Notes for "The Market for Multichannel Video Programming Subscription Service in the United States"

¹More precisely, Designated Market Areas (DMAs) are used by Nielsen Media Research to identify TV stations whose broadcast signals reach a specific area and attract the most viewers. A DMA consists of all counties whose largest viewing share is given to stations of that same market area. Non-overlapping DMAs cover the entire continental United States, Hawaii and parts of Alaska. There are currently 210 Designated Market Areas throughout the United States. ²Robert Crandall and Harold Furchtgott-Roth, Cable TV: Regulation or Competition? (Washington, DC: The Brookings Institution, 1996).

³This survey is discussed in greater detail below.

⁴The standard deviation is 34.7 percent.

⁵The Communications Act sets forth general franchise requirements and establishes limits on the franchise fees that local jurisdictions can impose as part of the franchise agreement. See 47 U.S.C. § 541,542.

⁶Warren Communications News, Inc., Television and Cable Factbook (Washington, DC: Warren Communications, 2002).

⁷The standard deviation is 14.4 percent.

⁸The standard deviation is 14.1 percent.

⁹Note that these and other summary statistics from the 2002 FCC Annual Cable Price Survey are computed as weighted averages from the results of the survey. The weights are based on the total number of subscribers (i.e., analog plus digital subscribers) from the systems responding to the survey.

¹⁰Federal Communications Commission, Ninth Annual Report, MM Docket 02-145 (Washington, DC: Federal Communications Commission, December 31, 2002).

¹¹There are four companies licensed by the Federal Communications Commission to provide DBS service: DirecTV, EchoStar (marketed as the DISH network), Dominion Video Satellite, Inc (marketed as Sky Angel), and Cablevisions’s Rainbow DBS. Three of these companies - DirecTV, EchoStar, and Dominion - currently provide service.

¹²It is important to note in the context of modeling the demand for multichannel video programming subscription service for cable service and DBS service the distinction between digital television and digital signal transmission. Digital televisions connect to a cable transmitting digital signals without a set-top box. This allows subscribers access to high definition television services offered by cable systems. A digital signal is a signal that has a limited number of discrete states prior to transmission. This is in contrast to an analog signal which varies in a continuous manner and may be said to have an infinite number of states.

¹³The standard deviation is 12.5 percent.

¹⁴Only cable television and DBS systems are considered here as the market for multichannel video programming subscription service. There are other components to this market including wireless cable, private cable or satellite master antenna TV (SMATV) systems, C-band satellite dishes, and local telephone companies. These have very small market shares and are not considered because of the absence of adequate, credible, and consistent data. Moreover, they are becoming less relevant because of their inability to compete on the basis of price and technology with cable service and DBS service.

¹⁵The standard deviation is 11.2 percent.

¹⁶The standard deviation is 12.6 percent. This change is not statistically significant at the five percent level. The Ninth Annual Report gives a market penetration value of 20.3 percent for 2002 (Federal Communications Commission, Ninth Annual Report, MM Docket 02-145 (Washington, DC: Federal Communications Commission, December 31, 2002).

¹⁷General Accounting Office, The Changing Status of Competition to Cable Television, GAO/RCED 99-158 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 1999).

¹⁸The president of DirecTV is quoted in the CableFAX Daily note as saying “... cable is not delivering on its digital promise and we are ... we are all digital” CableFAX Daily (April 15, 2003, p. 1).

¹⁹This number is different than the value reported above. This value was compiled by the Federal Communications Commission. The previously reported value was compiled by Warren Communications, Inc. Use of the Warren Communications value is intended to give a sense of what has happened to the growth in the number of cable systems. Federal Communications Commission data do not go back to 1970. The difference in the number arises due to the different definitions of precisely what constitutes a cable system.

²⁰Robert Crandall and Harold Furchtgott-Roth, Cable TV: Regulation or Competition? (Washington, DC: The Brookings Institution, 1996).

²¹Technically, a MSO is a company that operates multiple cable systems. There are in excess of 1000 MSOs in the United States. In the context being used here, however, the focus is just on the major MSOs.

²²This greater concentration of cable properties, combined with ownership ties to cable programming, raised concerns about the market power of the cable industry and the need for more competition in the delivery of multichannel video programming subscription service. Congress responded to these concerns by passing the Cable Television Consumer Protection and Competition Act of 1992. Under this act, among other things, cable television rates were regulated and exclusive contracts between cable systems and their affiliated program suppliers were eliminated. With the Telecommunications Act of 1996, however, most cable rate regulation ended as of March 31, 1999, although rates for the basic service tier of programming on cable systems could still be regulated by local franchising authorities. See 47 U.S.C. § 532.

²³The standard deviation is 11.3 percent.

²⁴George Ford and John Jackson, “Horizontal Concentration and Vertical Integration in the Cable Television Industry,” Review of Industrial Organization 12 (1997): 501-518.

²⁵Lewis Hindery, “Testimony before the Senate Commerce Committee on Commerce, Science, and Transportation,” U.S. Senate, Washington, DC, July 28, 1998.

²⁶Park Center for Investment Research, Clear Channel Communications (Ithaca, NY: Johnson Graduate School of Management, Cornell University, 2003).

²⁷The standard deviation is 39.2 percent.

²⁸William Emmons and Robin Prager, “The Effects of Market Structure and Ownership on Prices and Service Offerings in the Cable Television Industry,” Rand Journal of Economics 28 (1997): 732-750.

²⁹The standard deviation is 2.3 percent.

³⁰The standard deviation is 13.1 percent.

³¹This market share is based on results from the 2002 FCC Annual Cable Price Survey. Alternative estimates have slightly different market shares. For example, Veronis Suhler report an expected market share for cable service of 77.1 percent in 2002 (Veronis Suhler, Communications Industry Forecast [New York, NY: Veronis Suhler Media Merchant Bank, July 2001].) The Ninth Annual Report (Federal Communications Commission, Ninth Annual Report, MM Docket 02-145 (Washington, DC: Federal Communications Commission, December 31, 2002) gives a value of 79.0 percent for 2002. Nevertheless, the implications do not change - cable systems are the dominant market.

³²Kip Viscusi, John Vernon, and Joseph Harrington, Economics of Regulation and Antitrust (Cambridge, MA: The MIT Press, 2002).

³³A dominant position is frequently considered to be a 40 percent or greater market share (Frederic Scherer and David, Industrial Market Structure and Economic Performance, Third Edition (Boston, MA: Houghton Mifflin Company, 1990).).

³⁴Note that absent data on DBS prices there is no way to empirically verify this assumption. The DBS industry, which has no mandate to provide information on its pricing and operating costs and characteristics, remains a difficult entity to model.

³⁵Don Waldman and Elizabeth Jensen, Industrial Organization: Theory and Practice (Reading, MA: Addison Wesley, Inc., 1998) and Noel Uri and Wilson Mixon, “On the Optimal Pricing of a Dominant Firm,” De Economist 34 (1986): 225-227.

³⁶Seth Kahai, David Kaserman, and John Mayo, “Is the ‘Dominant Firm’ Dominant? An Empirical Analysis of AT&T’s Market Power,” Journal of Law and Economics 39 (1996): 499-517.

³⁷This is equal to not less than four percent or more than 7 percent of its capacity. See 47 U.S.C. § 335.

³⁸47 U.S.C. § 335.

³⁹Leonard Blank, David Kaserman, and John Mayo, “Dominant Firm Pricing with Competitive Entry and Regulation: The Case of IntraLATA Toll,” Journal of Regulatory Economics 14 (1998): 35-53 measure the competitive presence of the fringe as the number of fringe firms. In the multichannel video programming service market context, there are two DBS providers operating in every local market. Hence, the Blank, Kaserman, and Mayo measure cannot be used. Because there is no variation across franchise areas in the number of providers of DBS, however, the DBS share variable is used to measure the degree of competition cable systems face from noncable providers.

⁴⁰These factors impact the supply of analog and digital cable service.

⁴¹Section 623(k) was adopted as Section 3(k) of the 1992 Cable Act, Pub. L. No. 102-385, 106 Stat. 1460, codified at 47 U.S.C. § 543(k).

⁴²The 1992 Cable Act defines basic cable service as the tier of service that includes the retransmission of local television broadcast signals (47 U.S.C. § 522). Cable programming service is defined as any video programming other than video programming carried on the basic service tier, and video programming offered on a per channel or per program basis. Equipment refers to a converter box, remote control, and other equipment necessary to access programming.

⁴³Cable systems cannot be subject to rate regulation in areas where the FCC has made a finding of “effective competition.” A cable system is subject to effective competition when any one of the following conditions is met: (1) Fewer than 30 percent of the households in its franchise area subscribe to the cable service of a cable system. (2) The franchise area is served by at least two unaffiliated multichannel video programming distributors each of which offers comparable programming to at least 50 percent of the households in the franchise area and the number of households subscribing to multichannel video programming other than the largest multichannel video programming distributor exceeds 15 percent of the households in the franchise area. (3) A multichannel video programming distributor, operated by the franchising authority for that franchise area, offers video programming to at least 50 percent of the households in the franchise area. (4) A local exchange carrier or its affiliate (or any multichannel video programming distributor using the facilities of such carrier or its affiliate) offers video programming services directly to subscribers by any means (other than direct-to-home satellite services) in the franchise area of an unaffiliated cable operator which is providing cable service in that franchise area, but only if the video programming services so offered in that area are comparable to the video programming services provided by the unaffiliated cable operator in that area. In other franchise areas, local communities have the authority to regulate the rates of the basic service tier and equipment, but may or may not choose to exercise that authority.

⁴⁴The term service tier means a cable service for which the operator charges a separate rate. The major CPST tier typically meets two criteria: (1) offers the greatest number of channels among the CPST tiers, and (2) has the highest number of subscribers among the CPST tiers. Cable systems require subscribers to purchase the BST in order to purchase the CPST.

⁴⁵Local channel refers to those channels that carry local broadcast stations (either through must-carry requirement or retransmission agreement), public, educational, or government programming, commercial leased access, and other programming that originates locally. The term satellite channels refers to nationally-delivered channels on networks that are, predominately, delivered by satellite to the cable headend. Satellite channels include major regional sports networks.

⁴⁶There is a possibility that the number of DBS subscribers was systematically overestimated by the cable systems surveyed. Based on the results of the survey, for 2002 the ratio of the number of DBS subscribers to cable subscribers is 0.35. Other data sources report this ratio to be somewhat smaller for 2002. Veronis Suhler, Communications Industry Forecast (New York, NY: Veronis Suhler Media Merchant Bank, July 2001)., for example, report a value of 0.29 while the Ninth Annual Report of the Federal Communications Commission (Federal Communications Commission, Ninth Annual Report, MM Docket 02-145 (Washington, DC: Federal Communications Commission, December 31, 2002) gives a value of 0.27. The number of DBS subscribers in each cable system’s franchise area is not available. It is estimated by the cable systems on the 2002 FCC Annual Cable Price Survey. An exact count was not required.

⁴⁷Federal Communications Commission, Report on Cable Industry Prices, MM Docket 92-266 (Washington, DC: Federal Communications Commission, July 8, 2003.

⁴⁸A cable system is defined as the area served by a single headend. A headend is the control center of a cable television system, where incoming signals are amplified, converted, processed, and combined into a common cable along with any original cablecasting, for transmission to subscribers. A system operator is the individual, organization, company, or other entity that operates a cable television system.

⁴⁹Overbuilding occurs when more than one cable operator provides service in a specific geographic area. Each cable operator has its own proprietary cable wires.

⁵⁰The azimuth is the horizontal angular direction from a fixed reference point (i.e., the DBS subscriber) to a geosynchronous satellite. Elevation is the angle up needed to receive the satellite signal and skewness is required the dish rotation.

⁵¹John Mayo and Yasuji Otsuka, “Demand, Pricing, and Regulation: Evidence from the Cable TV Industry,” Rand Journal of Economics, Vol. 22 (1991), pp. 396-410.

⁵²Robert Rubinovitz, “Market Power and Price Increases for Basic Cable Service since Deregulation,” Rand Journal of Economics 24 (1993): 1-18.

⁵³Richard Biel, Thomas Dazzio, Robert Ekelund, and John Jackson, “Competition and the Price of Multiple Cable Television Services: An Empirical Study,” Journal of Regulatory Economics 6 (1993): 401-415.

⁵⁴George Ford and John Jackson, “Horizontal Concentration and Vertical Integration in the Cable Television Industry,” Review of Industrial Organization 12 (1997): 501-518.

⁵⁵Randolph Beard, Robert Ekelund, George Ford, and Richard Saba, “Price-Quality Tradeoffs and Welfare Effects in Cable Television Markets,” Journal of Regulatory Economics, 20 (2001): 107-123.

⁵⁶There are also the General Accounting Office studies (i.e., General Accounting Office, The Changing Status of Competition to Cable Television, GAO/RCED 99-158 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 1999) and General Accounting Office, The Effect of Competition from Satellite Providers on Cable Rates, GAO/RCED-00-164 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 2000) and General Accounting Office, Issues in Providing Cable and Satellite Television Services, GAO-03-130 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 2002). These rely essentially on the previous studies noted here. The 2002 General Accounting Office study explores the DBS issue but relies on proprietary data that are not readily available nor are the estimation results reproducible. There is also a series of Federal Communications Commission studies (e.g., Federal Communications Commission, Report on Cable Industry Prices, MM Docket 92-266 (Washington, DC: Federal Communications Commission, February 14, 2001) and Federal Communications Commission, Report on Cable Industry Prices, MM Docket 92-266 (Washington, DC: Federal Communications Commission, April 4, 2002). These, however, essentially reproduce the Ford and Jackson study using a different data set.

⁵⁷With a log-linear specification, all of the variables are transformed by loge before proceeding with the empirical analysis. There are theoretical reasons for doing this involving the distribution of the stochastic terms on the equations being estimated. That is, the size of the error terms frequently changes as the size of the endogenous (dependent) variables change usually in a simple multiplicative fashion. Michael Intrilligator, Econometric Models, Techniques, and Applications Englewood Cliffs, NJ: Prentice-Hall, Inc., 1978) explores these theoretical issues in detail. Whether such a transformation is appropriate here is something that needs to be verified empirically.

⁵⁸There are reception issues is some areas that preclude viewers from receiving an adequate signal. That is, for example, the signal strength is so weak in some rural areas that these viewers cannot receive over-the-air broadcast channels with a conventional antenna.

⁵⁹Occasionally, premium channels will be offered as part of the major programming tier. This is most often just a promotional device and offered for a short time.

⁶⁰Diane Anstine, “How Much Will Consumers Pay? A Hedonic Analysis of the Cable Television Industry,” Review of Industrial Organization 19 (2001): 129-147.

⁶¹Jith Jayaratne, “A Note on the Implementation of Cable TV Rate Caps,” Review of Industrial Organization 11 (1996): 823-840.

⁶²National Cable and Television Association, Broadband Services (Washington, DC: National Cable and Television Association, 2003).

⁶³DBS also offers tiers. The prices for the various tiers, as for the price of the basic tier, are uniform nationally.

⁶⁴See Jith Jayaratne, “A Note on the Implementation of Cable TV Rate Caps,” Review of Industrial Organization 11 (1996): 823-840 for a discussion of this.

⁶⁵Louis Phlips, Applied Consumption Analysis (Amsterdam, The Netherlands: North-Holland Publishing Company, 1974).

⁶⁶Even if the requisite data were available, the DBS supply equation would not be very interesting since the elasticity of supply of DBS service is almost infinite. That is, the marginal cost of supplying an additional subscriber once the satellite system is in place and broadcasting is approximately zero.

⁶⁷Hal Singer, Does Clustering by Incumbent Cable MSOs Deter Entry by Overbuilders? (Washington, DC: Criterion Economics, May 2003).

⁶⁸Russell Davidson and James MacKinnon, “Several Tests for Model Specification in the Presence of Alternative Hypotheses,” Econometrica 49 (1981): 781-793 and Russell Davidson and James MacKinnon, Estimation and Inference in Econometrics (Oxford, UK: Oxford University Press, 1993).

⁶⁹Roger Bowden and Darrell Turkington, Instrumental Variables (Cambridge, UK: Cambridge University Press, 1984) and Dane Wu, “Alternative Tests of Independence between Stochastic Regressors and Disturbances,” Econometrica 40 (1973): pp. 733-750.

⁷⁰Since the variable is constant, the impact of the number of DBS channels will simply be on the constant term in the DBS demand equation.

⁷¹The number of digital channels in the most highly subscribed tier is defined as the number of channels in the major digital tier.

⁷²Randolph Beard, Robert Ekelund, George Ford, and Richard Saba, “Price-Quality Tradeoffs and Welfare Effects in Cable Television Markets,” Journal of Regulatory Economics, 20 (2001): 107-123.

⁷³Gregory Crawford, “The Impact of the 1992 Cable Act on Household Demand and Welfare,” Rand Journal of Economics 31 (2000): 422-449.

⁷⁴Neil Ericsson and John Irons, Testing Exogeneity, (Oxford, UK: Oxford University Press, 1994).

⁷⁵Douglas Spencer and Karl Berk, “A Limited Information Specification Test,” Econometrica 49 (1981): 1079-1085.

⁷⁶William Greene, Econometric Analysis, Third Edition (Upper Saddle River, NJ: Prentice Hall, Inc, 1997).

⁷⁷David Belsley, Edwin Kuh, and Roy Welsch, Regression Diagnostics: Identifying Influential Data and Sources of Collinearity (New York, NY: John Wiley and Sons, Inc., 1980).

⁷⁸David Belsley, Edwin Kuh, and Roy Welsch, Regression Diagnostics: Identifying Influential Data and Sources of Collinearity (New York, NY: John Wiley and Sons, Inc., 1980).

⁷⁹One of the details worthy of note is the impact that the presence of collinearity among the exogenous variables has on the parameter estimates. Considerable time was spent on mitigating the influence of collinearity. This was done by a judicious examination of the data for the presence of a high degree of correlation between the exogenous variables individually or in concert. When collinearity appeared to be an insurmountable problem, the variables were either combined or one or more deleted from the specification. One disappointing result was the general lack of finding any sort of statistically significant relationships between the variables included in the large set of supplementary data compiled from the 2000 Census of Population Summary Files provided by the U.S. Census Bureau and the demand and supply of analog and digital cable service and DBS service. These supplementary data were quite comprehensive consisting of information on such things as school enrollment, educational attainment, marital status, disability, language spoken, nativity, employment status, commuting patterns, occupation, income, type of housing unit, number of occupants, house heating fuel, mortgage status, rent, and tenure of occupants.

⁸⁰In the sense being used here, robust estimates are estimates that are not extremely sensitive to the functional specification or to which variables are included or excluded from the specification.

⁸¹Michael Intrilligator, Econometric Models, Techniques, and Applications (Englewood Cliffs, NJ: Prentice-Hall, Inc., 1978).

⁸²Log of the likelihood function = -29743.4

⁸³The definitions are sequential and given just once. They are not repeated for subsequent equations even though they are included.

⁸⁴Franklin Fisher, The Identification Problem in Econometrics (New York, NY: McGraw-Hill Book Company, Inc., 1966).

⁸⁵Note that for the coefficient estimates to be meaningful, it was necessary to instrument the analog cable price and the digital cable price. See, e.g., Roger Bowden and Darrell Turkington, Instrumental Variables (Cambridge, UK: Cambridge University Press, 1984) for a discussion of the theoretical econometric issues. The instruments for analog cable price include the price of digital cable, the number of households in the franchise area passed, the type of cable system (e.g., large, very large, municipal), the number of BST plus CPST analog channels available less the number of broadcast channels, whether there was local regulation for the basic service tier, population density, and the capacity of the system. The instruments for digital cable price include the price of analog cable service, the number of households in the franchise area passed, the type of cable system (e.g., large, very large, municipal), the number of BST plus CPST analog channels available less the number of broadcast channels, the capacity of the system, and population density. In preliminary analysis the penetration of DBS subscribers was also considered as a possible explanatory variable for both price relationships but it did not prove to be a significant explanatory factor.

⁸⁶The price of digital cable service typically includes the price for BST and CPST analog service plus a surcharge for digital service because digital service typically is not offered separately or as a standalone service.

⁸⁷George Ford and John Jackson, “Horizontal Concentration and Vertical Integration in the Cable Television Industry,” Review of Industrial Organization 12 (1997): 501-518.

⁸⁸General Accounting Office, Issues in Providing Cable and Satellite Television Services, GAO-03-130 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 2002).

⁸⁹George Ford and John Jackson, “Horizontal Concentration and Vertical Integration in the Cable Television Industry,” Review of Industrial Organization 12 (1997): 501-518.

⁹⁰Randolph Beard, Robert Ekelund, George Ford, and Richard Saba, “Price-Quality Tradeoffs and Welfare Effects in Cable Television Markets,” Journal of Regulatory Economics, 20 (2001): 107-123.

⁹¹It is interesting to note the relevance of these estimates. From the 2002 FCC Annual Cable Price Survey, it is estimated that 66.4 percent (with a standard deviation of 33.1 percent) of the households passed subscribed to cable or another subscription television service rather than relying on over-the-air broadcast technology for their multichannel video programming. This represents an average value across all cable systems surveyed. The results presented here are marginal values. That is, they indicate the change in the number of subscribers as the number of households passed changes.

⁹²For this comparison it is necessary to consider both the coefficient estimates and the standard errors of the estimates on the analog cable demand and digital cable demand and the analog cable supply and digital cable supply equations.

⁹³That is, construction in areas already served by a cable operator.

⁹⁴George Ford and John Jackson, “Horizontal Concentration and Vertical Integration in the Cable Television Industry,” Review of Industrial Organization 12 (1997): 501-518.

⁹⁵John Mayo and Yasuji Otsuka, “Demand, Pricing, and Regulation: Evidence from the Cable TV Industry,” Rand Journal of Economics, Vol. 22 (1991), pp. 396-410.

⁹⁶Hendrick Houthakker and Lester Taylor, Consumer Demand in the United States (Cambridge,, MA, Harvard University Press, 1970).

⁹⁷Noel Uri and Keith Brown, A Hedonic Price Analysis of Cable Channel Offerings (Washington, DC: Media Bureau, Federal Communications Commission, September 2003).

⁹⁸Charles Ferguson, Microeconomic Theory (Homewood, IL: Richard D. Irwin, Inc., 1972).

⁹⁹Randolph Beard, Robert Ekelund, George Ford, and Richard Saba, “Price-Quality Tradeoffs and Welfare Effects in Cable Television Markets,” Journal of Regulatory Economics, 20 (2001): 107-123.

¹⁰⁰General Accounting Office, Issues in Providing Cable and Satellite Television Services, GAO-03-130 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 2002).

¹⁰¹The wrong sign indicates that the estimated coefficient is negative implying that analog cable service is an inferior good. This can hardly be the case.

¹⁰²Also recall that the data on the number of DBS subscribers are reported by the cable systems. There appears to be slight measurement error in these data. This obviously raises the question of the reliability of the coefficient estimates on the DBS demand equation (William Greene, Econometric Analysis, Third Edition [Upper Saddle River, NJ: Prentice Hall, Inc, 1997] and Noel Uri, “A Note on the Estimation of the Demand for Sugar in the Presence of Measurement Error in the Data,” Applied Economics 27 [1995]: 83-94).

¹⁰³General Accounting Office, Issues in Providing Cable and Satellite Television Services, GAO-03-130 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 2002).

¹⁰⁴By the way they are introduced, they serve only to impact the estimate of the constant (intercept) term.

¹⁰⁵As before, different measures of the number of analog cable channels were considered. None proved to be statistically significant.

¹⁰⁶Table A in the Appendix gives a precise definition of the variable.

¹⁰⁷General Accounting Office, Issues in Providing Cable and Satellite Television Services, GAO-03-130 (Washington, DC: Resources and Community Division, U.S. General Accounting Office, 2002).

¹⁰⁸Richard Biel, Thomas Dazzio, Robert Ekelund, and John Jackson, “Competition and the Price of Multiple Cable Television Services: An Empirical Study,” Journal of Regulatory Economics 6 (1993): 401-415.

¹⁰⁹Robert Rubinovitz, “Market Power and Price Increases for Basic Cable Service since Deregulation,” Rand Journal of Economics 24 (1993): 1-18.

¹¹⁰William Emmons and Robin Prager, “The Effects of Market Structure and Ownership on Prices and Service Offerings in the Cable Television Industry,” Rand Journal of Economics 28 (1997): 732-750.

¹¹¹By definition a cluster is a group of systems in close geographic proximity owned by the same cable operator. That is, clustered systems are owned by a MSO. But also, it is only the largest operators that cluster.