Building the NII: Will the Shareholders Come? (And if they
Don't, Will Anyone Really Care?)
More often than not, public policy debates concerning the national information infrastructure (NII) begin with a presumption that the proverbial "information superhighway" will be built regardless of what the government does. The only thing public policymakers really have to worry about, the reasoning goes, is to ensure that users, rival service vendors, and equipment vendors have affordable access to the nation's interoperable network of networks.
Many knowledgeable observers further assume that the telecommunications and cable television industries will move aggressively to upgrade their respective networks over the next 5 to 10 years. They presumably will do this to take advantage of new market opportunities spawned by interactive multimedia services, and to respond to competition. Still others, including several key federal officials, contend that the government will play a positive and constructive role in facilitating significant amounts of capital investment needed to extend broadband networks to households, businesses, schools, hospitals, and other public institutions throughout the country.1
This paper examines these expectations from the perspective of telephone company shareholders. Several key issues are addressed. How do returns on Bell company investment in local network facilities compare with returns on investment opportunities outside local telephone markets? Have shareholders been rewarded or penalized by Bell company decisionsgiven the prevailing regulatory environmentto upgrade their respective wireline telephone networks in recent years? On balance, do shareholder returns matter much to anyone other than the shareholder and telephone company managers, and if so, to whom, how, and why?
In a recent Delphi survey, the consulting firm of Deloitte & Touche questioned 120 executives in the information, communications, and entertainment industries about how soon they expect various new communications products and services to arrive on the scene, and how rapidly markets for these products and
NOTE: Robert T. Blau, Ph.D., CFA, is executive director of policy analysis for BellSouth Corporation. He would like to thank Stephen Barreca, manager, infrastructure planning at BellSouth Telecommunications, for his valuable comments and assistance in developing aspects of this analysis that concern Bell company deployment of advanced network technologies. Views expressed in this paper are solely those of the author and are not necessarily shared by BellSouth or its subsidiaries.
services will develop.2 With regard to the commercialization of interactive multimedia networks and devices, the panel came to the following conclusions:
Interestingly, information industry executives surveyed by Deloitte & Touche also had a generally favorable view about the role of government in promoting investment in the NII. Approximately 62 percent thought the net effect of government actions by 1998–2000 will be positive (41 percent) or neutral (21 percent) "in terms of encouraging investment, fostering research and development, and promoting the rapid spread of advanced information, communications, and entertainment offerings."3
Satisfying User Expectations
Are expectations for building the information superhighway realistic? In several key respects, the answers to that question will rest with those individuals who will be asked to put up the significant sums of capital needed to upgrade the NII, and particularly ubiquitous local telecommunications networks where the lion's share of these costs will be incurred.
If shareholders believe that risk-adjusted returns on investment in advanced network technologies will remain competitive with returns on alternative investment opportunities, then those technologies will be deployed and the new service features they make possible will be brought to the market in a timely manner. If, on the other hand, shareholders do not regard prospective returns on network investment to be high enough to compensate for risk incurred, then lesser amounts of discretionary capital spending will be committed to new network technologies. In that event, telephone company deployment of new technologies will slow, possibly to the point of lagging user expectations and needs. This could be especially problematic for developers and users of new multimedia service applications requiring substantially more bandwidth than is readily available over the public switched-telephone network (PSTN).
The balance of this paper analyzes relationships between shareholder returns, network investment, and the deployment of the types of advanced network technologies that will make up the NII. It begins with a discussion of recent trends in each and their implications for future investment. The paper concludes with a brief discussion of steps that telecommunications policymakers must take to create a more technology-friendly market environment.
Relationships Between Network Investment and Shareholder Returns
Figure 1 highlights relationships between regional Bell company (RBC) investment in regulated wireline networks and total shareholder returns between 1988 and 1994. Network investment is expressed as the percentage of Bell telephone company operating cash flow used to acquire wireline network plant and equipment. Unlike reported earnings, which are subject to the vagaries of different financial and regulatory accounting practices, operating cash flow provides an accurate measure of cash that a business generates after it pays its out-of-pocket operating expenses, taxes, and interest on its debt.4
Besides financing the acquisition of new plant and equipment for their regulated wireline networks, local Bell telephone companies principally use their operating cash flow in one of two ways: to pay dividends to their regional holding company shareholders, or to finance investment opportunities outside local networks. The ratio of wireline network investment to cash flow from local telephone operations, therefore, is a good comparative measure of how the individual companiesand their shareholdersview the relative attractiveness of using available cash flow to upgrade their local network facilities.5 This is particularly true of the RBCs, since they currently finance nearly all capital expenditures (and pay all dividends) with internally generated funds (e.g., operating cash flow).
Figure 1 also depicts cumulative changes in total shareholder returns for each of the regional Bell companies between 1988 and 1994. Total shareholder returns include the percentage change in the price of an individual stock plus its dividend yield (i.e., dividends paid divided by the price of the stock at the time). For purposes of this analysis, cumulative shareholder returns are based on monthly returns and assume that all dividends paid on a particular stock are immediately reinvested in that stock.
Figure 1 highlights a definite inverse relationship between the ratio of network investment to operating cash flow from local telephone operations and total shareholder return. Differences in shareholder returns between the individual regional companies over the 1988–94 period also were quite substantial.
Given the size of the RBCs and the capital intensity of their local telephone operations, these differences in shareholder returnsand the emergence of an inverse relationship between capital spending on wireline network plant and equipment and shareholder returnscould have an important bearing on future investment in local network facilities. If the recent past is prologue, the level of discretionary capital spending on local wireline networks (i.e., capital expenditures over and above those required to maintain the quality of local telephone service at current levels) also will determine how rapidly broadband multimedia and other advanced service features are brought to the market.
Relationships Between Capital Spending on Local Telephone Plant andEquipment and the Deployment of Advanced Network Technologies
Tables 1 and 2 highlight the degree to which differences in wireline network investment among the seven RBCs are reflected in how rapidly each of the regional companies upgraded their respective local wireline networks. The tables depict penetration rates and substitution ratios, respectively, for ten advanced network technologies that the Federal Communications Commission routinely tracks through its ARMIS Infrastructure Reports.6 The RBCs and GTE file these statistics with the Commission annually, and they are currently available for 1989–93.
Penetration rates shown in Table 1 represent the percentage of a company's total number of switches, access lines, and communications channels equipped with a particular technology (e.g., ISDN, Signaling System 7, digital stored program control, fiber optics, etc.). Substitution ratios are based on the Fisher-Pry model commonly used to project ongoing increases in the penetration of new technologies expressed as a percentage of total usage.7
Technologies depicted in Table 2 also are categorized in three broad groupings designed to provide comparative measures of the following:
Substitution ratios for each of these groups were calculated by averaging the substitution ratios for the individual technologies that make up that group. These composite measures of technology deployment are presented along with the ratio of network investment to operating cash flow, and cumulative shareholder returns for the 1988–94 period.
Finally, Tables 1 and 2 rank order the seven RBCs by penetration and substitution rates for each of the individual technologies and the three composite technology groups described above, and by the ratio of network investment to operating cash flow and total shareholder returns.
As expected, those Bell companies that reinvested larger portions of cash flow from their local telephone operations in wireline network plant and equipment generally deployed advanced network technologies more rapidly. Figure 2 shows, however, that decisions by Bell company managers to accelerate the introduction of advanced network gear did not have a positive effect on shareholder returns. If anything, the opposite has been true. Between 1989 and 1993, for instance, BellSouth ranked first in overall network modernization, but only
fifth in cumulative shareholder returns for the 1988–94 period.8 Southwestern Bell, on the other hand, ranked last among the Bell companies in overall network modernization but first in cumulative shareholder returns.
These relationships, of course, do not imply that shareholders have some inherent aversion to the Bell companies upgrading their portion of the nation's information infrastructure. What the data suggest, however, is that new or incremental revenue and earnings opportunities that investors expected to result from the deployment of wideband and broadband technologies have not been large enough, at least in recent years, to compensate for the capital cost and financial risk of installing these facilities sooner rather than later. This has been true even though the RBCs' local telephone operations accounted for roughly 86 percent of their total revenues between 1988 and 1994, 89 percent of their combined operating cash flow, 95 percent of their total earnings before extraordinary charges (e.g., write-offs of obsolete telephone plant and equipment), and 94 percent of their net income. (See Table 3.)
Implications for Upgrading the NII
In view of the Bell telephone company contributions to the regional holding companies' overall earnings, why do shareholders seem at all concerned about using internally generated funds from local telephone operations to upgrade their wireline networks? To understand why this is so and what it means for future investment in advanced network technologies requires at least some appreciation of factors that investors consider in valuing common stocks. Besides current earnings and dividends, these factors commonly include earnings and dividend growth, and the degree of financial risk associated with owning a given company's shares.9 And while the Bell telephone companies' current after-tax income might seem reasonable, if not ample todayeither in absolute terms or as a percentage of the holding companies' net incomeinvestor attitudes about their future earnings
growth and business risk are decidedly less positive. This is due largely to an unsettled regulatory environment.
Table 3 compares recent growth in average revenues, operating cash flow, earnings before extraordinary charges, and net income for the seven regional Bell holding companies, their local telephone companies, and their unregulated business operations. For comparative purposes, these same data are provided for the companies that currently make up the S&P 500, a broad market average. The data show that growth of the Bell telephone companies' after-tax earnings, whether before or after extraordinary charges, was negative and well below average earnings growth for the S&P 500 companies. By contrast, after-tax earnings from the regional Bell companies' unregulated businesses grew at an average annual rate of 28 percent over the 1988–94 period. This
explains why securities analysts now attribute as much as 36 percent of the regional holding companies' total market value to their unregulated, nonlocal telephone company operations.10
While lackluster earnings growth among the Bell telephone companies are attributable to several factors, including slow growth in the general economy between 1988 and 1994, two industry-specific developments stand out. One relates to competitive entry into the most lucrative segments of local telephone markets (e.g., exchange access and intra-LATA toll). Competition has limited growth in traffic carried over the Bell company networks while forcing the local telephone companies to reduce prices. The combination of modest traffic and access line growth and lower rates, in turn, has kept local telephone company revenues, earnings, and operating cash flow flat.
At the same time, legal and regulatory barriers to market entry like the Modified Final Judgement (MFJ) inter-LATA restriction, the video program prohibition, and until recently the MFJ information services restraint have walled off the Bell telephone companies from new revenue streams that could substantially hasten the recovery of capital invested in advanced network technologies. Investors understand this and are cautious about the Bell companies spending billions of dollars on network capacity that legal and regulatory barriers to market entry may prevent them from using.
Their concerns are compounded by expectations that deploying broadband technology in an increasingly competitive market environment will change network economics in ways that render telephone company earnings more volatile than they have been in the past, thereby increasing the risk of owning their stock.11 Unlike today's copper based, narrow band telephone networks, tomorrow's wideband and broadband architectures will greatly increase bandwidth available to end users. Quantum increases in network capacity also will reduce the marginal or incremental cost of transporting communications traffic both in absolute terms and relative to the fixed cost of building and maintaining network capacity. Similarly, as telephone company operating leverage (i.e., the ratio of a firm's fixed costs to its total cost) increases, the addition, or loss, of traffic will have an increasingly pronounced impact on earnings since large portions of those gains or losses in marginal revenue will flow directly to the bottom line.
Prospects that local telephone companies could lose a significant portion of their local telephone revenues to rival vendors have been made more apparent in recent years by growth in demand for wireless communications services and the Internet. The WEFA Group recently forecast, for instance, that over the next 10 years, increases in network capacity available on commercial wireless communications systems (e.g., cellular telephone, personal communication services, and so on) will be large enough to accommodate not only natural growth in demand for mobile telephone services, but also nearly all narrowband voice and data traffic carried over wireline networks, as illustrated in Figure 3.12 Should these forecasts pan out, wireless operators will no doubt attempt to leverage their "excess capacity" by integrating into traditional wireline telephone markets. If they do, price competition between wireless and wireline carriers will intensify, and wireless systems will capture voice and data traffic traditionally carried over wireline telephone networks.
The same could be said of the Internet. As Christopher Anderson of the Economist magazine recently observed.13
If the Internet does become a "data dialtone," regular dialtone might finditself out in the cold. It isalready possible to make telephone calls on the Internet from speciallyequipped computers; the spreadof multimedia PCS and faster Internet connections could make this commonplace.At the same timecompanies are turning an increasing amount of their telephone traffic intodigital data and sending itthrough private data networks, saving up to half their telecom costs.Regulation permitting, this trafficcould eventually move to the Internet. For the telephone companies, the onlydecision is whether theyparticipate in the cannibalization of their revenues or watch it happen.
In light of prospects for increased competition, low earnings growth, and increased business risk, it is understandable why shareholders appear to have rewarded the Bell companies for minimizing discretionary capital spending on advanced network technologies in recent years. Overall, Bell company investment opportunities outside their regulated wireline networks have simply been more attractive in recent years primarily because these investments offer better earnings potential, often with less risk.
Though certainly of interest to investors and Bell company managers, the question remains whether inverse relationships between network investment and shareholder returns really matter much to anyone else. The answer, it turns out, will likely depend on growth in demand for access to the Internet, and the types of multimedia services that are just beginning to emerge on the World Wide Web.
The Internet Community's Stake in Local Network Investment
According to the Internet Business Center, demand for access to the World Wide Web (WWW) increased at annual rates of 443,931 percent in 1993 and 1,713 percent in 1994, bringing the total number of WWW users in the United States to an estimated 4 million. The Internet Society further predicts that between June 1995 and January 2000 montly traffic on the WWW (measured in bytes) will increase by a factor of 50,000! Although no
one knows whether such forecasts will prove to be at all accurate, there is little question that the Internet will continue to expand at unprecedented rates for the foreseeable future.
Internet growth is of particular interest to local telephone companies for three key reasons. The first, as referenced above, has to do with prospects that significant amounts of voice and data traffic that traditionally have been carried over the PSTN could conceivably be routed over the Internet. Because use of the Internet today is effectively free, this shift would enable consumers to bypass the system of usage-based pricing arrangements and subsidies that the telecommunications industry has long relied on to recover its costs while keeping basic residential telephone rates at universally affordable levels (i.e., below cost).
At the same time, Internet traffic rides on network capacity leased from facilities-based telecommunications carriers. As such, growth in demand for services like the WWW also represents new market opportunities for local telephone companies. Routing Internet traffic should result in more intensive and more efficient use of embedded network facilities. If priced properly, it also could help recoup the cost of deploying
wider band technologies that Internet users will need if multimedia service applications available on the WWW are to continue to evolve.
Third, and perhaps most important, because multimedia applications on the WWW consume far more bandwidth than do electronic mail or other text-based services on the Internet, accommodating rapid growth in demand for access to the WWW could prompt local telephone companies to expand the capacity of their respective wireline networks significantly. The need for additional capacity should become all the more apparent as information resources available on the WWW continue to proliferate.
As new resources come online, demand for access to the WWW will increase along with its value to users as well as information service providers. Similarly, as the value of the WWW increases (e.g., by the square of the number of new users added to it during any given period), online sessions also should increase in duration (e.g., from current levels of 25 minutes per session versus an average of 5 minutes for local voice telephone call) for the simple reason that there will be more users and services to interact with. The combination of more businesses and residents spending more time online, accessing increasingly sophisticated multimedia services that require substantially larger amounts of bandwidth to transport, could press the limits of many local telephone networks within a relatively short period of time.
Near term, this demand for added capacity on the PSTN will be handled through the deployment of ISDN, asymmetrical digital subscriber loop (ADSL) equipment, and other advanced technologies that enable local telephone companies to increase bandwidth that can be made available to individual users over existing copper phone lines. Longer term, as demand for interactive full motion video applications develops, narrowband and wideband network technologies will need to give way to fiber optics or architectures that integrate fiber optics and coaxial cable capacity. Either way, the capital cost of accommodating WWW and other multimedia applications could be substantial.
Rendering a local telephone line ISDN capable, for instance, typically costs of $100 and $200 for lines already served by a digital switch, and between $300 to $500 if digital switch capacity has to be installed.14 At year end 1993, Southwestern Bell had only 11.2 percent of its access lines equipped with ISDN, while 45.5 percent of its lines were served by digital switches. At these levels of penetration (and assuming ISDN installation costs fall at the midpoints of the latter two ranges), the total cost of making the company's remaining 88.8 percent of access lines ISDN ready would be roughly $3.4 billion. As $3.4 billion represents 42 percent of Southwestern Bell's total capital expenditures during the 1988–93 period, a commitment by the company to make ISDN widely available, say in the next 5 years, would constitute a major capital spending decision. Unless prospective earnings from its local telephone operations improve significantly, making such a commitment also would arguably be at odds with the company's fiduciary responsibility to maximize shareholder value.
Public Policy Recommendations
There is no question that the government will play a key role in balancing shareholders' interest in realizing reasonable returns on network investment, and consumer interests in seeing the PSTN upgraded. How public policymakers play their part also will determine when advanced technologies will be deployed, where, and on what scale. And though there are several steps they could take to ensure that the capabilities of the PSTN keep pace with user needs, three changes in telecommunications policies should prove especially helpful in this respect.
Immediately Replace All Legal and Regulatory Barriers to Market Entry with User-Friendly Interconnection Requirements
Given the less-than-favorable relationships between network investment and shareholder returns, coupled with substantial increases in bandwidth that the deployment of fiber optics, ISDN, ADSL, and other digital network technologies could bring about, it is imperative that federal and state officials immediately abandon any and all regulations that restrict how the PSTN is used. Notwithstanding the value of abundant bandwidth to the Internet community, other users, and growing numbers of information intensive industry groups that rely on
telecommunications networks to produce, distribute or add value to their wares, local telephone companies will not invest in advanced network technologies and capacity that they may be precluded from using by law or regulation.
Policymakers should recognize this fact and focus on developing a set of user-friendly network interconnection requirements that will encourage the development of a wider array of new service applications over which the capital cost of upgrading local telephone networks can be spread. Telecommunications legislation recently adopted by the U.S. Senate represents an important and much needed step in this direction. If enacted, the bill would allow the RBCs to provide inter-LATA long distance and video program services and to manufacture telecommunications equipment provided they comply with a 14-point checklist of interconnection requirements that will open up local telephone markets to more competition.
As Frank Governali, a leading telecommunications securities analyst with First Boston, put it:15
We'd love to see a bill passed. One, we think it would be good for thecompanies and the stocks to getthis uncertain process behind us. Second, selfishly, we're tired of having toobserve the inane activitiesof Washington so close-up. Although there is no chance that the current Senatebill is exactly the sameas the one that may ultimately get [enacted into law], we think passage of areasonably similar billwould be viewed positively by the market and cause many of thetelecommunications companies' stocksto rise. By having a bill passed, investors could then observe the newoperating environment and try topick those companies that could do well in the new environment. Without abill, the selection processbecomes more difficult and the stocks more volatile, which is what we've seenover the past twelvemonths.
Complete the Transition from Cost Plus Rate-of-Return Regulation to Price Regulation
In addition to replacing legal and regulatory barriers to network utilization with pro-competitive, user friendly interconnection requirements, federal and state officials need to complete the transition from rate-of-return regulation to a pure form of price regulation for noncompetitive telecommunications services. Moving to a pure form of price regulation also should have a favorable impact on efficient network investment in two important respects.
First, like telecommunications legislation, price regulation would eliminate a considerable amount of regulatory uncertainty that has discouraged network investment in recent years. Price regulation would accomplish this by eliminating any need for regulators to concern themselves with the types of capital expenditures regulated telephone companies are allowed to undertake, how local network costs are allocated between different services for rate-making purposes, or how rapidly new network plant and equipment is depreciated. Because these types of regulatory requirements are especially susceptible to being "gamed" by rival service providers for purely anticompetitive ends, their removal would make for a much more predictable and investment-friendly market environment.
Equally important, price regulation would give local telephone companies a much-needed opportunity to increase returns on network investment, while affording consumers much better protection from having to pay the cost of bad investment decisions by regulated carriers. Under price regulation, a local telephone company's shareholders would bear the brunt of ill-conceived or poorly timed investment in fiber-optic feeder plant or some other advanced technology, since price constraints would preclude the company from automatically passing these costs through to the ratepayer. At the same time, however, if a company's investment in advanced network technologies and capabilities succeeded in raising earnings, then those benefits would accrue to its shareholders, much as they would in any other business. Incentives to invest in new technologies would then improve.
Shifting investment risk from the local telephone companies' customers to their shareholders is especially important in today's market environment. This is primarily because consumer demand for many new multimedia services that telephone companies will rely on to help recoup the cost of deploying wideband and broadband network capacity will be far less predictable and, in all likelihood, far more volatile than demand for plain old
voice grade telephone service. Policymakers need to recognize that the emergence of a less certain and more volatile market environment will necessarily raise the risk of investing in new telecommunications technologies. And that unless regulatory pricing policies are adjusted in ways that compensate shareholders for bearing that added risk, capital expenditures on network facilities will slow, and network capabilities will likely lag behind user needs.
Allow Local Telephone Companies to Fashion New Business Models for Accommodating Internet Users
Finally, if rapid growth in demand for Internet services continues to reshape the face of electronic communications, as many believe it will, local telephone companies and their regulators will need to experiment with new models for providing and pricing bandwidth that Internet users will no doubt want. What these models might entail is, at this point, unclear. Suffice it to say, however, that they will need to balance the telephone companies' need to recoup the cost of deploying advanced network technologies with the Internet community's desire to keep the Internet free of toll or other usage-based charges.
How these needs and interests ultimately get worked out will likely be accomplished through a considerable amount of trial and error. In some instances, regulators may need to give residents as well as business customers the latitude to pay part of the front end cost of installing advanced service features such as ISDN or ADSL in exchange for getting access to those features sooner and at lower monthly rates. In other areas, local telephone companies may want to provide customers free access to the Internet in exchange for agreeing to subscribe to a second telephone line for some specified amount of time.
In any case, new business models that telephone companies follow in responding to Internet user requirements may differ somewhat from ways local telephone service traditionally has been packaged and priced. Federal and state officials should recognize the need for flexibility in this area and allow any reasonable trials to go forward, preferably with minimal regulatory delay.
1. The estimated cost of deploying a ubiquitous broadband network in the United States ranges from $1,000 to $2,000 per business and residential access line depending on how much fiber and associated optoelectronics is deployed in telephone company feeder plant. See, for example, Egan, Bruce L. 1994. ''Building Value Through Telecommunications: Regulatory Roadblocks on the Information Superhighway," Telecommunications Policy 18(8):580–583.
2. Of the 120 executives surveyed by Deloitte & Touche, 25 percent were with telecommunications companies; 24 percent were with broadcast or cable television firms; 21 percent were with consumer electronics or personal computer manufacturers; 14 percent were with publishing/advertising firms; and 16 percent were with entertainment companies. Fifty-five percent of the respondents were chairs, presidents, or chief executive officers, 30 percent were executive vice presidents or vice presidents, and the remaining 30 percent were senior managers. See Deloitte & Touche. 1995. Interactive Multimedia Age II: Report on the 12-Question Update Survey. Deloitte & Touche, May 1995.
3. Ibid., pp. 2–3.
4. See Rappaport, Alfred. 1986. Creating Shareholder Value: The New Standard for Business Performance. Free Press, New York, pp. 19–45.
5. See Hackel, Kenneth S., and Joshua Livnat. 1992. Cash Flow and Security Analysis. Business One, Irwin, New York, pp. 138–214.
6. See Kraushauer, Jonathan M. 1995. Infrastructure of the Local Operating Companies Aggregated to the Holding Company Level. Federal Communications Commission, Washington, D.C., April.
7. See Fisher, John C., and Robert H. Pry. 1971. Technology Forecasting and Social Change. American Elsevier Publishing Company, New York.
8. Because there are time lags between capital expenditures on various technologies depicted in Table 1 and their actual deployment, and additional lags between the deployment of those technologies and incremental sales/earnings growth that should result from their availability, it was appropriate to compare network investment/shareholder returns for
1988–94 with the Federal Communications Commission's technology deployment data, which is currently available for the 1989–93 period only.
9. The discounted dividend model holds that P = E(D)/(k - g) where
See Bodie, Zvi, Alex Kane, and Alan J. Marcus. 1989. Investments. Irwin, Homewood, Ill., pp. 473–480.
10. See Yanis, Steven R., and Thomas J. Lee. 1995. "The Regional Holding Companies Are More Than Plain Old Telephone Companies," Telecommunications Services, Oppenheimer & Company Inc., January 26, p. 8.
11. Increased earnings volatility adds to the risk of owning a share of stock, because it raises the probability that the company in question may not be able to pay future dividends or, for that matter, sustain day-to-day business operations. See Bodie, Zvi, Alex Kane, and Alan J. Marcus. 1989. Investments. Irwin, Homewood, Ill., pp. 130–143.
12. WEFA Group. 1995. Economic Impact of Deregulating U.S. Communications Industries. WEFA Group, Burlington, Mass., February, p. 29.
13. Anderson, Christopher. 1995. "The Internet Survey," The Economist, July 1, p. 18.
14. See Egan, Bruce L. 1994. "Building Value Through Telecommunications: Regulatory Roadblocks on the Information Superhighway," Telecommunications Policy 18(8):580–583.
15. See Governali, Frank. 1995. Weekly Industry and Valuation Report, First Boston, June 16.