SECTORAL RISK AND RETURN EFFECTS OF THE DEREGULATION OF THE U.S. TELECOMMUNICATIONS INDUSTRY: IS GROWING COMPETITION INCREASING INVESTMENT RISK? by Randall S. Billingsley, Ph.D., CFA and Doug Eckel, Ph.D. Department of Finance Center for Wireless Telecommunications Virginia Polytechnic Institute & State University Blacksburg, VA 24061-0221 U.S.A. (540) 231- 7374 Phone (540) 231-4551 FAX e-mail: RANDALL@vt.edu To be Presented at the Economics of Telecommunications and Media Sectors Session of the 2nd International Conference of the Society on Computing in Economics and Finance Society of Computational Economics Geneva , Switzerland, 26-28 June 1996 The authors thank Eric Blazer for computing assistance and Douglas Wilburne, CFA of Bell Atlantic and William Keck and Douglas Schaller of BellSouth Telecommunications for helpful discussions of competitive and regulatory developments in the telecommunications industry. The first author gratefully acknowledges financial support from Virginia Tech through the Center for Wireless Telecommunications and the Virginia Tech Foundation for an international conference travel grant awarded by Research and Graduate Studies (ICTSG Program). The authors are solely responsible for the given results, their interpretation, and any remaining errors. SECTORAL RISK AND RETURN EFFECTS OF THE DEREGULATION OF THE U.S. TELECOMMUNICATIONS INDUSTRY: Is Growing Competition Increasing Investment Risk? The death of distance as a determinant of the cost of communications will probably be the single most important economic force shaping society in the first half of the next century. It will alter, in ways that are only dimly imaginable, decisions about where people live and work; concepts of national borders; patterns of international trade. ... There was a time when telecoms seemed to be a natural monopoly. ... But as new technologies reduce the costs of entry, competition is spreading....The direction of these changes is fairly clear, but their speed is uncertain. ... [T]he pace will be set not just by technology but by the interplay of regulation and competition. (The Economist, Telecommunications Survey, September 30, 1995, pp. 5-6). 1. Introduction In this study we examine the evidence to date for relative changes in the profitability and riskiness of the competing sectors in the telecommunications industry. The Telecommunications Act of 1996, combined with the rapid evolution in new telecommunications technology, will bring significant changes to the competitive structure of the telecommunications industry. Although the Act itself ostensibly provides for an "orderly transition from regulated to competitive and deregulated telecommunication markets consistent with the public interest, convenience and necessity," technological developments alone are increasing the threat of entry by new competitors, the rate of development of substitute telecommunications products and services, and the ability of existing competitors to challenge rivals across sectors. These changes, along with enhanced regulatory risk at both the state and the federal levels, should have increased investor uncertainty in the industry as a whole. However, security analysts, academics, and government regulators have for years anticipated that the Act's new regulatory regime and more competitive markets will create both winners and losers among the different sectors of the industry. Ownership of local area networks, competitive marketing expertise, and access to new technology or informational/entertainment content, for example, will all provide competitive advantages, even under completely deregulated markets. A comprehensive empirical study that systematically measures the nature, extent, and significance of any performance and/or risk changes induced in U.S. telecommunications firms due to recent deregulation and competitive changes is timely. Although the economic efficiency motivation for deregulation of the telecommunications industry is clear, increases in the riskiness of sectors of the industry are of interest to shareholders and potential investors. These changes provide the rationale for seeking certain regulatory policy actions on behalf of the individual sectors, e.g., greater allowance for business combinations or relaxation of anti-trust concerns, or relief from the phasing in of deregulatory measures (such as cross-sectoral fees for local area network access). Our goal is to determine whether the investment community views deregulation as a net benefit or cost for each of the individual segments. We provide evidence on the changing investment characteristics of telecommunications firms useful to equity investors and to telecommunications policy analysts evaluating the competitive effects of deregulation.To do this we perform time-series analyses of key measures of business and investment risk in recent years for prominent U.S. and non-U.S. telecommunications firms. Specifically, we evaluate the operating performance of telecommunications firms using balance sheet, income statement, and cash flow data in order to measure the effects, if any, of the increased competition presumed to accompany deregulation. We also use measures of systematic or market-related and firm-specific risk in our analysis. Finally, we examine the impact of key competitive telecommunications industry events in recent years from the perspective of the equity investors in the different sectoral portfolios. The stock return reactions of the portfolios reveal any net changes in the risk perceptions of telecommunications firms' investors in recent years. We first divide firms active in the industry during the 1987-1994 timeframe into the following segments: long-distance providers, the so-called "Baby Bells" and GTE, the independent (i.e., non-"Baby Bell" or GTE) providers of local exchange services, wireless telecommunications providers, cable (CATV) services, and various non-U.S. telecommunications services providers. Although our results are preliminary, to summarize briefly, we find that although profitability (return on equity, ROE) has fallen for the Regional Bell Holding Companies and GTE (RBHC/GTE) between the 1987-90 and 1991-94 periods, market value (price-to-book and price-to-sales) indicators, an increase in free cash flow per share, and a decrease in the RBHC/GTE sector's market risk (beta) and financial risk (interest burden), suggest that the RBHC/GTE segment of the industry may be well-positioned competitively. Only firms in the non-U.S. group show similar decreases in interest burden and increases in market valuation indices. Independent firms also had declining ROE, and less negative free cash flow per share, but did not experience a favorable change in their equity market valuation. Long-distance firms did not experience any statistically significant changes in their financial position over the period studied. Wireless sector firms had a significant increase in capital expenditures accompanied by a decrease in research and development (R&D) expenditures. Firms in the cable segment experienced an increase in total asset turnover, and an increase in interest burden. In summary, the numbers do not suggest major, systematic differences for the sectors between the two periods. However, the changing environment seems to have affected the RBHC's and GTE and non-U.S. sectors' market prospects more favorably than the other sectors. When we analyze a preliminary set of events signaling changes in the industry's competitive environment, we find significant portfolio reactions to the MCI/British Telecom announcement to jointly build a trans-Atlantic cable, to the Bell Atlantic/TCI merger plans, and to the Bell Atlantic/NYNEX merger announcement. The RBHC/GTE segment responded significantly negatively to the first event, while firms in the independent and wireless sectors responded significantly positively to this event. The cable sector reponds very positively (0.05) to the Bell Atlantic/TCI plan. The market interpreted this later event as beneficial for the RBHC/GTE sector firms. When we analyse the significance of absolute differences between the market responses of different sector firms to the set of announcements, we find that the MCI/British Telecom trans-Atlantic cable announcement produced four significant pairwise comparisons. Both the long distance and RBHC/GTE sectors respond significantly more negatively than the wireless and independent sectors. The cable sector responds significantly more positively than all the other sectors on the date of the Bell Atlantic/TCI merger. Section 2 below reviews the regulatory developments over the 1987-1994 period from the perspective of the different sectors. Section 3 reviews our data sources and analytical methodology. Section 4 presents our results, and Section 5 our conclusions, and suggestions for future work. 2. Recent Competitive and Deregulatory Trends in the U. S. Telecommunications Industry 2.1. Competition from the Local Exchange Telephone Company Perspective The primary providers of local telephone services in the U.S. are the seven regional Bell holding companies (RBHCs), GTE, and the so-called independent companies that do not fall into the prior categories. As such, they are commonly referred to as the local exchange companies (LECs). The economic "bottleneck" position previously occupied by the LECs has been eroded through the growing ability to at least partially bypass their services and through the progressive development of substitute services like wireless technology. For example, large businesses are able to bypass the LECs' private line and access services using fiber optic networks, microwave transmission, and very small aperture terminals (VSATs). The growth of competitive access providers (CAPs) such as Metropolitan Fiber Systems (MFS) and the Teleport Communications Group (TCG) has allowed large business customers in major U.S. cities to connect with long distance carriers (inter-exchange carriers, IXCs) without paying an access charge to a LEC. Similarly, while the current expense of cellular technology makes it more of a complement or a substitute for traditional telephone service, the cost of this technology is likely to decline over time and thereby make it a viable substitute for such services. Technological advances in conjunction with broad-based deregulation have exposed the LEC business to much greater competition and risk. Standard & Poor's provides a representative summary of the dominant view of the investment community concerning competition in the industry (1993, page T-19): Competition has been the driving force in the industry in the recent past. Competition, competition and more competition will be the driving force in the future. Technology and regulatory initiatives have been chipping at old structural barriers, building on earlier moves to create a truly competitive marketplace. Similarly, in a more recent evaluation Standard & Poor's discusses the benefits of competition in a global context (1994, page T-15): Growing competition will benefit telecom users. New service offerings will expand the size of the telecommunications pie globally and will attract new players. Established companies may see some segments of their business stolen away, while other segments will grow in importance. These are the risks and rewards of new opportunities in today's rapidly evolving telecommunications marketplace (emphasis added). "Established companies" like the RBHCs and GTE are expected to be hurt by the continually declining interconnection/access rates associated with the more competitive, deregulated environment that will put pressure on margins and overall revenues. 2.2. Competition from a Cable Company Perspective The cable industry is poised to compete effectively with the LECs because its existing networks provide direct links to customers comparable to those of the LECs. Many of the major CATV providers have indicated their intent to provide essential residential telephony services. Indeed, their coaxial cable in place has much greater capacity than the copper wire that predominates the LECs' networks. Yet the LECs have two-way signal or switching capability while the cable industry does not. Thus, in order to compete effectively with the LECs, cable companies must upgrade their networks at considerable expense. These economic realities have prompted a number of cable and telephone companies to form joint ventures in the U.S. and abroad. Consider the scope of recently reported alliances: Sprint/TCI/Cox Cable/Comcast, Intermedia Partners/Viacom Cable, US West/Time Warner, Southwestern Bell (SBC)/Hauser, NYNEX/Viacom Cable, BCI/Jones, US West/Wometco, and Airtouch Communications/Bell Atlantic/NYNEX/US West, to name a few. A common strategic goal in these alliances is to develop the capacity to offer both telephone and video services over coaxial cable and fiber optic networks. Further, CATV companies are pursuing strategic alliances with LECs, IXCs, and CAPs that will create networks for the distribution of multimedia products on a broad geographic scale. 2.3. Competition from the Competitive Access Provider Perspective Historically, the LECs have had the responsibility to provide affordable, universal telephone service. This created a social pricing scheme that left business and long distance services priced uneconomically. Specifically, long distance and business telephone service prices were placed above their true costs in order to subsidize the pricing of local residential telephone service. The AT&T divestiture in 1984 addressed some of these pricing issues. However, the fact that the LECs are required under their regulatory contract to foster the achievement of universal telephone service has left them vulnerable to being underpriced by unregulated (or less regulated) competitors providing local exchange services. CAPs have grown because they can price their services more aggressively than the LECs since they do not have the universal service responsibility shouldered by the LECs. Recent decisions by the Federal Communications Commission (FCC) have reduced the remaining barriers to entry in the provision of competitive access services. The FCC introduced more competition into the LECs' environment by granting CAPs the right of expanded interconnection with the LECs in order to provide special and switched access services. This is of enormous economic significance since it allows CAPs not only to compete with the LECs for large business customers but for almost any other customer as well. The fees associated with the expected proliferation of CAPs are not expected to offset the expected loss in access charges (see telecommunications section, Standard & Poor's Industry Surveys, 1993, page T-20). Consequently, the increased competition in the local exchange market is expected to erode the LECs' revenue base even further. 2.4. Competition from the Inter-Exchange (Long Distance) Carrier Perspective AT&T and MCI are positioning themselves to provide a full array of telecommunications services, which include local exchange services. For example, AT&T has indicated that its merger with McCaw Cellular Communications will facilitate the development of a national and international wireless network. The integration of McCaw's cellular network with AT&T's long distance network will allow AT&T to provide both local and access services. MCI's purchase of Western Union's ATS division and a $2 billion investment in MCI Metro evidences a desire to provide comprehensive telecommunications services that will compete with LECs. Indeed, in late 1993, MCI announced plans to build competitive access networks and install switches in about 20 major U.S. metropolitan areas by the end of 1996. British Telecomm's purchase of a 20% stake in MCI for $4.3 billion provides MCI with the resources to complete these plans. 2.5. Competition from the Wireless Services Perspective Wireless services, first in the form of cellular telephone services and now moving toward personal communications systems (PCS), promise to be a competitive substitute for traditional telephone services in the future. While the growth of cellular telephone services has been phenomenal, it is currently only viewed as a complement to traditional services and commonly makes use of LEC networks. PCS technology is cheaper to apply than that of cellular and thus PCS services are expected to be offered to customers at cheaper prices and with more network capacity than cellular services. On the basis of PCS license applications to date, CAPs and CATV operations are expected to be major competitors in the provision of PCS services. IXCs such as AT&T and MCI are also actively seeking to develop PCS networks. The provision of wireless services such as PCS systems by CAPs, CATV operators, and electric utilities enhances the ability of customers to completely bypass the local exchange services of the LECs. Wireless services are becoming a viable consumer alternative to LEC services. These alternatives will only increase the competitiveness of that environment and thus magnify the business risk of LEC operations. 2.6. Competitive Implications of Recent Regulatory and Legislative Developments As discussed above, at the federal level the FCC has increased the degree of competition faced by LECs through its order that allows expanded interconnection with LEC networks. Further, the FCC has enhanced competition by its decision to allocate additional spectrum to PCS services. Finally, the recently passed Telecommunications Act of 1996 allows the telecommunications companies to get into one another's businesses. For example, the Act allows LECs to move into the CATV and long distance markets and allows long distance and CATV companies to provide local telephone services. Importantly, the LECs will not be permitted to provide long distance services until they can demonstrate the presence of significant competition in their local telephone service markets. Further, the Act deregulates cable rates. Thus, the Telecommunications Act of 1996 will greatly enhance the competitiveness and possibly the riskiness of the U.S. telecommunications industry. 3. Data and Methodology 3.1. Data Sources and Telecommunications Industry Segments The current study collects data on companies in the following segments of the telecommunications industry: 1) the regional Bell holding companies and GTE; 2) independent telecommunications (i.e., non-"Baby Bell" or GTE) companies; 3) long distance service providers; 4) wireless telecommunications companies; 5) cable and other pay television service providers, and 6) non-U.S. providers of long distance, local telephone, and/or cable telecommunications services. There is of course some overlap among the industry segments. For example, while the wireless segment defined in Table 3 has essentially stand-alone companies, the RBHCs clearly have wireless subsidiaries that operate within this industry segment as well. Further, in 1994 AT&T announced its planned acquisition of McCaw Cellular, thereby eliminating it as a stand-alone member of the defined industry segment. Notwithstanding these overlaps, the telecommunications industry segments defined in Table 3 are as distinct as the data permit. Analysis of accounting-based and cash flow-based performance measures is based on annual (year-end) data obtained from COMPUSTAT and beta coefficient and event study data are obtained from the New York Stock Exchange (NYSE)/American Stock Exchange (ASE) and National Association of Securities Dealers (NASD) daily returns tapes provided by the Center for Research in Security Prices (CRSP). The non-U.S. telecommunications companies all have American Depository Receipts (ADRs) for which price data are available. While radio and television broadcasting are important segments of the telecommunications industry, they are excluded from the current analysis in order to focus on the telephone and cable components. The chosen financial data from COMPUSTAT covers the period from 1987 to 1994 while the chosen event study data is from the period from November 11, 1989 to September 12, 1994. Table 2 indicates the composition of the telecommunications industry sector portfolios used in the analysis and Table 1 describes the SIC codes of the members of those portfolios. 3.2. Methodology and Testable Hypotheses 3.2.1. Time Series Analysis of Sectoral Performance: Period 1 vs. Period 2 The first step of the analysis evaluates the validity of the conventional wisdom that the progressive deregulation and increasing competitiveness of the telecommunications industry has significantly affected its operating performance and riskiness in recent years. This is done by measuring key measures of financial, cash flow, and market performance for each of the above-indicated segments of the telecommunications industry over the period from 1987 to 1994. In order to evaluate whether the changes in the competitive/regulatory environment have materially affected the telecommunications industry segments, these measures are systematically compared over two sub-periods: from 1987 to 1990 (Period 1) and from 1991 to 1994 (Period 2). Performance measures are estimated and evaluated within but not across telecommunications industry segments. For example, the present analysis determines whether the performance of the long distance segment changed significantly over the indicated time period but does not assess whether the long distance segment performed significantly differently than the wireless segment over that time period. The specific performance measures used in the analysis are defined in Table 3. Observed operating performance is evaluated under the following testable null hypothesis: H01: There is no significant difference in the given performance measure i for telecommunications industry sector j between Period 1 and Period 2. There is consequently no significant difference between the mean of the indicated variable Xi within sector j for Period 1 and the mean of the indicated variable Xi for Period 2 for the given industry sector j. Thus, the null hypothesis is that Xij1 = Xij2 within each given sector j. The resulting t-tests are adjusted in light of whether the evidence supports the homogeneity of variance assumption of the given mean comparison. In the event that the variances cannot be assumed to be equal on the basis of the appropriate F-test, Satterthwaite's (1946) approximation is used to compute the degrees of freedom and the approximate t-statistic is reported. Failure to reject H01 would imply that the changes in the competitive/regulatory environment were not associated with significant differences in the performance or riskiness of the telecommunications industry between the periods of 1987-90 and 1991-94, as measured on a sector-by-sector basis. 3.2.2. Analysis of Responses to Selected Competitive Events in the Telecommunications Industry Table 5 identifies and briefly explains the significance of the competitive events in the telecommunications industry from 1990 to 1994 selected for analysis. The event dates are the earlier of the first published announcement or the first mention of the event on the news wire, as reported by Dow Jones News/Retrieval. The current study determines whether the given competitive events were associated with significant excess returns for the portfolios of the telecommunications industry sectors identified in Table 2. The presence of returns that are inconsistent with a given sector's systematic risk suggests that the given event may have influenced shareholders' wealth. Similarly, the comparative analysis of shareholders' return reactions across the sectors of the telecommunications industry should reveal any wealth redistributions among those sectors that are consistent with the expected winners and losers resulting from the change in the competitive landscape. Since all of the firms considered are in the telecommunications industry, albeit in different though related sectors, they were all affected simultaneously, if at all, by the given competitive events. This implies that the segment portfolio stock returns are most likely cross-sectionally correlated. Following Binder (1985), Smirlock and Kaufold (1987), and Schipper and Thompson (1985), the seemingly unrelated regression (SUR) technique proposed by Zellner ( 1962) is used to correct the residual terms for the probable covariances across the telecommunications sector portfolios used in the study. The responses of the segment portfolios are evaluated using the following system of n equations estimated over the period starting 250 days before the first event noted in Table 5 (5 November 1990) and ending one trading day after the last event (9 September 1994). Each event is evaluated as a two-day window that includes the explicit event date noted in Table 5 and the prior trading day to cover any potential error in the identification of the event date: where: Rjt = the observed value-weighted return for sector portfolio j for day t; n = the number of telecommunications industry sector portfolios as described in Table 2; Rmt = the observed return on the market value-weighted (including dividends) market index on day t; Dtk = a dichotomous dummy variable equal to 1 on the event day and the prior trading day and 0 otherwise for event k, and îjt = the residual error term. The overall responses of each of the telecommunications industry sector portfolios are examined under two null hypotheses: H02: There is no response to the given competitive event in the telecommunications industry. The event dummy parameter ëij is consequently equal to zero for each company in each telecommunications industry sector portfolio. Thus, the null hypothesis is that ë1j = ë2j = ... = ënj = 0. H03: There is no difference in the reaction to the given event across the various sectors of the telecommunications industry. Thus, pairwise comparisons reveal no significant differences between the dummy coefficients ëij for the two given different telecommunications industry sectors. The null hypothesis is consequently that ë1 - ë2 = 0 for all of the pairwise comparisons of the average values of the dummy coefficients across the sectors of the telecommunications industry. Failure to reject H02 would provide evidence that the given competitive event did not materially influence stockholders expectations concerning the value of their investment. Failure to reject H03 would suggest that the various segments of the telecommunications industry did not react significantly differently to the given competitive event. This would suggest that stockholders did not distinguish among the specific telecommunications sector portfolios in reacting to the given event. 4. Empirical Results 4.1. Time Series Analysis of Sector Performance: Period 1 vs. Period 2 As noted above, Table 3 defines the various performance measures that are compared between 1987-90 (Period 1) and 1991-94 (Period 2) for each of the telecommunications sector portfolios identified in Table 2. Table 4 shows the empirical results of these time-series comparisons and highlights statistically significant mean differences in bold. The RBHC/GTE segment portfolio shows a lower average ROE, lower interest burden, and lower average beta coefficient in Period 2 than in Period 1. The lower ROE could reflect greater competitive pressures in the later period while the lower interest burden suggests that interest charges actually fell per dollar of total asset, which has no clear competitive implication other than that financial structure risk moderated in the presence of no significant change in the amount of debt relative to equity (TACE). The lower average beta of 0 .7272 in Period 2 indicates that systematic risk declined for the portfolio, which is counter to the expected impact of greater competition. However, it should be remembered that this is the net change in the systematic risk of the RBHC/GTE portfolio, which could reflect significant offsetting changes in the systematic risks of the portfolio's components. Thus, the distribution of beta changes bears additional scrutiny. Free cash flow per share (FCFSHR) became significantly less negative. This indicates that the segment's cash flows improved but that it remained quite aggressive in Period 2 as it financed growth by relying significantly on external sources of funds. Consistent with its negative FCFSHR, total equity cash flows to shareholders relative to net income (ECFNI) declined over the two periods. Finally, the sector shows higher price-to-book (PBOOK) and price-to-sales (PSALE) ratios in the second period. The performance measures do not strongly support the contention that the competitive/regulatory environment dramatically altered the performance of the RBHC/GTE portfolio over the two indicated time periods. However, the sector's market risk decreased and market value increased more than all other sectors except the non-U.S. sector. The independent telecommunications companies industry segment also experienced a significant decline in ROE over the two time periods and its FCFSHR also became less negative. Interestingly, its capital expenditures per dollar of free cash flow became less negative as well. Perhaps this reflects the group's inability to sustain the same proportionate level of capital expenditures in light of declining profitability. Consistent with the RBHC/GTE portfolio, total cash flows to equity holders relative to net income also declined significantly. In contrast to the case for the RBHC/GTE portfolio, the favorable equity market valuations of telecommunications stocks generally in Period 2 are not reflected in significantly higher PSALE and PBOOK variables for the independents. Further, there is no significant difference in the systematic risk of the independents across the two time periods. Interestingly, the long distance segment portfolio does not exhibit any significant changes in its performance measures over the two time periods. Perhaps this is because this segment was confronted with significant competition soon after the AT&T break-up in 1984 and had adapted to a generally competitive environment prior to the periods studied in the current analysis. However, as the subsequent analysis of the long distance segment's reactions to key competitive events indicates, shareholders' expectations concerning the prospects of the segment clearly varied during the periods of analysis. The wireless telecommunications industry segment shows no significant performance changes other than a significant decline in research and development expenses relative to sales (R&DS) and a more negative amount of capital expenditures relative to free cash flow. This could reflect a reallocation of R&D spending toward the acquisition and/or development of additional capital assets associated with the rapid growth in cellular service sales over these time periods. The cable segment portfolio only experienced a significant increase in its average total asset turnover (TATO) and an increase in its interest burden (IBURDEN). This is consistent with more efficient use of cable assets (plant) to generate sales and with the presence of generally high leverage and its associated interest costs. The non-U.S. portfolio of telecommunications firms exhibited a significant decrease in ROE, interest burden, financial leverage (TACE), and capital expenditures relative to free cash flow. The portfolio also reflected a significantly higher PBOOK variable in the later period. In summary, analysis of the performance measures of the various telecommunications industry sectors does not generally support the hypothesis that increased competition in the industry has dramatically affected performance or increased shareholder risk. The only consistent finding is a significant decline in profitability, as measured by ROE, for the RBHC/GTE, independents, and non-U.S. company segment portfolios. 4.2. Analysis of Responses to Selected Competitive Events in the Telecommunications Industry 4.2.1. Overall Stock Return Reactions to Selected Competitive Events Table 5 identifies and briefly explains the significance of seven competitive events in the telecommunications industry that occurred between 1990 and 1994. Tables 6 and 7 present the average stock return reaction of each of the industry segment portfolios to each of the events and tests whether the segment portfolios reacted significantly differently in a series of pairwise comparisons, respectively. There were significant stock return reactions to the MCI/British Telecom announced intent to build a trans-Atlantic cable in 1990, the announcement of the Bell Atlantic/TCI combination, and the announcement of the Bell Atlantic/NYNEX cellular consolidation plan in 1994. The RBHC/GTE segment portfolio reacted significantly negatively to the MCI/British Telecom announced intent to build a trans-Atlantic cable. This is consistent with shareholders' concern that this cable would potentially make it harder for the RBHCs and GTE to compete in the overseas market. This was relevant since the RBHCs and GTE did not face the same constraints on the provision of long distance services outside as inside the U.S. and also because they anticipated the ability to enter into more extensive long distance operations as a result of future deregulation. Interestingly, the independent segment portfolio reacted positively to the MCI/British Telecom event, perhaps under the assumption that the negative effect of the event on its RBHC/GTE rivals would be a net benefit to their sector. The long distance, cable, and non-U.S. company segments did not react significantly to this event, however, the wireless segment reacted positively. A plausible rationale for this reaction is not immediately apparent. As expected, the cable segment responded very positively to the announced Bell Atlantic/TCI merger. The RBHC/GTE segment portfolio reacted significantly positively to the announcement of the Bell Atlantic/NYNEX cellular operations consolidation plan in 1994. This is no doubt due to the perceived benefits of the consolidation of geographically contiguous cellular operations for both Bell Atlantic and NYNEX. While the positive reactions of these two prominent members of the segment probably overwhelm any non-positive reactions of the other members, the net positive portfolio reaction also likely results from the positive signal to the market concerning the feasibility and desirability of other such combinations. Interestingly, the independent segment portfolio exhibited a marginally significant negative reaction to Pacific Telesis' decision to spin-off its wireless operation as AirTouch Communications. This presumably reflects the concern that the independents could face an increased number of competitors emanating from their strong RBHC and GTE rivals. 4.2.2. Tests for Different Reactions to Selected Events Across the Industry Sectors Table 7 presents the results of the pairwise comparisons of industry sector portfolio reactions to the given events that were significantly different. There were two events for which the some of the industry sector portfolios reacted differently: 1) the MCI/British Telecom announcement of their plan to build a trans-Atlantic cable in 1990 and 2) the Bell Atlantic/TCI announcement of their intended merger in 1993. When we analyze the significance of absolute differences between the market responses of different sector firms to the set of announcements, we find that the MCI/British Telecom trans-Atlantic cable announcement produced four significant pairwise comparisons. Both the long distance and RBHC/GTE sectors respond significantly more negatively than the wireless and independent sectors. The cable sector responds significantly more positively than all the other sectors on the date of the Bell Atlantic/TCI merger. 5. Summary and Conclusions Based on this preliminary analysis of a selected set of financial ratios and events during a period over which competitive pressures increased in the telecommunications industry, it is difficult to pinpoint any broad systematic changes in the performance or riskiness of the various sector portfolios. Thus far we are only able to conclude that the changing competitive envirnonment appears to have affected the RBHCs and GTE more favorably than the other sectors. Further, we find evidence of different sectoral reactions to selected competitive events in the telecommunications industry in the early 1990s. Additional work will attempt to expand the analysis to events closer to passage of the Act in 1996, and develop additional measures of portfolio risk for the comparative analysis. We then hope to develop a cross-sectional analysis to help explain the reasons for variation between the sectors in the response to deregulation. REFERENCES Binder, J., 1985, Measuring the effects of regulation with stock price data, Rand Journal of Economics 16, 2, 167-183. Brown, S.J. and J.B. Warner, 1980, Measuring security price performance, Journal of Financial Economics 8, 205-258. The Economist, Telecommunications survey, September 30, 1995, survey section 1-27. Judge, G., R. Hill, W. Griffiths, H. Lutkepohl and T. Lee, 1982, Introduction to the theory and practice of econometrics (John Wiley & Sons, New York, NY). Satterthwaite, F.E., 1946, An approximate distribution of estimates of variance components, Biometrics Bulletin 2, 110-114. Schipper, K. and R. Thompson, 1985, The impact of merger-related regulations using exact distributions of test statistics, Journal of Accounting Research 23, Spring, 408-415. Smirlock, M. and H. Kaufold, 1987, Bank foreign lending, mandatory disclosure rules, and the reaction of bank stock prices to the Mexican debt crisis, Journal of Business 60, July, 347-364. Standard & Poor's Corporation, April 13, 1993, Telecommunications, Basic analysis, Standard & Poor's Industry Survey. Standard & Poor's Corporation, June 2, 1994, Telecommunications, Basic analysis, Standard & Poor's Industry Survey. Zellner, A., 1962, An efficient method of estimating seemingly unrelated regressions and tests for aggregation bias, Journal of the American Statistical Association 57, 348-368. Table 1 Telecomunications Industry Sector SIC Code Descriptions SIC Code Description 4812 Radiotelephone Communications (Wireless) 4813 Phone Communications Ex Radiotelephone 4841 Cable and Other Pay TV Services Table 2 Telecommunication Industry Sector Portfolios Panel A: Regional Bell Holding Companies and GTE: SIC = 4813 Company Ameritech Corp Bell Atlantic Corp BellSouth Corp GTE Corp Nynex Corp Pacific Telesis Group Southwestern Bell Corp U S West Inc Panel B: Independent Telecommunications Companies: SIC = 4813 Company ALC Communications Inc. AllTel Corp. BCE Inc. Century Telephone Enterprises Cincinnati Bell Inc. Intelcom Group Inc. Rochester Telephone Co Southern New England Telecommunications Telephone & Data Panel C: Long Distance Services: SIC = 4813 Company AT&T Corp. MCI Communications Sprint Corp. Panel D: Wireless Telecommunications Services: SIC = 4812 Company Airtouch Communications Inc. American Paging Inc. Lin Broadcasting McCaw Cellular Comm.-CL A Page America Group US Cellular Corp Panel E: Cable and Other Pay TV Services: SIC = 4841 Company Cablevision Systems -CL A Century Comm.-CL A Comcast Corp. -CL A SPL Falcon Cable Systems -LP Galaxy Cablevision -LP Jones Intercable -LP-CL A Spectravision Inc. Tele-Communications-CL A Panel F: Non-U.S. Telecommunications: SIC = 4812, 4813, & 4832 (ADRs) Company British Telecom PLC Cable & Wireless PLC Comp De Tele De Chile Hong Kong Telecom Ltd. Philippine Long Distance Tel. Telecom CP New Zealand Telefonica De Argentina Telefonica De Espana Telefonos De Mexico Vodafone Group PLC Table 3 Performance Variable Definitions Panel A: Accounting-Based Performance Measures Variable Definition/Interpretation ROE Net Income/Common Equity Return from common equityholders' perspective. PMARGIN (Earnings Before Interest and Taxes)/Sales Profit Margin Effectiveness of transforming sales into operating profit. TATO Sales/Total Assets Total Asset Turnover Effectiveness of using assets to generate sales. ROA (Earnings Before Interest and Taxes)/Total Assets Operating Return on Assets Effectiveness of using assets to generate operating income. IBURDEN Interest Expenses/Total Assets Interest Burden Interest expenses relative to assets generating cash flow to cover such expenses. NBTAXA Net Income Before Taxes/Total Assets Pre-tax profitability relative to asset base. TACE Total Assets/Common Equity Equity Multiplier Reflects amount of debt relative to equity in capital structure. NBTAXE Net Income Before Tax/Common Equity Pre-tax profitability relative to equity base. R&DS Research and Development Expenses/Sales Research and development commitment per dollar of sales. Panel B: Cash Flow-Based Performance Measures Variable Definition/Interpretation FCFSHR [Income Before Extraordinary Items* - Capital Expenditures*)/Common Shares Outstanding] Free Cash Flow to Equity Per Share Cash available for distribution to shareholders irrespective of dividend policy. CAPFCF Capital Expenditures*/Free Cash Flow Capital ependitures relative to available cash flow. DIVFCF Cash Dividends*/Net Income Cash dividends paid relative to income. ECFNI [Cash Dividends* + Purchase of Common and Preferred Stock*]/Net Income Total equity cash flows relative to income. Panel C: Market-Based Performance Measures Variable Definition/Interpretation PBOOK Closing Annual Common Equity Share Price/Book Value Per Share Summary measure of market valuation relative to accounting net worth. PSALE Closing Annual Common Equity Share Price/Sales Per Share Summary measure of market valuation relative to sales, which is presumed to be more standardized than is the case for reported earnings per share. BETA Beta coefficient computed using the CRSP daily NYSE/AMEX excess returns file. Measure of the systematic or market-related risk of the equity security. *Based on the Statement of Cash Flows. Table 4 Univariate Analysis of Telecommunications Industry Sectors: 1987-90 vs. 1991-94 Panel A: Regional Bell Holding Companies and GTE* Mean Variable** 1987-90 1991-94 t*** Prob > t ROE 0.1491 0.1124 2.7010 0.0087 PMARGIN 0.0930 0.0907 0.3768 0.7069 TATO 0.5223 0.5143 0.9230 0.3576 ROA 0.0480 0.0466 0.4835 0.6299 IBURDEN 0.0312 0.0300 1.9820 0.0495 NBTAXA 0.0800 0.0762 1.1001 0.2743 TACE 3.2761 3.3140 -0.4990 0.6186 NBTAXE 0.2590 0.2516 0.7260 0.4695 FCFSHR -7.4588 -6.6917 -1.7471 0.0829 CAPFCF -1.8267 -1.8127 -0.1705 0.8649 R&DS 0.0133 0.0118 1.0857 0.8369 DIVFCF -0.5276 -0.5754 0.9168 0.3609 ECFNI 0.7396 0.4114 2.8011 0.0065 PBOOK 1.7193 1.9173 -2.8726 0.0050 PSALE 1.0552 1.1757 -1.8533 0.0660 BETA 1.1607 0.7272 5.5296 0.0001 *Specific members of industry segment are listed in Table 2. **Defined as indicated in Table 3. ***Reported t is adjusted (Satterthwaite) when the homogeneity of variance test is rejected at the .05 level of significance. Table 4 Univariate Analysis of Telecommunications Industry Sectors: 1987-90 vs. 1991-94 Panel B: Independent Telecommunications Companies* Mean Variable** 1987-90 1991-94 t*** Prob > t ROE 0.1229 0.0916 1.6263 0.1066 PMARGIN 0.0849 -0.0042 1.1393 0.2573 TATO 0.5318 0.5538 -0.0020 0.9984 ROA 0.0375 0.0423 -0.6672 0.5054 IBURDEN 0.0347 0.0314 1.4321 0.1536 NBTAXA 0.0718 0.0738 -0.3018 0.7631 TACE 2.8030 2.8622 -0.3734 0.7092 NBTAXE 0.2364 0.2034 1.4010 0.1640 FCFSHR -4.2976 -3.8862 -2.0796 0.0390 CAPFCF -1.5509 -1.4248 -1.6973 0.0915 R&DS 0.0234 0.0207 0.7115 0.4776 DIVFCF -0.3936 -0.3602 -1.3383 0.1822 ECFNI 0.6932 0.6339 1.6195 0.1073 PBOOK 1.5325 1.7471 -0.7112 0.4785 PSALE 1.1403 1.3450 -1.4844 0.1395 BETA 1.2111 1.1876 0.0962 0.9284 *Specific members of industry segment are listed in Table 2. **Defined as indicated in Table 3. ***Reported t is adjusted (Satterthwaite) when the homogeneity of variance test is rejected at the .05 level of significance. Table 4 Univariate Analysis of Telecommunications Industry Sectors: 1987-90 vs. 1991-94 Panel C: Long Distance Services Mean Variable** 1987-90 1991-94 t*** Prob > t ROE 0.0624 0.0621 0.0420 0.9665 PMARGIN 0.0127 0.0152 -0.6487 0.5174 TATO 0.9133 0.9191 -0.0839 0.9332 ROA 0.0012 .0038 -0.8536 0.3945 IBURDEN 0.0279 0.0270 0.6447 0.5199 NBTAXA 0.0291 0.0308 -0.4071 0.6844 TACE 3.3024 3.2472 0.8354 0.4046 NBTAXE 0.1061 0.1075 0.0905 0.9280 FCFSHR -5.2686 -5.256 -0.0367 0.9708 CAPFCF -1.1907 -1.1945 0.1730 0.8628 R&DS 0.0235 0.0232 0.0912 0.9274 DIVFCF -0.3217 -0.3192 -0.1639 0.8700 ECFNI 7.1864 7.1041 0.0844 0.9328 PBOOK 1.7854 1.7649 0.3069 0.7593 PSALE 0.7028 0.7056 -0.0914 0.9273 BETA 1.2989 0.8074 1.3269 0.3158 *Specific members of industry segment are listed in Table 2. **Defined as indicated in Table 3. ***Reported t is adjusted (Satterthwaite) when the homogeneity of variance test is rejected at the .05 level of significance. Table 4 Univariate Analysis of Telecommunications Industry Sectors: 1987-90 vs. 1991-94 Panel D: Wireless Telecommunications Services Mean Variable** 1987-90 1991-94 t*** Prob > t ROE 0.2299 0.1908 0.2103 0.8350 PMARGIN -0.2740 -0.0866 -1.2389 0.2333 TATO 0.3610 0.3570 0.0514 0.9594 ROA -0.0740 -0.0368 -0.6845 0.5037 IBURDEN 0.0505 0.0367 0.9405 0.3616 NBTAXA -0.0235 -0.0001 -0.5379 0.5979 TACE -1.4685 0.8405 -1.2375 0.2265 NBTAXE 0.2348 0.1515 0.4137 0.6824 FCFSHR -0.8694 -1.5429 0.6935 0.4941 CAPFCF -0.2273 -0.7419 2.7776 0.0098 R&DS 0.00004 0.000004 2.7266 0.0177 DIVFCF -0.0127 -0.0621 1.1186 0.2783 ECFNI 0.1092 0.8168 -1.0809 0.2946 PBOOK 2.1070 1.7186 0.1335 0.8949 PSALE 10.2566 5.8352 1.3073 0.2130 BETA 1.3268 0.9540 0.5620 0.6134 *Specific members of industry segment are listed in Table 2. **Defined as indicated in Table 3. ***Reported t is adjusted (Satterthwaite) when the homogeneity of variance test is rejected at the .05 level of significance. Table 4 Univariate Analysis of Telecommunications Industry Sectors: 1987-90 vs. 1991-94 Panel E: Cable and Other Pay TV Services Mean Variable** 1987-90 1991-94 t*** Prob > t ROE 0.2216 -1.8501 1.1001 0.2805 PMARGIN 0.4735 -0.4769 0.0263 0.9962 TATO 0.2779 0.4014 -3.6482 0.0008 ROA -0.1322 -0.1861 0.8096 0.4246 IBURDEN 0.0692 0.0932 -1.8704 0.0669 NBTAXA -0.0630 -0.0929 0.5122 0.6124 TACE -1.2049 20.9493 -1.2822 0.2094 NBTAXE 0.1711 -1.7835 1.0491 0.3031 FCFSHR -3.5922 -5.6254 1.2651 0.2139 CAPFCF -0.5131 -0.4435 -0.6819 0.4983 R&DS 0.0017 0.0021 -0.3166 0.7528 DIVFCF -0.3063 -0.1368 -1.6513 0.1046 ECFNI -0.3146 -3.4650 0.9828 0.3340 PBOOK -0.5184 14.2295 -1.2954 0.2053 PSALE 3.3158 1.9273 1.5659 0.1279 BETA 0.6303 0.9849 -0.9384 0.3726 *Specific members of industry segment are listed in Table 2. **Defined as indicated in Table 3. ***Reported t is adjusted (Satterthwaite) when the homogeneity of variance test is rejected at the .05 level of significance. Table 4 Univariate Analysis of Telecommunications Industry Sectors: 1987-90 vs. 1991-94 Panel F: Non-U.S. Telecommunications* Mean Variable** 1987-90 1991-94 t*** Prob > t ROE 0.4231 0.3365 3.1496 0.0020 PMARGIN 0.1394 0.1624 -1.2671 0.2072 TATO 0.4067 0.4393 -1.0753 0.2840 ROA 0.0718 0.0862 -0.9137 0.3624 IBURDEN 0.0532 0.0437 2.4958 0.0137 NBTAXA 0.1250 0.1299 -0.3735 0.7093 TACE 4.0089 3.3508 2.6654 0.0086 NBTAXE 0.4485 0.3791 2.3791 0.0188 FCFSHR -1.2087 -0.7385 -0.8621 0.3903 CAPFCF 39.3803 28.3220 2.3406 0.0206 R&DS 0.0029 0.0027 0.1590 0.8739 DIVFCF 5.7037 4.6435 1.2959 0.1971 ECFNI 0.2932 0.3731 -1.1485 0.2526 PBOOK 1.7593 2.7673 -1.9919 0.0486 PSALE 1.2811 2.3652 -2.9150 0.0042 BETA 0.9553 1.2285 -0.9953 0.3365 *Specific members of industry segment are listed in Table 2. **Defined as indicated in Table 3. ***Reported t is adjusted (Satterthwaite) when the homogeneity of variance test is rejected at the .05 level of significance. Table 5 Selected Competitive Events in the Telecommunications Industry: 1990-94 Date* Description/Significance of Event 5 November 1990 MCI and British Telecom announce decision to build trans-Atlantic fiber-optic cable. International co-operative venture of potential significance to competitors. 2 June 1993 British Telecom announces intended acquisition of 20% interest in MCI for $4.3 billion in cash. Significant additonal resources to finance MCI's strategic initiatives. 13 October 1993 Bell Atlantic and TCI announce merger plans. Significant potential combination of telephone and cable. 4 April 1994 Pacific Telesis spins off AirTouch Communications (wireless). Effort to explicitly separate traditional regulated telephone operations and wireless, and to develop long-distance business by circumventing Modified Final Judgment (MFJ) restrictions. 30 June 1994 Bell Atlantic and NYNEX announce plan to combine cellular service operations. Consolidation of potential significance to competitors. 6 July 1994 Bell Atlantic, BellSouth, and Southwestern Bell file motion to remove MFJ restrictions. Visible move to challenge legality of regulatory constraints on nature of competition in U.S. telecommunications industry. 9 September 1994 AT&T announces intended acquisition of McCaw Cellular for $12.6 billion. Provides AT&T with significant wireless operations and potential ability to re-enter the local telephone market. *Event dates are the earlier of the first published announcement or the first mention of the event on the news wire, as reported by DowJones News/Retrieval. Table 6 Seemingly Unrelated Regression Results: Event Impact Analysis by Telecommunications Industry Sector* Panel A: Regional Bell Holding Companies and GTE Parameter Estimate Standard Error t Prob > t à 0.000026 0.000197 0.132 0.8949 i 0.969064 0.028795 33.654 0.0001 ë Event 5 November 1990 -0.008828 0.004863 -1.816 0.0697 MCI/BT trans-Atlantic cable plan. 2 June 1993 0.002610 0.004852 0.538 0.5908 BT 20% intended MCI stock purchase for $4.3 billion. 13 October 1993 0.001229 0.004851 0.253 0.8000 Bell Atlantic/TCI merger intention. 4 April 1994 0.002307 0.004858 0.475 0.6350 Pacific Telesis decision to spin-off AirTouch Comm. 30 June 1994 0.011137 0.004851 2.296 0.0219 Bell Atlantic/NYNEX cellular consolidation plan. 6 July 1994 -0.000702 0.004851 -0.145 0.8850 Bell Atlantic, BellSouth, Southwestern Bell file to remove MFJ restrictions. 9 September 1994 -0.004021 0.004852 -0.829 0.4073 AT&T intent to acquire McCaw Cellular for $12.6 billion. *System weighted R2 = 0.3142. Table 6 Seemingly Unrelated Regression Results: Event Impact Analysis by Telecommunications Industry Sector Panel B: Independent Telecommunications Companies Parameter Estimate Standard Error t Prob > t à 0.000145 0.000160 0.908 0.3638 i 0.559873 0.023334 23.994 0.0001 ë Event 5 November 1990 0.010018 0.003940 2.542 0.0111 MCI/BT trans-Atlantic cable plan. 2 June 1993 0.000592 0.003932 0.151 0.8803 BT 20% intended MCI stock purchase for $4.3 billion. 13 October 1993 -0.002303 0.003931 -0.586 0.5581 Bell Atlantic/TCI merger intention. 4 April 1994 -0.006048 0.003936 -1.536 0.1247 Pacific Telesis decision to spin-off AirTouch Comm. 30 June 1994 0.001594 0.003931 0.405 0.6853 Bell Atlantic/NYNEX cellular consolidation plan. 6 July 1994 -0.002800 0.003931 -0.712 0.4764 Bell Atlantic, BellSouth, Southwestern Bell file to remove MFJ restrictions. 9 September 1994 -0.003918 0.003932 -0.997 0.3192 AT&T intent to acquire McCaw Cellular for $12.6 billion. *System weighted R2 = 0.3142. Table 6 Seemingly Unrelated Regression Results: Event Impact Analysis by Telecommunications Industry Sector Panel C: Long Distance Services Parameter Estimate Standard Error t Prob > t à -0.000026 0.000281 -0.092 0.9265 i 1.145885 0.041034 27.925 0.0001 ë Event 5 November 1990 -0.004122 0.006929 -0.595 0.5521 MCI/BT trans-Atlantic cable plan. 2 June 1993 0.001196 0.006915 0.173 0.8627 BT 20% intended MCI stock purchase for $4.3 billion. 13 October 1993 0.001539 0.006913 0.223 0.8239 Bell Atlantic/TCI merger intention. 4 April 1994 -0.000491 0.006922 -0.071 0.9435 Pacific Telesis decision to spin-off AirTouch Comm. 30 June 1994 0.001626 0.006913 0.235 0.8141 Bell Atlantic/NYNEX cellular consolidation plan. 6 July 1994 -0.003674 0.006913 -0.531 0.5952 Bell Atlantic, BellSouth, Southwestern Bell file to remove MFJ restrictions. 9 September 1994 -0.004924 0.006914 -0.712 0.4765 AT&T intent to acquire McCaw Cellular for $12.6 billion. *System weighted R2 = 0.3142. Table 6 Seemingly Unrelated Regression Results: Event Impact Analysis by Telecommunications Industry Sector Panel D: Wireless Telecommunications Services Parameter Estimate Standard Error t Prob > t à 0.000095 0.000473 0.201 0.8405 i 1.499683 0.069069 21.713 0.0001 ë Event 5 November 1990 0.021106 0.011664 1.810 0.0706 MCI/BT trans-Atlantic cable plan. 2 June 1993 0.008549 0.011639 0.734 0.4628 BT 20% intended MCI stock purchase for $4.3 billion. 13 October 1993 0.008736 0.011637 0.751 0.4529 Bell Atlantic/TCI merger intention. 4 April 1994 -0.008238 0.011652 -0.707 0.4797 Pacific Telesis decision to spin-off AirTouch Comm. 30 June 1994 -0.003895 0.011637 -0.335 0.7379 Bell Atlantic/NYNEX cellular consolidation plan. 6 July 1994 -0.002853 0.011636 -0.245 0.8064 Bell Atlantic, BellSouth, Southwestern Bell file to remove MFJ restrictions. 9 September 1994 0.004873 0.011638 0.419 0.6755 AT&T intent to acquire McCaw Cellular for $12.6 billion. *System weighted R2 = 0.3142. Table 6 Seemingly Unrelated Regression Results: Event Impact Analysis by Telecommunications Industry Sector Panel E: Cable and Other Pay TV Services Parameter Estimate Standard Error t Prob > t à -0.000179 0.000575 -0.3111 0.7556 i 1.689376 0.083893 20.1370 0.0001 ë Event 5 November 1990 -0.000291 0.014167 -0.021 0.9836 MCI/BT trans-Atlantic cable plan. 2 June 1993 0.000485 0.014137 0.034 0.9726 BT 20% intended MCI stock purchase for $4.3 billion. 13 October 1993 0.051196 0.014134 3.622 0.0003 Bell Atlantic/TCI merger intention. 4 April 1994 -0.002819 0.014153 -0.199 0.8421 Pacific Telesis decision to spin-off AirTouch Comm. 30 June 1994 -0.006652 0.014134 -0.471 0.6380 Bell Atlantic/NYNEX cellular consolidation plan. 6 July 1994 -0.006700 0.014134 -0.474 0.6355 Bell Atlantic, BellSouth, Southwestern Bell file to remove MFJ restrictions. 9 September 1994 0.009853 0.014136 0.697 0.4859 AT&T intent to acquire McCaw Cellular for $12.6 billion. *System weighted R2 = 0.3142. Table 6 Seemingly Unrelated Regression Results: Event Impact Analysis by Telecommunications Industry Sector Panel F: Non-U.S. Telecommunications Parameter Estimate Standard Error t Prob > t à 0.000731 0.000304 2.402 0.0165 i 0.972810 0.044386 21.917 0.0001 ë Event 5 November 1990 0.003609 0.007495 0.481 0.6303 MCI/BT trans-Atlantic cable plan. 2 June 1993 -0.005801 0.007479 -0.776 0.4381 BT 20% intended MCI stock purchase for $4.3 billion. 13 October 1993 0.009628 0.007478 1.288 0.1981 Bell Atlantic/TCI merger intention. 4 April 1994 -0.009471 0.007488 -1.265 0.2062 Pacific Telesis decision to spin-off AirTouch Comm. 30 June 1994 0.002764 0.007478 0.370 0.7117 Bell Atlantic/NYNEX cellular consolidation plan. 6 July 1994 0.001979 0.007478 0.265 0.7914 Bell Atlantic, BellSouth, Southwestern Bell file to remove MFJ restrictions. 9 September 1994 -0.002588 0.007479 -0.346 0.7294 AT&T intent to acquire McCaw Cellular for $12.6 billion. *System weighted R2 = 0.3142. Table 7 Evaluation of Seemingly Unrelated Regression Results: Pairwise Comparative Analysis of Event Reactions by Telecommunications Industry Sector* Panel A: Analysis of Event of 5 November 1990 - MCI/British Telecom Trans-Atlantic Cable Plan Absolute Return Industry Segment Pairwise Comparision Difference FProb > F Long Distance vs. Wireless -.025 3.6841 0.0550 Long Distance vs. Independents -.014 3.2531 0.0713 Regional Bells + GTE vs. Independents -.018 9.6738 0.0019 Regional Bells + GTE vs. Wireless -.029 5.2978 0.0214 *Only those pairwise comparisons for which reactions to a given event were significantly different from one another at the .10 level of significance are presented. There are 7290 degrees of freedom in all of the above calculations. Table 7 Evaluation of Seemingly Unrelated Regression Results: Pairwise Comparative Analysis of Event Reactions by Telecommunications Industry Sector* Panel B: Analysis of Event of 13 October 1993 - Bell Atlantic/TCI Merger Intention Plan Absolute Return Industry Segment Pairwise Comparision Difference FProb > F Long Distance vs. Cable -.0497 10.4165 0.0013 Wireless vs. Cable -.0425 6.2374 0.0125 Regional Bells + GTE vs. Cable -.05 10.9318 0.0009 Independents vs. Cable -.0535 13.8157 0.0002 Non-U.S. vs. Cable -.0416 6.9683 0.0083 *Only those pairwise comparisons for which reactions to a given event were significantly different from one another at the .10 level of significance are presented. There are 7290 degrees of freedom in all of the above calculations.