The Economic Benefits of Ubiquitous Broadband with Public Networks

A wave of opportunity for wireless broadband applications is in the making. Most of it lies in the form of broadband deployment. In their April 2001 white paper "The $500 Billion Opportunity: The Potential Economic Benefit of Widespread Diffusion of Broadband Internet Access," Robert Crandall and Charles Jackson point to an economic benefit of $500 billion per year for the American economy if broadband Internet access were to become as ubiquitous as landline phones.

The remainder of this chapter assumes that it is considerably cheaper (both in terms of hardware and lawyers) to deploy wireless broadband Internet to a residence than a similar service that depends on wiring (copper wire from the phone company or coax cable from the cable TV company). Both telephone wires and cable TV coax cable are accessible by almost 90 percent of American households. Even if they were not, the cost of copper wire, for example, is 15 cents per foot. The physical cost of connecting a home to the Internet in most residential applications is not that high. However, for a new market entrant, gaining the right-of-way from private landowners and public utilities to get to those households in most cases will not be possible without costly legal procedures. The legal costs of running wire or cable to a residence may not be offset by the revenue generated from subscription fees from that residence.

Using 802.11 as a means of access does not require legal dealings for rights-of-way and, compared to wired infrastructure, can be deployed much more quickly. Based on the fact that only 8 percent of U.S. households have broadband access via either telephone wires or cable TV coax cable, it will be assumed that these wired means of access are, for a variety of reasons, inadequate for achieving the same level of penetration in the market as telephone service. As evidenced by the efforts of CLECs to offer competitive residential telephone service using incumbent telephone poles and other incumbent-owned and -operated facilities, it is far easier to bypass PSTN facilities than to utilize them via legal means. 802.11 only requires an AP to be accessible by a residence. A wireless service provider only needs to install an AP and turn up the service. The remainder of this chapter explores the benefits of ubiquitous residential broadband Internet access, assuming the ease and economy of Wi-Fi is a catalyst for achieving the same levels of penetration for broadband Internet access as residential telephone service has today.

As the uses of broadband multiply, the value to subscribers rises far above the monthly subscription price. This is the consumer surplus from the innovation. Producers of new services that rely on the broadband (for example, i-mode-type services, Net2Phone, and so on), of products used in conjunction with broadband service (soft-switches, media gateways, IP phones, and residential gateways), and even of the broadband service itself also benefit from the greater diffusion of broadband. The producer surplus that is generated by sales is a real benefit to producers and, therefore, to the economy. Currently, no more than 8 percent of American households subscribe to a broadband service; only slightly more than 50 percent subscribe to an Internet service of any kind; and 94 percent subscribe to ordinary telephone service.[10] If broadband becomes ubiquitous, it would resemble current telephone service in its household penetration.

Producer Benefits

One of the reasons many IP backbone carriers went bankrupt is that they could not deliver bandwidth to a broad market. The bottleneck to the last mile remains the access controlled largely by telephone companies with their ubiquitous twisted-pair copper wire. Cable TV companies now service a majority of American homes.

Like Ethernet, Wi-Fi is not just about the technology itself-it is about what the technology empowers and the applications that result from the technology. Understanding the indirect effects of Wi-Fi is key to an investor's ability to profit from the technology. Figure 8-7 illustrates the Wi-Fi pull-through phenomenon on various segments of the PC and communications tech food chain. The following pages explain in more detail how Wi-Fi affects each technology segment discussed in the following sections. One example is the proliferation in Wi-Fi-enabled laptops. Intel will include 802.11a and 802.11b technology on their computer chips. Wi-Fi will probably become as common in computers as universal serial bus (USB) or Ethernet connections are now.

click to expand
Figure 8-7: The economic pull-through of wireless broadband, Source- Goldman Sachs

Wi-Fi will create a cycle of adoption that will drive technology purchases and upgrades by enterprises, retailers, service providers, government, and individuals for the three following reasons. First, it offers a means of delivery that is cheaper, simpler, smaller, and more convenient than a wired (telephone and cable TV) means of delivery. Wi-Fi service requires either the presence or installation of a broadband connection and/or traditional LAN equipment. The larger the environment-that is, the number of coverage areas, or hot spots, the number of users supported, and so on-the more infrastructure equipment and network bandwidth are required, thus spurring sales of APs, wireless routers, antennas, and so on.

Once Wi-Fi is available in a given area, it will spur the purchase of more mobile computers, personal digital assistants (PDAs), and other networked devices. This is particularly relevant in the home, where Wi-Fi enables broadband connections to be shared easily among multiple PCs and, ultimately, other devices. Major PC vendors are beginning to include Wi-Fi support now, and PC cards can be purchased for as low as $30, with prices continuing to drop. Chip-makers and laptop computer manufacturers now include Wi-Fi capabilities in their products.

New applications will drive the uptake of Wi-Fi. These market drivers include home networking, home multimedia, smart appliances, and voice over 802.11. These applications require new platforms such as home APs and Vo802.11 telephony devices.[11]

Computer Sales The expansion of the demand for broadband will create additional demand for computers and networked home appliances. As of June 2000, approximately 40 percent of all U.S. households do not have a computer.[12] Clearly, these households are not equipped to connect to the Internet at any speed. Of the 60 percent of households with computers, many will need to upgrade their equipment to obtain greater processing speed, more random access memory (RAM), or greater hard-drive capacity. Still others will choose to buy more advanced equipment such as storage devices, MP3 players for music downloads, and liquid crystal display (LCD) projectors for viewing video downloaded via a high-speed Wi-Fi connection. The applications that the following sections explore (video and telephony) could very well drive much of the remaining 40 percent of households without computers to make the leap and install a computer in their homes.

Crandall and Jackson estimate broadband's stimulus on household purchases of broadband-related equipment would be that U.S. household spending on computer equipment, peripherals, and software would resume its growth rate of 14.3 percent per year during 1991 through 1995, rather than continuing at its growth rate of 10.4 percent per year during 1995 through 1999. If growth returns to its pace in 1991 to 1995, by 2006, total spending would be $80 billion, rather than $66 billion-an increase of $14 billion. By 2011, the difference would be $53 billion per year. If the broadband revolution accelerated household equipment expenditures by another 3 percent per year to 17.3 percent annual growth, the additional spending in 10 years would be $110 billion per year.[13]

Consumer Benefits

The most straightforward estimate of the value of the enhanced availability of broadband derives from information on consumer subscriptions to broadband services. Currently, no more than 8 percent of households subscribe to a broadband service; only slightly more than 50 percent subscribe to an Internet service of any kind; and 94 percent subscribe to ordinary telephone service. Were broadband to become ubiquitous, it would resemble current telephone service in its household penetration.

An Estimate Based on Current Demand The price elasticity of demand is a relationship of the change in demand to the change in price. Given a current broadband penetration of 8 percent and an average price of the service of $40 per month, total broadband revenues may be estimated at $480 times 8.4 million, or $4 billion per year. Assuming that the demand for such service is linear with an elasticity of -1.0, the value of the service to these consumers-the consumer surplus-is $2 billion per year in addition to the $4 billion they pay. If the demand elasticity is -1.5, the consumer surplus falls to $1.4 billion.

If broadband spread to 50 percent of households at $40 per month through a shift of a linear demand curve with constant slope, the annual expenditure on the service would rise to $31.2 billion. At 50 percent penetration, the additional value to consumers would rise to between $80 billion and $121 billion per year at these two price elasticities. If broadband service became truly ubiquitous, similar to ordinary telephone service, annual consumer expenditures on the service would rise $58.7 billion per year, assuming the continued shift of the linear demand curve had a constant slope and an annual price of $480. The additional value to consumers-over and above their expenditures on the service-would be $284 billion to $427 billion per year, assuming that the linear demand curve with a current elasticity of -1.0 or -1.5 simply shifted outward. Table 8-8 illustrates this price elasticity of demand.

Table 8-8: Estimated ultimate annual consumer surplus from increased broadband penetration

Current Price Elasticity of Demand ($ Billions)

 

-1.5

-1.0

At 8 percent penetration

1.4

2.0

At 50 percent penetration

80.0

121.0

At 94 percent penetration

284.0

427.0

Source: Crandall and Jackson

Home Entertainment By the mid-1960s, a majority of homes in the United States had a television. This would now be called a wireless residential video service. Subscribers were limited in content to three channels of programming focused on the evening hours known as prime time. Those subscribers were forced to be present in front of their video monitors at precisely the time of the broadcast. There was no means of storing the program for viewing at a later time. The coming of cable TV and videocassette recorders (VCRs) in the following decades added some flexibility to the TV viewing experience.

Before cable TV and VCRs, subscribers were entirely at the mercy of the programmers. They had to watch what the programmers offered. The ability to choose programming drove the growth of cable and VCRs, leading to myriad new businesses, including cable TV companies and video rental firms. For most programming (films and prime-time TV shows), the production costs were very high and distribution was costly. This presented a high barrier to entry for any competitors.

The coming of the Internet held out some promise for a break in this regime. However, one barrier remained to a mass market for video delivered via the Internet-the bandwidth bottleneck. At the time of this writing, less than 10 percent of U.S. households enjoy broadband Internet. Dial-up Internet access at speeds up to 56 Kbps is too slow to download a feature-length film. DSL Internet access with speeds of around 256 Kbps is markedly better, but still requires a few hours of download time to receive a feature-length film.

What exactly constitutes a viable online video service? Is a short news or sports broadcast from a TV station made available via the Web a viable video service paid or unpaid? Is delivering an advertorial from an advertiser a commercially viable video service? Does the video have to be viewed in real time using video streaming or can the video file(s) be scheduled for an overnight download on the subscriber's computer? Can the subscriber have multiple computers to download video files around the clock? Is a feature-length film the only commercially viable form of Internet-based video service? Or are 30-minute serials or sitcoms also commercially viable online? What if the subscriber downloads or shares video files of all shapes and sizes at a completely random sequence, stores them in a digital video library, and views them completely at his or her leisure. One efficient form of delivery is to cache popular video files (films) on a local server, thus speeding the download. The subscriber can then use a PC to download video for viewing at his or her leisure, not unlike Personal Video Recorder technology like DirectTV's TiVo.

The freedom to obtain video files at little or no expense and store them in personal libraries presents a marked departure from the programmer-centric paradigm of the 1960s. This marks a shift in power of selection from the content programmer to the subscriber as the subscriber now chooses what he or she will view. More importantly, it determines what the subscriber is willing to pay for-the ultimate test of a product's commercial viability in a free market economy.

At the time of this writing, file sharing of video files, including feature-length films, has been available online for years. In October 2002, a start-up firm named Cflix launched a paid video download service offering a variety of feature-length films and some video serials such as the popular animation South Park. One month later, a consortium of Hollywood firms launched a service called MovieLink, which offers recent Hollywood releases for a fee per download via broadband Internet connections.

Broadband Internet via 802.11 with download speeds at around 6 Mbps makes the delivery of video services much more viable. A number of web sites coordinate the sharing of even the most recent releases of feature-length films. This sharing of files validated the technical viability of video file sharing and downloading long before the emergence of Cflix and MovieLink.

Making Money with Online Video So how does one make money on video on the Internet? MovieLink charges $3 to $5 per feature-length film. The subscriber doesn't get to keep it for that price. After downloading, the buyer has 30 days to activate the movie, but once the video file is opened, the subscriber can watch it as often as he or she wants within a 24-hour period. Then it disappears from the hard drive.

A Cflix subscription, which includes some basic programming, costs $4 per month-considerably cheaper than other video services. Cflix subscribers pay an additional $1.99 for older movies and $3.99 for new releases. They can attach equipment to their computers that enables them to watch the movies on a TV set.[14]

MovieLink is backed by Warner Brothers, Paramount, Universal Studios, MGM, and Sony Pictures, and offers people with broadband Internet connections downloads of about 175 recent and classic movies. To be truly useful, MovieLink has to offer thousands of titles, not a couple hundred. Even with an Internet connection capable of transferring 1 Mbps, it would take around 10 hours to download a 2-hour movie at that compression setting. More advanced compression technology, such as MPEG-4, may be able to cut the bandwidth in half while maintaining equal quality, but the download times would still be prohibitive.[15]

How can the economic scope and scale of video over broadband be measured? History is probably the best teacher. The multichannel video revolution of the 1980s and 1990s created enormous value for consumers. This explosion in choice created between $77 billion and $142 billion in annual value beyond the costs of the service (consumer surplus).

The Napster sensation probably provides only a prelude to what is possible over household broadband connections. At its height, Napster traffic was over one-third of the total traffic on the Internet. Downloading motion pictures or other video material, interactive television, interactive games, and even home editing of digitized entertainment material have not even begun in earnest. It is reasonable to assume that eventually the contribution to the consumer surplus of new video and other entertainment options created by the widespread diffusion of broadband Internet access would be at least as great as that already created by cable TV and direct broadcast satellites. This provides the model with an estimate of $77 billion to $142 billion per year. This estimate is consistent with the remarkable growth in home-entertainment spending by U.S. consumers since 1980, the year that cable began to grow as the result of the FCC's deregulation of cable signal carriage (see Table 8-9). Total expenditures have risen by about $56 billion since 1980 (nominal dollars). If the price elasticity of demand at the current prices is substantially less than one, the consumer surplus from this increase could easily be more than $100 billion per year. Crandall and Jackson forecast the next broadband revolution-delivered by the broadband Internet (potentially wireless)-should be of equivalent value.[16]

Table 8-9: Total spending on home entertainment (1980-1999)

Year

Expenditures by U.S. Households ($ Millions)

1980

4,657

1981

6,292

1982

8,199

1983

10,374

1984

12,742

1985

14,708

1986

16,915

1987

19,869

1988

23,250

1989

23,685

1990

29,822

1991

32,160

1992

34,009

1993

38,016

1994

39,513

1995

42,380

1996

46,647

1997

50,730

1998

55,231

1999

60,765

Broadband Access and Telephone Services

A previous chapter outlined the technical aspects of routing voice over 802.11. What are the economics of replacing dial tone provided by the telephone company with voice over 802.11? As Vo802.11 is VoIP using 802.11 as a means of access, the best approach is to study how VoIP replaces the PSTN dial tone in residential and small business telephone service.

Advantages of Vo802.11 Vo802.11 has a number of distinct economic advantages over using the PSTN and cell phone services. First, the cost of cell phone service is decreased by using Vo802.11 telephony in an office or any 802.11-serviced locale. Some new technologies allow a voice-enabled PDA to be dual-channel 802.11 and Code Division Multiple Access/Global System for Mobile Communications (CDMA/GSM). This capability enables an employee to talk over the 802.11 WLAN at the office or 802.11-serviced home or home office when in those offices. Once the employee leaves the office, he or she can switch over to a cell phone service provider if he or she has to make or receive calls.

Using Vo802.11 in the office can eliminate the cost of long-distance interoffice phone bills. Nearly 70 percent of corporate telephony is interoffice calling. This is an expense that can be eliminated by moving a company's telephony onto its corporate network. If 802.11 becomes a primary means of access within the company, then Vo802.11 would potentially eliminate much of a firm's phone bill.

A firm could eliminate all of its interoffice long-distance expenses by deploying a VoIP and 802.11 system. Calls routed over the corporate WAN would free the company from costs associated with longdistance phone service. Local phone service costs could be eliminated as well. If firms employed dual-frequency telephone handsets, all interoffice calls could be made on the corporate WAN. Local calls could also be routed to other 802.11 or IP-enabled handsets without contact with the PSTN. Other handsets could be reached using the cell phone network.

Soon the demand for broadband will reflect not only the growing potential uses of the Internet, but also the prospect for using these broadband connections to obtain voice telephone services currently provided over a narrowband connection. The use of broadband access to carry voice-ordinary telephone calls-as well as data will deliver substantial savings to consumers that are not captured in current demand estimates. Voice communications can be compressed, put in packets, and sent over an IP connection.

Microsoft Windows XP includes Session Initiation Protocol (SIP) software, which enables voice calls over the Internet-usually all one has to do to enable voice communications on a networked PC is to plug a headset into the audio ports on a computer and run the software. A broadband connection can support several voice connections-the exact number depends on the speed of the connection and the degree of compression of the voice signal. The current structure with two networks in the home (a voice and an IP network) and two connections to the outside world (a narrowband analog connection and a high-speed digital connection) appears inefficient. However, the transition to Internet telephony will take many years.

The cost savings from integrated access will be significant. Reliable Internet telephony would eliminate the need for second or third lines in households for teenagers or fax machines. The FCC estimated that the average household spent $55 per month on local and long-distance telephone service in 1999, and each household with telephone service had 0.289 additional lines.[17]

Within a few years, broadband access will permit consumers to substitute other services for these services that now cost $55 per month. The FCC estimates that the average residence spends $34 per month for local telephone service and $21 for long-distance telephone service. Part of that local telephone service cost is for the loop that is used for the broadband service. Consumers continue to incur most of those loop costs when broadband service is used, but they avoid the cost of the analog line card, the voice switch, and the voice transmission lines. Vo802.11 should lower the costs of both local and long-distance telephone service while providing residences with the equivalent of several telephone lines. Crandall and Jackson estimate that such savings could average $25 per month per household. In addition, households with broadband service would get the equivalent of multiple voice telephone lines. It is estimated that this additional service or option of service could be worth $10 per month to the average household.[18] Thus, in the longer run (say, a decade from now), broadband access could deliver voice communications benefits of about $35 per month, or $420 per year, to the average household with telephone service. If it is assumed that 122.2 million households have telephone service, these benefits would total $51.4 billion per year, assuming no growth in voice usage occurs. The actual value could be much higher.

The substantial economic benefits (principally savings from expenditures on telephone service) created by providing multiple services over a high-speed line almost cover the cost of a high-speed line-it has been estimated that benefits of $35 per month are created by a broadband connection that costs $40 per month. These savings are one reason why it seems reasonable to expect that the fraction of households with high-speed access services will ultimately approach the fraction that has telephone service today.

Speeding up the adoption of broadband access provides benefits sooner. The present value of the difference between the base adoption scenario and the much faster adoption scenario of the previous example is 140 percent of one year's worth of the benefits of ubiquitous broadband adoption by households.[19] Thus, if one assumed that, when fully adopted, broadband would generate benefits of $300 billion per year to U.S. consumers, a policy change that moves our society from the baseline adoption curve to the much faster curve would generate benefits with an NPV of about $420 billion.[20] The increase in the present value of producers' surplus would be about $80 billion. This acceleration is therefore worth $500 billion to U.S. consumers and producers. Table 8-10 lists the consumer benefits of universal broadband deployment.

Table 8-10: Summary of consumer benefits from universal broadband deployment ($ billions per Year)

Source

Low Estimate

High Estimate

Direct Estimates

Broadband access subscription

284

427

Household computer and network equipment

13

33

Total Benefits

297

460

Alternative Estimates-Benefits Deriving from:

Shopping

74

257

Entertainment

77

142

Commuting

30

30

Telephone services

51

51

Telemedicine

40

40

Total Benefits

272

520

How could speeding up the adoption of a technology have such massive benefits? The key lies in the substantial benefits that ubiquitous broadband can convey to consumers. Once virtually everyone has the service, the network effects from developing new services will become very large. Moving these benefits forward a few years can create very large benefits-even when evaluated from today's perspective. The powerful advantage of 802.11 over the current, dominant broadband technologies (DSL and cable modem) is that in the words of Clayton Christensen,[21] it is "cheaper, simpler, smaller, and more convenient to use." The lack of a requirement for wires and their incumbent, expensive rights-of-way could potentially give the wireless service provider a significant advantage over the wired incumbent.

[10]The number of broadband subscribers (DSL and cable modems) was 7.3 million as of March 2001. See Crandall, Robert and Charles Jackson, "Failure of Free ISPs Triggers First-Ever Dip to 68.4 Million Online Users: Cable Modem Boom Continues as DSL Sign-ups Lag," Telecommunications Reports (April 2001): 1. The estimates for Internet and telephone service are from the authors' tabulations using the Current Population Survey for August 2000.

[11]Chris Fine, "Watch Out for Wi-Fi," a white paper from Goldman Sachs, September 26, 2002.

[12]The most recent estimate from the Bureau of the Census for June 2000 was 41.5 percent. More recent estimates from TNS suggest that about 50 percent of households now have access to the Internet. See TNS Telecoms, ReQuest Market Monitor National Consumer Survey, v.3 (2001).

[13]Robert Crandall and Charles Jackson, "The $500 Billion Opportunity: The Potential Economic Benefit of Widespread Diffusion of Broadband Internet Access," a white paper from Criterion Economics, LLC, July 2001.

[14]Dan Luzadder, "Video Service Gives It That New College Try," Denver Post, www.denverpost.com/cda/article/print/0,1674,36%257E33%257E940472,00.html, October 22, 2002.

[15]Steven Wildstrom, "Hardly a Boffo Start for MovieLink," Business Week, www.busi-nessweek.com/technology/content/nov2002/tc20021115_6283.htm, November 15, 2002.

[16]Robert Crandall and Charles Jackson, "The $500 Billion Opportunity: The Potential Economic Benefit of Widespread Diffusion of Broadband Internet Access," a white paper from Criterion Economics, LLC, July 2001.

[17]FCC, "Trends in Telephone Service," 2nd Report, www.fcc.gov/Bureaus/Common_Carrier/Reports/FCC-State_Link/IAD/trend200.pdf, 2000.

[18]The FCC's numbers indicate that the average household with telephone service has 1.289 access lines and pays local service fees of $34 per month. Assuming that all lines cost the same (which is not quite right but is reasonable), the average household with telephone service in 1999 paid $7.62 per month for additional line service. If those households without a second line today place an average value of no more than $3.40 per month for a second line of service, then the average household will value a second line at $10 per month or more.

[19]This was calculated using a discount rate of 10 percent and assuming a 2 percent per year growth in the economy.

[20]These present values are 2.8 and 4.2 times the ultimate value of broadband adoption when evaluated at an interest rate of 10 percent per year.

[21]Christensen, Clayton, "Innovator's Dilemma-When New Technologies Cause Great Firms to Fail (Management of Innovation and Change Series)", Harvard Business School Press, 1997. This book coined the term "disruptive technology" which is defined as being "cheaper, simpler, smaller and more convenient to use."



Wi-Fi Handbook(c) Building 802.11b Wireless Networks
Wi-Fi Handbook : Building 802.11b Wireless Networks
ISBN: 0071412514
EAN: 2147483647
Year: 2003
Pages: 96

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net