Health of the Telecommunications Sector: A Perspective from Investors and Economists

Subcommittee on Telecommunications and the Internet
February 5, 2003
1:00 PM
2123 Rayburn House Office Building 

Mr. Robert Crandall
Senior Fellow
The Brookings Institute
1775 Massachusetts Avenue, NW
Washington, DC, 20036

Mr. Chairman and Members of the Subcommittee, I am pleased to be here this afternoon to express my views on the current state of the U.S. telecommunications sector. This is an important topic, given the severe downturn that has occurred in equity values and capital expenditures in the sector. Indeed, telecommunications is suffering the greatest financial distress and general turmoil that I have seen in the roughly 25 years that I have been studying this industry.

As we focus on the current dire condition of the telecommunications sector in this proceeding, we should recall that we have been through similar problems in other sectors after they were first opened to entry and subjected to deregulation. For example, similar, but less severe problems gripped the airline and trucking industries shortly after they were deregulated in 1978 and 1980, respectively. A surge in investment by the airlines led to excess capacity that was exacerbated by the rather deep 1982 recession. As a result, airline stocks tumbled in 1982. Similarly, the natural gas pipeline industry suffered through much more severe adjustments to natural-gas and oil price deregulation in the 1980s. Finally, no one needs to be reminded of the recent difficulties that California utilities faced when “deregulation” was instituted in that state in the 1990s. In each case, policymakers, industry participants, and economists could not predict how these industries would adjust to entry and deregulation or who the winners and losers would be. We simply knew that prices would eventually be lower under deregulation, and that output would be greater. 

   These observations are important, because, as my colleague Clifford Winston has found, deregulation has created much larger benefits in virtually every instance than many economists and other observers would have expected.[1] Given the turbulence that often follows from opening markets to competition and from deregulation, we should have expected the 1996 Telecommunications Act to create similar turmoil. The 1996 Act was not as deregulatory as many of the earlier statutes, but it opened a regulated market to competition and thereby threatened to place considerable pressure on an incredibly distorted regulatory rate structure. For this reason, in October 1996, I opined in a speech in Maine that “all hell would soon break loose” in the telecom sector. As it turned out, my prediction was premature, and it may have been right for at least some of the wrong reasons. Neither I nor anyone else could have predicted the 1998-2001 stock market bubble that swept the sector, the nature of the competition that would develop, the regulatory policies that would be adopted under the new Act, nor the telecommunications market’s response to all of these events. 

The Telecom Surge -- Rhetoric Meets Reality 

   The years leading up to the 1996 Act were a period of excited discussion about the potential of the “Information Superhighway”[2] and the promise of network convergence. The declining cost of fiber-optics transmission and the technological progress driving microprocessor technology (at a rate described by Moore’s Law) led some to predict that communications bandwidth would soon be virtually free.[3] Households, physicians, teachers, and businesses would be able to send and receive high-speed video images that would substitute for personal diagnoses, provide remote monitoring, allow remote tutoring, and supply much more personalized access to entertainment. Once the telecommunications sector was opened to competition and deregulated, innovation could flourish, thereby allowing subscribers access to new services and providing existing services at dramatically lower prices.

Capital Spending and Market Valuations 

The 1996 Act was not deregulatory. It created a vast new system of wholesale-price regulation of local services that was only vaguely spelled out in the statute. Indeed, local telecommunications would continue to be intensely regulated by the states and the FCC. On the other hand, long distance and wireless services were essentially deregulated before the Act was passed. Both sectors had begun to invest heavily in infrastructure in the early 1990s, and the lure of the Internet would entice them to accelerate this investment after 1996. Moreover, new local carriers sprouted from everywhere and were able to attract enormous amounts of capital. The result was an investment boom that continued for more than five years.  

   Source: U.S. Department of Commerce, Bureau of Economic Analysis.(“BEA”)  

   Figure 1 shows the acceleration in capital spending in the telecommunications sector that occurred after 1995 in both nominal dollars and constant dollars.[4] Between 1987 and 1996, nominal and real (inflation-adjusted) capital spending increased at average rates of 4.8 and 4.5 percent per year, respectively. In the next four years, however, the growth rate soared to more than 20 percent per year. By 2000, real capital spending had risen 148 percent from its 1996 level. This surge in capital spending was accompanied by an even greater rise in the prices of telecommunications industries equities—a stock market “bubble” that burst with a vengeance in 2000-01. (Figure 2)[5] 

The bubble gripped three of the four major groups of carriers: the new CLECs, the wireless carriers, and the long-distance companies.[6]  The largest rise was in the CLEC index in Figure 2, which also shows the greatest collapse. The wireless stocks were next in the upsurge, and the long-distance stocks were third. The Bell companies enjoyed much less of a surge in 1998-2000 and suffered less in the downturn as they essentially tracked the S&P 500, which had a much more modest bubble.

   Source: Author’s calculations from closing prices on   www.finance.yahoo.com

   Of the facilities-based long distance companies, only AT&T remains with any market capitalization attributable to long distance, and even AT&T has lost roughly one-third of its market cap in the two months since it spun off its broadband division. It now appears that the long distance companies must offer a wider bundle of services to survive.

   At this point, it appears that very few of the new competitive local carriers (“CLECs”) are likely to survive and prosper. Once the repository of more than $80 billion in market capitalization, the publicly traded CLECs now have a scant $1 billion in total market cap after reporting more than $40 billion of spending on capital facilities between 1996 and 2001. As was the case in the airlines and trucking industries two decades ago, a large number of new entrants have foundered on bad business plans and a disappointing market.

   Even the stocks of the Bell companies are only slightly above their 1996 levels, hardly a stunning result in the wake of the 1999-2001 surge in their capital expenditures. The wireless sector’s equity prices are about two-thirds of their 1996 values. The cable television companies, who are not included in my analysis, also appear to be relatively stable. Thus, the problems in telecom are heavily concentrated in the companies with large national fiber networks who offer long distance services, the new local entrants, who offer little new after investing more than $40 billion, and to a lesser extent, the wireless carriers. This is not to say that the incumbent local carriers operators are prospering in this environment, but the equity markets are suggesting that they are not in long-term difficulty.

   As a result of the collapse in market valuations, capital spending declined substantially in 2001-02, a decline that was exacerbated by an incredible series of bankruptcies of telecommunications carriers. Total capital spending has fallen from more than $100 billion in 2000 to less than $40 billion last year, according to most estimates, and it is forecast to remain low for at least the next year. This decline in capital spending is clearly the most alarming aspect of the current telecom malaise for it portends a slowdown in the deployment of new technology and even the possibility of a degradation of traditional services if it continues. Capital spending is now less than it was when the act was passed seven years ago.

   The capital spending boom is now widely acknowledged to have created excess capacity in data and voice transmission,[7] but the rise in investment spread far beyond fiber-optic transmission facilities. Capital spending by the new local carriers increased from virtually nothing to nearly $20 billion in 2000.[8] The wireless sector increased its capital outlays from $8.5 billion in 1996 to $18.4 billion in 2000.[9] And the regional Bell Companies (including GTE) increased their wireline capital spending from $20.8 billion in 1996 to $35.7 billion in 2000 even though they were largely banned from interstate communications.[10] All of these companies have pulled back substantially since 2000. 

Telecom Revenues 

   Perhaps the most surprising feature of the telecom industry since 1996 has been the absence of growth in carrier revenues despite the explosion of the Internet and the strong growth of the economy. Between 1987 and 1995, the growth in telecom output (its contribution to the “gross domestic product” of the entire economy) and capital spending growth were virtually identical.[11] But after 1995, the industry’s output did not accelerate very much in nominal dollars, rising from a 4.6 percent growth rate during 1987-95 to just a 6 percent growth rate after 1995. Figure 3 provides annual data from the Commerce Department on investment and output, as measured by gross product, for the telephone and telegraph industry. Given that current-dollar GDP grew at a 5.8 percent average annual rate during this period, the growth in telecom spending was surprisingly low and far below the growth in nominal capital spending. Indeed, the value of telecommunications output in nominal dollars has grown more slowly than has durable-goods manufacturing in recent years.[12]

   Source: BEA

   Most of the growth in telecommunications services output and revenues in recent years has come from the wireless sector. The local exchange companies, long distance carriers, and wholesale fiber-optic transmission companies, have seen little or no growth in their total revenues. (See Figure 4.) Between 1996 and 2001 (2002 data are not available yet), end-user wireline revenues increased by less than $8 billion, from $159.4 billion to $167 billion, or only 0.9 percent per year. In real, inflation-adjusted terms, wireline revenues actually decreased by about 1 percent per year.[13] This was surely not the explosive growth that had been anticipated from the IT revolution and “deregulation.”

   Source: FCC.

   The lack of revenue growth does not mean that there was no real output growth in telecommunications, but that the output growth was not great enough to more than offset the substantial decline in prices that was occurring. Prices of transmitting the trillions of bits of information generated by the Internet fell dramatically, but the demand response to these price declines was not sufficient to boost revenues. Real output growth was substantial, but not as spectacular as many have suggested. Figure 5 shows the recent trend in the telecom industry’s contribution to gross domestic product in real terms along with its contribution to gross domestic product in nominal dollars that was shown previously in Figure 3.

   Source: BEA.

   Regulatory Changes

   Though the 1996 Act was not deregulatory, telecom regulation was changing during the period preceding that Act and in the first few years of implementing it. Between 1993 and 2000, the following regulatory changes combined to make the telecom sector a treacherous environment for investment when combined with the technological changes also buffeting the industry:

1.   Wireless communications were opened to competition for the first time in 1993, and wireless rates were largely deregulated;

2.   Government auctions were initiated in 1995 to provide the requisite spectrum for new wireless competitors;

3.   AT&T’s interstate long distance rates were deregulated by the FCC in 1995;

4.   Local telecommunications services for residential and small business customers were opened to competition by the Telecommunications Act of 1996;

5.   Local incumbents were required by the 1996 Act to lease network facilities to new entrants at regulated prices; but the extent of the required unbundling and “line sharing” remained uncertain due to court reversals of FCC decisions;

6.    In 1997, the FCC launched a sweeping new program to reduce international “accounting rates” and to increase competition in international services.

7.   In 1999-2000, the “UNE Platform” began to replace other forms of competitive local entry as AT&T, WorldCom, and others responded to the deep discounts that states offered for leasing (essentially reselling) the incumbent’ networks.

All of these changes occurred as equity and debt capital poured into a large number of telecom companies, funding hastily constructed investment plans. In this environment, one had to expect a large number of failures, and this expectation was surely fulfilled. A tabulation of recent telecom bankruptcies found that 59 firms have filed for bankruptcy protection in the last few years.[14] 

Why Was Growth So Slow? 

   The lack of revenue growth for wire-based telecom carriers after 1996 can be attributed to perhaps three factors. First, competition in wireless and long distance services began to drive prices down substantially. Because the demand for traditional telecommunications services is price inelastic, these price declines translate into lower subscriber expenditures unless there are important new uses of these services. Second, the revolution in wireless communications has siphoned enormous amounts of traffic from the wireline network, particularly the over-priced long distance traffic. Third, new services, such as broadband, were slow to develop, in part because of regulatory uncertainty.  

   Household Spending.

Approximately 60 percent of all end-user telephone expenditures are made by households. Census data show that the share of household expenditures devoted to telephone service remained remarkably constant throughout the1980s at about 2 percent of their overall expenditures.[15] Beginning in 1993, however, this share rose gradually to 2.3 percent. In nominal dollars, the average household spent $877 in 2000 compared to $658 in 1993.[16] However, household data collected by TNS and reported by the FCC, reproduced in Table 1, show that all of this increase reflected a growth in wireless spending, not expenditures on traditional wireline services. Rising expenditures on local service, reflecting principally the increase in FCC mandated subscriber line charges, could not offset the decline in long distance spending.

   Why have these revenues fallen in an era of explosive growth of the Internet? The reason must be that revenues from the new uses of the telephone network have not offset the decline in revenues from falling long-distance rates. Given that the residential demand for local access has an estimated price elasticity of less than -0.05, any increase in local rates would lead to higher expenditures on local services. The modest increases in local rates that occurred after 1995, reflecting principally the increase in FCC mandated subscriber line charges as a substitute for per-minute carrier charges, could not offset the decline in long distance spending as Table 1 shows. This decline in spending is much greater than can be accounted for by lower rates; it obviously reflects a substantial shift to wireless services. Competition is truly working.

Table 1

Average Annual Household Expenditures on Telephone Service

   ($/Year) 

   Source: FCC.

   The Growth of Wireless.

The 1996 Act provided major policy changes towards the wireline telecommunications sector, but it largely ignored the wireless sector. Competition in the delivery of mobile wireless services had been limited to two carriers per local market by FCC policy since the 1970s, but liberalization was thrust on this sector by Congress three years before the 1996 Act was passed. The 1993 Omnibus Budget Reconciliation Act instructed the FCC to begin auctioning spectrum for commercial wireless uses. These auctions began in 1995, and construction of the new digital “PCS” networks was just beginning when the 1996 Act was passed.

   The 1993 legislation that established the spectrum auctions essentially eliminated price regulation of wireless services. States may now regulate these rates only if the carriers have “market dominance,” an unlikely condition in today’s wireless sector, even in rural areas. The auctions of 120 MHz of spectrum essentially allowed four new entrants into wireless services in each local market to compete with the two carriers that were already operating there. Since 1996, the wireless industry has been transformed by mergers and consolidations into an industry with six, large national carriers

   Although the U.S. launched its “second-generation” digital wireless service somewhat after Europe and Japan, wireless is now growing very rapidly. (See Figure 6) By mid-2002, the number of wireless subscribers had risen to almost 135 million; by 2004 or 2005 the number of wireless subscribers is likely to exceed the number of fixed access lines.[17] The substitution of wireless for traditional wire-based telephony is developing very rapidly. Many households, particularly those with young adults, do not even have a traditional copper-wire telephone service. Others are using their wireless service rather than their home telephone for long distance calls. The effect on traditional wire-based telephone carriers is obvious. Long distance revenues for wire-based carriers are now declining rapidly, and the number of fixed access lines is now also falling after decades of steady growth. 

   The potential stumbling block for wireless operators is the technology required to provide higher-speed Internet access. In Europe and other parts of the world, carriers have paid billions of dollars in auctions for spectrum designated for “Third Generation” wireless services (3G). But these services are likely to be much slower and less easy to use than fixed-wire broadband services, such as cable modems and DSL, or even “WiFi” services. 

   Source: FCC

   Regulatory Uncertainty.

In most other sectors, the liberalization of entry has been accompanied by, or at least followed by, rate deregulation. This is clearly not the case in telecommunications. The profound changes that have gripped, if not overwhelmed the telecommunications sector have not changed the traditional regulatory policy towards retail rates very much. State regulatory commissions continue to regulate local retail rates very much as they did in 1996 despite the growth in wireless, the steady expansion of competitive local carriers, and the lurking threat of cable telephony.

   The 1996 Act also ushered in a major new form of regulation: network unbundling at cost-based rates. After seven years of intense intra-industry battles, the FCC has maintained a rather steady course of intense wholesale regulation and has not placed any pressure on the states to deregulate retail rates. In fact, unbundling and line-sharing requirements have actually increased with the passage of time. And new services, such as broadband services, have been subject to asymmetric regulation since 1996.

   Fortunately, there are numerous proceedings open at the FCC that could alter the regulatory landscape substantially.

    a. Network Unbundling

   The FCC’s approach to wholesale unbundling has been controversial from the outset, in part because of the vague language in the 1996 Act. Facilities are to be unbundled and made available to entrants if without them the entrants would be “impaired” in competing with the incumbents.[18] But what is the measure of “impairment?” Moreover, should a facility be unbundled everywhere if it is determined that the entrants would be “impaired” somewhere without it? The FCC on two separate occasions essentially decided that virtually everything in the incumbents’ networks must be unbundled and that there should be no differences between rural and urban areas or across states. This broad approach to unbundling is unique to the United States and has been the source of controversy for more than six years.

   The most recent challenge to this broad-based approach to unbundling was mounted by the incumbent telephone carriers in 2001 in the form of a petition to the U.S. Court of Appeals for the District of Columbia. The court’s opinion in May 2002 requires the FCC to reconsider its unbundling requirements and, in particular, its requirement for line sharing with entrants seeking to offer DSL services, but not basic telephone service.[19] The court criticized the Commission for failing to account for the effect of competition in determining whether the absence of an unbundled element would “impair” the ability of entrants to compete. As of this writing,  the FCC has not responded to this court order.

   The Commission now has the opportunity to back off from its broad attempt to require everything to be unbundled and to bring U.S. policy to a less interventionist level that is consistent with that employed elsewhere in the world. First, if it decides that certain switching and transport functions need not be unbundled, it would bring an end to the use of the “UNE Platform” that now accounts for one third of all entrants’ lines. This would force entrants to build facilities or to scale back their local services where they are essentially reselling the incumbents’ services. Second, the Commission can end its unsuccessful attempt to force intra-platform competition in broadband through line-sharing requirements. Both decisions would lead the Commission in a deregulatory direction and reverse six years of increasing regulation of inter-carrier relationships.

   b. Broadband

   Economic regulation is generally premised on the existence of market failure due to monopoly. In telecommunications, competition is increasing rapidly in most markets. However, one cannot claim that the delivery of the new broadband Internet services are or are even likely to be plagued by problems of monopoly. Cable television systems compete actively with telephone-company DSL services, and a variety of wireless and satellite services are under development. Because these services have developed over facilities that were originally designed to carry other communications services, they have been subject to the threat or actuality of regulation under different provisions of the Communications Act. Recently, however, the FCC has decided that cable modem service is an “interstate information service,” and therefore subject to the FCC’s jurisdiction. It is also contemplating the appropriate regulatory approach to all wireline broadband services, including DSL.[20]

   Through a welter of different proceedings, the Commission has the opportunity to exercise “forbearance” from regulating any of these services under Section 706 of the Act. Unfortunately, it has been examining these options for a very long time without reaching any decision on whether to regulate cable modem service or to forbear from regulating any of these “advanced” services, allowing the market to drive technology, facilities deployment, and pricing of the services. Once again, the Commission has the opportunity to move in a deregulatory direction in a market that is evolving rapidly and has no clear tendency towards monopoly.

Policy Options for Recovery

   In a sense, the telecommunications industry is suffering from the effects of successful competition. The cost of accessing and using traditional telecommunications services is declining rapidly, led by the aggressive competition among six national wireless companies. Further competition will place more downward pressure on the traditional local and long distance rates, particularly as the fiber-based local carriers drive down business rates in central business districts. Wireless rates will continue to fall. All of this is beneficial to consumers and the economy, but it does not provide the resources for growth and expansion of the network. These forces will further reduce telecom revenues. As wireless continues to replace wireline services, the incentives to invest in the traditional telephone network will be further reduced.

   If we are to see a revival of capital spending, it must be stimulated by the development and deployment of new services that expand telecom revenues. But the incentive to develop these new services is clearly impeded by the continuing uncertainty over attempts to regulate the wholesale and retail access to these services. Wireless companies, local incumbent carriers, and cable companies cannot and will not underwrite large capital expenditures to develop new services if they must share the gains with rivals or be subjected to rate regulation in selling these services to customers. For this reason, the FCC should make it clear that traditional telephone companies and cable television companies will not face regulation of their new high-speed, broadband Internet services.

   Decisions regarding unbundling of network facilities for the delivery of traditional telephone service are also important, but not as important as the removal of regulation from the newer services. The spread of the UNE platform will increase the appearance of competition, but not the reality of it. Simply allowing other carriers to deliver the same service over the same facilities to the same customers at a greater social cost will not promote competition. The UNE platform is not stimulating the development of new local services. Nor are the companies offering local service over the UNE platform using this network strategy to gain a toe-hold before moving ahead to build their own networks. Indeed, I believe that the securities markets are already telling us that those using the UNE platform are not likely to thrive from such a strategy. For a while, the increase in transactions costs, the bickering over wholesale rates, and the uncertainty over the UNE-P’s effect on the incumbents’ cash flows will simply displace more productive uses of these resources. But eventually, I believe, the UNE platform will die because it is not an economically viable method of organizing a network industry in which there is so much technical change.

   The United States was not alone in experiencing a telecommunications “bubble.”  Virtually every other developed country suffered a similar boom-bust cycle. The exaggerated expectations for telecom created by the Internet and general IT revolution were met with stark reality in Europe, Asia, and Oceania at about the same time. Stock market valuations in many of these countries also soared in 1998-2000 only to collapse in 2000-02. Capital spending in telecommunications collapsed everywhere, placing most telecom equipment suppliers in severe difficulty. The only way out of these problems is to allow investors to find and fund productive new uses of telecommunications. Otherwise, declining prices will translate into declining revenues and little appetite for capital spending.

Related Documents

Printer Friendly

Comment On This Page

Related Documents

Document Menu

Hearing Webcast

Invited Witnesses

Member Statements

Printed Hearing Record
(transcript)