networks to maximize download speeds.
Table 3. Wired Speed Leaders
Rank
City
ISP
Download Speed
Upload Speed
Price (USD/PPP)
Network Technology
Data Cap (GB)
1(t)
Hong Kong
3 in HK
1000
1000
48.59
Fiber
None
1(t)
Chattanooga, TN
EPB
1000
1000
317.03
Fiber
None
3
Amsterdam
KPN
500
500
83.33
Fiber
--
4
Copenhagen
ComX
250
250
126.78
Fiber
None
5
Bristol, VA
BVU
250
30
259.95
Fiber
None
6(t)
Tokyo
Yahoo BB
200
200
26.85
Fiber
None
6(t)
San Francisco
Webpass
200
200
37.50
Fiber
--
6(t)
Riga
Lattelecom
200
200
62.80
Fiber
None
9
Toronto
Bell
175
175
106.52
Fiber
300
10
New York
Verizon
150
35
159.98
Fiber
None
11
Dublin
eircom
150
30
81.40
Fiber
500
12
Berlin
TeleColumbus
128
4
67.89
Cable
None
13
Prague
UPC
120
10
56.19
Fiber
None
14
Bucharest
UPC
120
6
37.58
Fiber
--
15
Washington DC
Comcast
105
10
105.00
Cable
300
16(t)
Seoul
LGplus
100
100
23.04
Fiber
None
16(t)
Lafayette
LUS
100
100
199.95
Fiber
8192
18
Paris
Free
100
50
34.47
Fiber
None
19
Zürich
Green
100
20
150.00
Fiber
--
20
London
BT
100
15
53.03
Fiber
None
21
Los Angeles
TimeWarner
50
5
99.99
Cable
None
22
Mexico City
Iusacell
40
--
211.99
Fiber
--
‘--‘ indicates that data could not be found.
Offers included additional bundled services.
*Reserves the right to enforce a bandwidth cap by charging a fee, suspending, or terminate service but does not define a specific limit.
Worldwide Speed Leaders
{"dataSourceUrl":"//docs.google.com/spreadsheet/tq?key=0AjXfjO3kbDzMdHppOXdQM05scS1fSUtINmdvb1FRUEE&transpose=0&headers=1&range=E1%3AH23&gid=21&pub=1","options":{"titleTextStyle":{"bold":true,"color":"#666666","fontSize":12},"series":[{"color":"#9fc5e8","type":"bars"},{"color":"#ea9999","type":"bars"},{"color":"#6aa84f","targetAxisIndex":1}],"curveType":"function","legendTextStyle":{"color":"#666666","fontSize":12},"animation":{"duration":0},"theme":"maximized","type":"line","hAxis":{"titleTextStyle":{"color":"#666666","italic":true,"fontSize":12},"useFormatFromData":true,"slantedTextAngle":90,"slantedText":true,"viewWindowMode":"pretty","textStyle":{"color":"#666666","fontSize":12},"viewWindow":{}},"vAxes":[{"useFormatFromData":true,"viewWindowMode":"pretty","textStyle":{"color":"#c27ba0","fontSize":"11"},"viewWindow":{}},{"useFormatFromData":true,"minValue":0,"viewWindowMode":"explicit","textStyle":{"color":"#38761d","fontSize":12},"viewWindow":{"min":0}}],"booleanRole":"certainty","domainAxis":{"direction":1},"legend":"in","focusTarget":"category","isStacked":false,"width":900,"height":450},"state":{},"chartType":"ComboChart","chartName":"Chart 2"}
We also include a separate category for mobile or wireless broadband services using a USB dongle or other means, but do not include smartphone data plans. For these wireless offerings we have listed the package with the lowest monthly data cap, although consumers often have the option to upgrade to a higher monthly data tier for additional fees.
Table 4. Wireless Speed Leaders
Rank
City
ISP
Download Speed
Upload Speed
Price (USD/PPP)
Network Technology
Data Cap (GB)
1(t)
Toronto
Rogers
100
50
17.88
Wireless (LTE)
0.1
1(t)
Hong Kong
3 in HK
100
50
18.39
Wireless (LTE)
0.4
1(t)
Berlin
Deutsche Telekom
100
50
98.70
Wireless (LTE)
30
4
Mexico City
Axtel-Acceso Universal
100
--
256.06
Wireless (WiMax)
None
5
Tokyo
Yahoo BB Softbank
76
10
53.70
Wireless (HDSPA)
5
6
Seoul
LGplus
75
-
36.38
Wireless (LTE)
5
7
Bucharest
Orange
43.2
5.76
40.59
Wireless
9
8(t)
Paris
Bouygues Telecom
42
5.8
11.38
Wireless
0.3
8(t)
Zürich
Swisscom
42
5.8
55.92
Wireless
10
10(t)
Chattanooga
T-Mobile
42
1
39.99
Wireless (HSPA+)
2
10(t)
Lafayette
T-Mobile
42
1
39.99
Wireless (HSPA+)
2
10(t)
Los Angeles
T-Mobile
42
1
39.99
Wireless (HSPA+)
2
10(t)
New York
T-Mobile
42
1
39.99
Wireless (HSPA+)
2
10(t)
San Francisco
T-Mobile
42
1
39.99
Wireless (HSPA+)
2
10(t)
Washington, DC
T-Mobile
42
1
39.99
Wireless (HSPA+)
2
10(t)
Prague
T-Mobile
42
--
63.22
Wireless
30
17
Amsterdam
Tele2
40
3
35.71
Wireless (HDSPA)
3
18
Dublin
Digiweb
30
1
47.91
Fixed wireless
100
19
London
Three.co.uk
21
5.76
11.92
Wireless
1
20
Copenhagen
Fullrate
15
1
34.14
Wireless
10
21
Bristol
Verizon
12
5
50.00
Wireless (LTE)
5
22
Riga
Tele2
6
--
51.14
Wireless
25
‘--‘ indicates that data could not be found.
It is important to note that the speeds listed represent the maximum speed advertised and not the actual speed of a service, which can vary widely depending on location and time of use. For example, a recent survey by PC Magazine found that actual 4G mobile broadband speeds ranged between 3 to 11 Mbps download and less than 1 Mbps to 7 Mbps for upload, which is considerably less than advertised.[xii]
Mirror, mirror, on the wall, who’s the fastest of all?
Comcast’s national advertising campaign states that its broadband speeds are “the fastest in the nation.” The claim stems from a 2011 PC Magazine study that ranked Comcast in the top speed tier of Internet service providers. However, the study only examined providers with a national presence, ignoring smaller providers.
Take Chattanooga, Tennessee, as an example. When you compare the download speeds of the highest tier residential broadband services offered by Comcast and Chattanooga EPB, the local fiber optic municipal broadband provider, there is a big difference. Comcast offers 105 Mbps, while EPB customers can get up to 1000 Mbps service in their homes. Chattanooga EPB’s maximum download speed is nearly 10 times as fast as the one offered by Comcast, but Comcast’s advertising in Chattanooga still calls their service the “fastest in the nation.”
In June 2012, the National Advertising Division of the Better Business Bureau publicly recommended that Comcast discontinue these claims or clearly qualify that they only apply in certain areas.
Policy Implications
When looking at the cost and speed of Internet access in major U.S. cities in comparison to other global cities, it is difficult to ignore the fact that the U.S. is so much more expensive than many of its international counterparts. As this report demonstrates, U.S. consumers often pay higher prices for slower service than many other parts of the world. For example, customers in Seoul, South Korea, can get a 100 Mbps connection for as low as the equivalent of $23 USD, while in Hong Kong, it's possible to subscribe to a 500 Mbps connection from 3 in HK for about $38 USD. By contrast, it costs between $100 and $300 per month to get a connection speed of over 100 Mbps in most U.S.cities—if such high speeds are even available at all.
Although various factors contribute to the cost and speed of Internet access in a community, a critical factor is competition. In markets where connections tend to be more affordable and faster, customers have access to at least three competitive ISPs offering similar plans. In the majority of U.S. cities, most consumers have a choice between a local telephone company and local cable company. According to the 2010 National Broadband Plan, 78 percent of households in the U.S. have a choice between two providers, while an additional 13 percent have just one option.[xiii]
There are some exceptions, however. For example, in San Francisco, several local ISPs have emerged to compete with the incumbent cable and telephone companies. Smaller providers such as Sonic.net and Astound are offering more affordable high-speed Internet plans compared to AT&T or Comcast. Moreover, in Lafayette, LA, in response to services of the municipal broadband provider, LUS Fiber, the incumbent cable operator, Cox changed its pricing strategy, ceasing to raise prices for three years straight when annual increases had been common before. Cox also introduced a new 50 Mbps speed tier in Lafayette in response to the higher speeds offered by LUS.[xiv]
The low level of competition that is common in much of the United Statesis not the case in many other developed nations. Unbundling policies, also known as open access, played a critical role in facilitating competitionin the first transition to broadband in Europe and parts of Asia and are now central to efforts for the next generation networks.[xv] Open access policies require that incumbent broadband providers offer to lease capacity on their networks to new entrants selling competing Internet services to consumers. They have been critical in creating a robust competitive marketplace in Japan and France, where new upstarts such as Yahoo BB and Free were able to compete with the telephone incumbents and now offer 100 mbps+ Internet connections at competitive prices. Moreover, contrary to criticisms of open access policies, many of the companies that got their start leasing access from incumbent providers have since invested substantially in building their own network infrastructure. (See sidebar “Unbundling Policies in Paris”)
Almost a decade ago, the FCC abandoned similar unbundling policies in the U.S. The decision reversed regulations implemented after the passage of the 1996 Telecommunications Act, which had required local telephone companies to open up access to their copper wiring to competitors. The original unbundling policy helped usher in new competition for telephone services, as well as enabling Internet service upstarts AOL and Compuserve to bring the first dial-up service to consumers. As a result, there were over 9,000 ISPs in the U.S. in the year 2000. By 2005, when a Supreme Court ruling upheld the FCC’s decision to deregulate broadband service, the number had fallen by 74 percent to just under 2,500.[xvi]
Unbundling Policies in Paris
Paris has been known as a leader in many areas, but broadband competitiveness would rarely be the first to come to mind. Yet in the past decade, Paris has emerged as a model in providing fast speeds at competitive prices—a shift which can be traced back to government policies instituted a dozen years ago.
In 2000, France instituted an unbundling policy, and ARCEP, the national regulator for postal and electronic communications, forced the former state-owned monopoly, France Telecom (which sells mobile and Internet services as Orange), to open its network to rival operators. Almost immediately, independent ISPs and carriers from other countries began to lease access to France Telecom’s infrastructure and compete to offer broadband services. New companies like Iliad (offering services since 2002 as Free) emerged as fierce competitors, benefitting from the fact that they did not have to pay the initial costs of laying wires. And now, those companies are beginning to build their own infrastructure in order to meet a growing demand for faster speeds at low prices.
Free, for example, has always charged the same monthly rate, but continued to add new features to its service. Today, Free offers competitive DSL prices in Paris, and has begun to lay its own fiber in the city to provide even faster speeds. Eligible customers can get a connection with download speeds of up to 100 mbps (50 mbps upload) for the monthly fee of €29.99 ($34.47 adjusted for PPP), the same cost as Free’s standard DSL package. To get comparable speeds in most major US cities would cost over $100 per month.
Unrealized Competition
To justify deregulation back in 2005, the FCC predicted that in the future, cable and telephone companies in each community would effectively compete with each other to provide high-speed Internet service. However, this was a flawed approach from the beginning. Cable was able to offer a more attractive bundle of video programming and broadband plans than DSL-based providers. Moreover, cable network upgrade costs, largely limited to replacing equipment, were extremely modest compared to the need of telephone providers to build-out completely new infrastructure. According to figures from J:Com, the largest cable company in Japan, costs for upgrading to the Docsis 3.0 (capable of providing peak download speeds in excess of 100 Mbps) was just $20 per home passed. By comparison, Verizon spent an average of $817 per home to deploy its fiber-based FiOS service.[xvii]
The FCC also predicted that wireless broadband would become a third pipe into the home, competing directly with high-speed cable and DSL service. Although wireless or mobile broadband services are now widely available, the speeds and costs (including data overage charges) in the U.S.are not competitive in comparison to wired offerings like cable or DSL. Competition in the mobile broadband market is also decreasing, as two of the nation's largest mobile and home telephone companies, Verizon and AT&T, have bought up rivals and approximately 75 percent of the wireless spectrum auctioned by the federal government.[xviii] The remaining mobile competitors have struggled to keep up, and new entrants face considerable hurdles to get into the market.
New technologies such as broadband over power line (BPL), and satellite were also expected to add to a vibrant marketplace of choices for consumers. However, for a variety of reasons, broadband over powerline never really took off. In 2010, the largest and most successful BPL demonstration in the United States shut down.[xix] Satellite broadband is largely a last resort for rural residents, and it is hardly competitive with cable or even DSL given its slower speeds, high latency, and low data usage caps.
Prices Will Continue to Increase for U.S. Consumers
In December 2011, Verizon Wireless, Comcast, Time Warner Cable, Bright House Networks, and Cox Communications announced the sale of a large chunk of mobile broadband spectrum held by the cable companies. For Verizon, the deal will net them additional spectrum for their already sizeable stockpile, while the cable companies receive several billion dollars in cash. As part of the deal, the companies agreed on a series of joint re-sale, marketing, and technology development agreements. Beginning in January 2012, Comcast and Verizon Wireless debuted joint promotions in Portland and Seattle, offering customers who sign a two-year contract for both wireless and cable service a prepaid Visa card of up to $300. In addition, Comcast will promote Verizon Wireless service through its call centers and retail stores.[xx]
The arrangements are certainly beneficial for companies and shareholders, but not so much for consumers. As Senator Herb Kohl (D-WI) offered in a statement at the Senate hearing on the deal, “Many now wonder if these agreements that we are examining... amount to a truce between one of the two largest phone companies and over 70 percent of the cable TV industry.” Though the deal has yet to be approved by the FCC and the Department of Justice is investigating, if it goes through, the end result is likely to mean even higher prices for U.S. consumers over the long term. Cable already has a sizeable lead in high-speed residential broadband, as DSL offered by telephone companies has steadily lost market share.[xxi] A 2012 FCC report indicates that over 82 percent of residential broadband subscriptions above 6 Mbps download and 1.5 Mbps upload are delivered via cable modems.[xxii]
In the future, consumers wishing to subscribe to higher speed Internet services will likely face a near-monopoly from cable providers, as telephone providers have halted wide scale upgrades of their networks. AT&T announced after reaching a little over half of its customer base with its more competitive fiber-to-the-neighborhood technology, U-Verse, that it would stop building at the end of 2011.[xxiii] Verizon reached only around 14 percent of the nation with its fiber-to-the home FiOS service, before halting further deployments in 2010.[xxiv] By partnering with cable companies, many have argued, Verizon has essentially “reliev[ed] itself of the need to expand FiOS.”[xxv]
Instead, upgrades will likely continue to take the shape of higher speed offerings in areas where infrastructure already exists, as Verizon announced plans for a new, higher tier of broadband services in June 2012.[xxvi] Meanwhile, unless consumers live in areas where Verizon and AT&T have already deployed FiOS or U-verse, they will have exactly one choice for high-speed Internet service.
The erosion in competition is also likely to reduce incentives for cable providers to upgrade their infrastructure to offer higher speeds, despite the fact that the data show how far the U.S. lags behind other countries. In 2006, even when DSL was more competitive, cable executives from the largest U.S. providers killed an upgrade to cable broadband technology that would have increased cable download speeds to at least 70 Mbps.[xxvii] As a cable executive revealed in a round-table discussion a year later, “there [wasn’t] any market pressure” to upgrade sooner.[xxviii]
U.S. Needs to Re-Examine its Current Policies
For much of the past decade, the majority of Americans have had only two choices for high-speed broadband service. And with the Verizon/Cable deal, that choice may dwindle to one for the home, while Verizon and AT&T maintain a duopoly on mobile service. This alarming trend suggests that the U.S. needs a new approach to create robust broadband competition if it is to keep pace with its international peers. The FCC failed to adequately address these issues in the 2010 National Broadband Plan, even as it concluded that three quarters of the U.S. would “likely have only one broadband service provider that can offer very high peak download speeds.” It also completely ignored the findings of its own study completed by the Berkman Center, which found that most of the world’s broadband leaders utilized some form of open access or unbundling regulations to promote competition.
Instead, the Administration and the FCC seem to be singularly focused on freeing up more spectrum for mobile broadband. Theoretically, this could enable new entrants and allow existing wireless competitors to increase network capacity and offer faster speeds. However, the amount of spectrum that will actually be freed up for mobile broadband—and that will not just end up in the hands of the two largest providers—remains unclear. Moreover, without other policies such as policing anti-competitive pricing for special access lines (links that connect cell phone towers to the Internet and telephone networks), or empowering consumers to more easily switch providers by requiring unlocked phones, it is far from certain that the U.S. will develop a vibrant, competitive marketplace for mobile broadband.
Rather than focusing on spectrum auctions and the promise of wireless broadband, policymakers need to address the lack of competition in most of the U.S. and how policies can enable new competitors to enter the marketplace. Moreover, policymakers should encourage competition from all sectors, including the public sector. In the past few years, a number of state legislatures have created significant barriers for local governments to build their own networks, even as community-owned networks in cities such as Chattanooga, TN, are offering world-leading broadband speeds. It should be no surprise that telephone and cable companies have often thrown their weight behind these state laws, claiming that it is unfair for government to compete with the private sector and promising to upgrade local network infrastructure, even as the they tell their shareholders the exact opposite.[xxix] The legal battle over these networks raises the question of who will invest in infrastructure and offer broadband services to compete with the cable and telecom monopolies if local communities are prevented from doing so in the future. As individuals and businesses become increasingly reliant on high-speed Internet access for work, healthcare, education, and other services, decreased investment, slower speeds, and higher prices will be a burden on the entire nation.
References
[i] Statistics for advertised speed, penetration, usage, pricing, and coverage available at “OECD Broadband Portal,” Organization for Economic Co-operation and Development, http://www.oecd.org/document/54/0,3746,en_2649_34225_38690102_1_1_1_1,00.html [accessed July 17, 2012].
[ii] Akamai Technologies, The State of the Internet, 4th Quarter, 2011 Report, ed. David Belson, http://www.akamai.com/stateoftheinternet/.
[iii] Josh Levy, "Hey America! We're Ranked #16 in Broadband!" The Huffington Post, April 30, 2012, http://www.huffingtonpost.com/josh-levy/broadband-rankings-worldwide_b_1400630.html [accessed July 17, 2012]. Also see “In broadband race, USA is not No. 1,” Editorial, USA Today, December 11, 2011, http://www.usatoday.com/news/opinion/editorials/story/2011-12-06/Guess-whos-No-1-in-broadband-Hint-Its-not-USA/51683876/1 [accessed July 17, 2012].
[iv] Robert McDowell, “Broadband Baloney,” Wall Street Journal, July 24, 2007, http://online.wsj.com/article/SB118524094434875755.html [accessed July 17, 2012].
[v] For more information on the municipal networks in Bristol, Chattanooga, and Lafayette, see Christopher Mitchell, Broadband at the Speed of Light: How Three Communities Built Next Generation Networks, Institute for Local Self Reliance and the Benton Foundation, April 2012, http://www.ilsr.org/broadband-speed-light/.
[vi] Criteria for the OECD Broadband Price Collections, available at“OECD Broadband Portal.”
[vii] For the PPP conversion tables used in this report, see 2010 data at “PPP conversion factor, GDP (LCU per international $),” World Bank, http://data.worldbank.org/indicator/PA.NUS.PPP [accessed July 17, 2012].
[viii] PPP definition taken from the World Bank Data & Statistics, available at “Quick reference table,” World Bank, http://econ.worldbank.org/WBSITE/EXTERNAL/DATASTATISTICS/0,,contentMDK:20399244~menuPK:1390200~pagePK:64133150~piPK:64133175~theSitePK:239419,00.html [accessed July 17, 2012]. Also see Fredric A. Vogel, “What Is A Purchasing Power Parity?” The World Bank, http://siteresources.worldbank.org/ICPINT/Resources/270056-1255977254560/6483625-13388 [accessed July 17, 2012].
[ix] Julian Clocer, “Triple play subscriptions to quadruple,” Broadband TV News, July 18, 2011, http://www.broadbandtvnews.com/2011/07/18/triple-play-subscriptions-to-quadruple [accessed July 17, 2012].
[x] National Telecommunications and Information Administration, Exploring the Digital Nation: Home Broadband Internet Adoption in the United States, November 2010, http://www.ntia.doc.gov/files/ntia/publications/esa_ntia_us_broadband_adoption_report_11082010_1.pdf [accessed July 17, 2012].
[xi] For historical trends on the increase of connection speeds, see The State of the Internet, 48.
[xii] Mark Sullivan, “3G and 4G Wireless Speed Showdown: Which Networks Are Fastest?” PCWorld, April 17, 2012, http://www.pcworld.com/article/253808/3g_and_4g_wireless_speed_showdown_which_networks_are_fastest.html [accessed July 17, 2012].
[xiii] Federal Communications Commission, Connecting America: The National Broadband Plan, March 2010, http://www.broadband.gov/plan/ [accessed July 17, 2012].
[xiv] Broadband at the Speed of Light: How Three Communities Built Next Generation Networks, 28-9. In Lafayette, LA, Cox had raised prices six times in four years, but when LUS Fiber was about to come online, Cox kept its prices stable for three straight years. Additionally, Cox chose Lafayette as the first market to deploy a 50 Mbps offering because it is a "competitive market."
[xv] For more information on open access policies worldwide and their impact on competition, see Section 4: Competition and Access from The Berkman Center for Internet & Society at Harvard University, Next Generation Connectivity: A Review of Broadband Internet Transitions and Policy Worldwide, February 2010, http://cyber.law.harvard.edu/pubrelease/broadband/ [accessed July 17, 2012].
[xvi] Teletruth, “Killing Off 7000 Independent Internet Service (ISPs) Providers by the FCC Created Net Neutrality Problems,” November 29, 2007, http://www.newnetworks.com/parttwosummary.htm [accessed July 17, 2012].
[xvii] Saul Hansell, "World's Fastest Broadband at $20 per home," New York Times: Bits Blog, April 3, 2009, http://bits.blogs.nytimes.com/2009/04/03/the-cost-to-offer-the-worlds-fastest-broadband-20-per-home/ [accessed July 17, 2012].
[xviii] Stacey Higginbotham,"Verizon's spectrum deal with cable is the end of broadband competition," GigaOm, December 2, 2011, http://gigaom.com/broadband/verizons-spectrum-deal-with-cable-is-the-end-of-broadband-competition/ [accessed July 17, 2012].
[xix] Karl Bode, "Most Successful US Broadband Over Powerline Network Shut Down," Broadband DSL Reports, April 8, 2010, http://www.dslreports.com/shownews/Most-Successful-US-Broadband-Over-Powerline-Network-Shut-Down-107812[accessed July 17, 2012].
[xx] Jesse Ward, "Industry Demands Scrutiny of Verizon-Cable Agreements," NCTA Blog New Edge, January 23, 2012, http://www.ntca.org/new-edge/wireless/industry-demands-scrutiny-of-verizon-cable-agreements [accessed July 17, 2012].
[xxi] Susan P. Crawford, "The Looming Cable Monopoly," Yale Law & Policy Review, Vol. 29, December 2010, http://yalelawandpolicy.org/29/the-looming-cable-monopoly [accessed July 17, 2012].
[xxii] Federal Communications Commission, Internet Access Services: Status as of June 30, 2011, June 2012, http://transition.fcc.gov/Daily_Releases/Daily_Business/2012/db0614/DOC-314630A1.pdf [accessed July 17, 2012].
[xxiii] Dave Burstein, "AT&T's Stankey: U-Verse Build Virtually Over," Broadband DSL Reports, May 18, 2011, http://www.dslreports.com/shownews/ATTs-Stankey-Uverse-Build-Virtually-Over-114279 [accessed July 17, 2012].
[xxiv] Karl Bode, "The Press Realizes the FiOS Party is Over," Broadband DSL Reports, March 30, 2010, http://www.dslreports.com/shownews/The-Press-Realizes-The-FiOS-Party-Is-Over-107639 [accessed July 17, 2012].
[xxv] "Verizon's Worrisome Cable Deals," Editorial, New York Times, December 24, 2011, http://www.nytimes.com/2011/12/25/opinion/sunday/verizons-worrisome-cable-deals.html [accessed July 17, 2012].
[xxvi] Jon Brodkin, "Faster Than Your Router Speed? Verizon Doubles FiOS Speeds to 300Mbps," Ars Technica, May 30,2012, http://arstechnica.com/information-technology/2012/05/faster-than-your-router-verizon-doubles-fios-speeds-to-300mbps/ [accessed July 17, 2012].
[xxvii] Alan Breznick, "CableLabs Kills Interim DOCSIS Standard," Light Reading, June 28, 2006, http://www.lightreading.com/document.asp?doc_id=98169 [accessed July 17, 2012].
[xxviii] CTO Roundtable Discussion, compiled by Leslie Ellis, "How Sexy is HFC? (Answer: Plenty)," CED Magazine, May 1, 2007,http://www.cedmagazine.com/articles/2007/05/how-sexy-is-hfc%3F-(answer%3A-plenty.) [accessed July 17, 2012].
[xxix] Denise Roth Barber, "Dialing Up the Dollars: Telecommunications Interests Donated Heavily to NC Lawmakers," National Institute on Money in State Politics, March 20, 2012, http://www.followthemoney.org/press/ReportView.phtml?r=484&utm_campaign=nc-broadband-report-e-alert&utm_medium=email&utm_source=nimsp-contacts [accessed July 17, 2012].
Sidebar References
“The San Francisco Story: Local ISPs”
“Pando Networks Releases 2011 American Broadband Study,” Pando Networks, http://www.pandonetworks.com/Pando-Networks-Releases-2011-American-Broadband-Study [accessed July 17, 2012].
Cathy Bussewitz, “Sonic.net builds super-fast network for future,” Santa Rosa Press Democrat, May 6, 2012, http://www.pressdemocrat.com/article/20120506/BUSINESS/120509761/1036/business?Title=Sonic-net-builds-super-fast-network-for-future [accessed July 17, 2012].
Thomas Claburn, “Monkey Brains: Silly Name, Serious Broadband,” InformationWeek, August 3, 2010, http://www.informationweek.com/news/infrastructure/management/226500209?pgno=1[accessed July 17, 2012].
Rob Pegoraro, “Gigabit Internet for $70: the unlikely success of California's Sonic.net,” Ars Technica, February 26, 2012, http://arstechnica.com/tech-policy/2012/02/gigabit-internet-for-80-the-unlikely-success-of-californias-sonicnet/ [accessed July 17, 2012].
“Mirror, Mirror, On The Wall: Who’s the Fastest of All?”
Carey O'Neill, "Comcast Asked To Change Ad Claims," Chattanooga Times Free Press, June 14, 2012, http://timesfreepress.com/news/2012/jun/14/comcast-asked-to-change-ad-claims/ [accessed July 17, 2012].
“Unbundling Policies in Paris”
Scott J. Wallsten and Stephanie Hausladen, “Net Neutrality, Unbundling, and their Effects on International Investment in Next-Generation Networks,” Review of Network Economics, Vol. 8, Issue 1, 2009, http://econpapers.repec.org/article/bpjrneart/v_3a8_3ay_3a2009_3ai_3a1_3an_3a6.htm [accessed July 17, 2012].
Jennifer Schenker, “ViveLaHighSpeedInternet!” Bloomberg Businessweek, July18, 2007, http://www.businessweek.com/stories/2007-07-18/vive-la-high-speed-internet-businessweek-business-news-stock-market-and-financial-advice [accessed July 17, 2012].
Full Data Set (click here to download)