|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 DETERMINANTS OF THE TIMING OF DIGITAL TV ADOPTION IN THE U.S. BROADCAST INDUSTRY By HEEJUNG KIM A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR T HE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 201 1
2 2011 Heejung Kim
3 In the m emory of m y f ather, Kim Jin sik
4 ACKNOWLEDGMENTS I appreciate Dr. David H. Ostroff for his support and guidance during my doctoral studies and the dissertation research process I also wish to thank my dissertation committee members, Dr. Sylvia Chan Olmsted Dr. Ron ald Rodgers, and Professor Sanford V. Berg, for their direction and feedback In particular, Professor thoughtful and thorough co mments and edits benefited the style and organization of my dissertation. My sincere thanks go to P rofessor David Waterman who guided me to focus on media economics during my master s program at Indiana University I am greatly indebted to Professor Kim Y oung seok and Professor Kang Tae young at Yonsei University, and Professor Park Chun il at Sook m yung Women s University for their encouragement that enabled me to pursue doctoral studies I want to thank Lee Jeong yeon Kim Woo sik, Won Young sun Shin Hy un jong and Lee Sung mi for being good listeners and keeping me on track My special thanks go to Heo Jun and Lee Chunsik for coffee break s and discussions during my years in the doctoral program I thank my brot h er and sisters for their encouragement an d support My deepest gratitude is to my mother, Chin Young cha for her love and belief in me. There would be no way I could have finished this dissertation without her.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 Digital TV in the Media Industry ................................ ................................ .............. 13 Significance of the Cur rent Study ................................ ................................ ........... 15 Purpose of the Study ................................ ................................ .............................. 17 Broadcast TV Industry ................................ ................................ ............................ 19 D igital TV Transition ................................ ................................ ................................ 21 Development of Digital TV ................................ ................................ ................ 25 Mandated Transition ................................ ................................ ......................... 27 FCC : Regulatory decisions ................................ ................................ ........ 28 Congress: Legislative decisions ................................ ................................ 30 Timeline ................................ ................................ ................................ ..... 32 Technical standards ................................ ................................ ................... 33 Multicasting ................................ ................................ ................................ ...... 37 Digital Strategies ................................ ................................ ................................ ..... 40 Competitive Advantage ................................ ................................ .................... 40 Diversification ................................ ................................ ................................ ... 41 Chapter Overview ................................ ................................ ................................ ... 42 2 LITERATURE REVIEW AND RESEARCH QUESTIONS ................................ ....... 43 Theories of Firm Behavior ................................ ................................ ....................... 43 Industrial Organization Traditio n ................................ ................................ ....... 43 Strategic Management Tradition ................................ ................................ ...... 44 Theories of Innovation Adoption ................................ ................................ ............. 46 Concepts of Adoption, Innovation, and Diffusion ................................ .............. 47 Patterns of Innovation Adoption and Diffusion ................................ .................. 49 Digital TV as Inno vation ................................ ................................ .................... 50 Innovation Adoption and Diffusion Theories ................................ ..................... 51 Levels of Innovation Adoption Studies ................................ ............................. 52 Individual level approach ................................ ................................ ........... 53 Organizational level approach ................................ ................................ ... 54 Innovation Adoption by Firms ................................ ................................ ................. 56 Timing of Innovation Adoption ................................ ................................ .......... 56
6 Early Adopters ................................ ................................ ................................ .. 57 Determinants o f the Timing of Adoption ................................ ................................ .. 58 Firm Characteristics ................................ ................................ ......................... 58 Ownership ................................ ................................ ................................ .. 59 Horizontal integration ................................ ................................ ................. 60 Market Characteristics ................................ ................................ ...................... 64 Competition ................................ ................................ ................................ 65 Market size ................................ ................................ ................................ 66 Market Demographics ................................ ................................ ...................... 66 3 METHODS ................................ ................................ ................................ .............. 70 Research Design ................................ ................................ ................................ .... 70 Sample Selection ................................ ................................ ................................ .... 71 Data Collection ................................ ................................ ................................ ....... 74 Measurement ................................ ................................ ................................ .......... 74 Dependent Variable ................................ ................................ .......................... 74 Independent Variables ................................ ................................ ..................... 75 Firm characterist ics ................................ ................................ .................... 75 Market characteristics ................................ ................................ ................ 77 Market demographics ................................ ................................ ................ 78 Data Ana lysis Methods ................................ ................................ ........................... 80 4 RESULTS ................................ ................................ ................................ ............... 82 Measurement of the Dependent Variable ................................ ............................... 82 Descriptive Statistics ................................ ................................ ............................... 84 Correlation Analysis ................................ ................................ ................................ 86 Determinants of the Timing of Digital TV Adoption ................................ ................. 89 The Largest 25 TV Group owned Stations ................................ ....................... 90 The Largest 11 to 25 TV Group owned Stations ................................ .............. 90 Firm Characteristics ................................ ................................ ......................... 90 Market Characteristics ................................ ................................ ...................... 91 Market Demographics ................................ ................................ ...................... 92 Comparison of Independent Variables ................................ ............................. 94 Independent Samples T tests ................................ ................................ ................. 95 ANOVA ................................ ................................ ................................ ................... 97 Discriminant Analysis ................................ ................................ .............................. 99 Summary ................................ ................................ ................................ .............. 103 5 DISCUSSION AND CONCLUSION ................................ ................................ ...... 110 Summary of Major Findings ................................ ................................ .................. 110 Firm Characteristics ................................ ................................ ....................... 112 Ownership ................................ ................................ ................................ 113 Horizontal integration ................................ ................................ ............... 113 Vertical integration ................................ ................................ ................... 114
7 Market Characteristics ................................ ................................ .................... 116 Competition ................................ ................................ .............................. 117 Market size ................................ ................................ ............................... 117 Market Demographics ................................ ................................ .................... 118 Income ................................ ................................ ................................ ..... 119 Population density ................................ ................................ .................... 119 Measures of the Timing of Adoption ................................ ............................... 121 Sample Compositions ................................ ................................ .................... 121 Early Adopters and Late Adopters ................................ ................................ .. 122 Theoretic al Implications ................................ ................................ ........................ 1 24 Industry Implications ................................ ................................ ............................. 126 Policy Implications ................................ ................................ ................................ 128 Lessons from U.S. Digital TV Transition ................................ ............................... 131 Concluding Remarks ................................ ................................ ............................ 134 Limitations and Suggested Further Studies ................................ .......................... 135 LIST OF REFERENCES ................................ ................................ ............................. 137 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 147
8 LIST OF TABLES Table page 1 1 Digital TV transition timeline: Legislative and regulatory measures .................... 31 1 2 Digital TV formats in the ATSC standard ................................ ............................ 34 2 1 Programming sources of network affiliates ................................ ......................... 62 2 2 Literature on determinants of innovation adoption ................................ .............. 68 3 1 Group ownership in U.S. commercial TV stations ................................ .............. 73 3 2 The largest 25 TV group s ................................ ................................ ................... 73 3 3 Measurement of variables and data source for the model ................................ .. 79 4 1 Timing of digital TV adoption by affiliates and independents (1997 2009) ......... 83 4 2 Descriptiv e statistics: The largest 25 TV group owned stations .......................... 85 4 3 Descriptive statistics: The largest 11 25 TV group owned stations .................... 86 4 4 Correlation matrix : The largest 25 TV group owned stations .............................. 88 4 5 Correlation matrix: The largest 11 to 25 TV group owned stations ..................... 89 4 6 Regression analysis for the largest 25 TV group owned stations: Year of adoption ................................ ................................ ................................ .............. 93 4 7 Regression analysis for the largest 25 TV group owned stations: Adoption period ................................ ................................ ................................ ................. 93 4 8 Regression analysis for the largest 11 25 group owned stations: Year of adoption ................................ ................................ ................................ .............. 93 4 9 Regression analysis for the largest 11 25 group owned stations: Adoption period ................................ ................................ ................................ ................. 94 4 10 Comparison among independent variables: The largest 25 TV group owned stations ................................ ................................ ................................ ............... 94 4 11 Comparison among independent variables: The largest 11 to 25 TV group owned stations ................................ ................................ ................................ .... 95 4 12 Independent samples t test: The largest 25 TV group owned stations ............... 96 4 13 Independent samples t test: The largest 11 to 25 TV group owned stations ...... 97
9 4 14 ANOVA: The largest 25 TV group owned stations ................................ ............. 98 4 15 ANOVA: The largest 11 to 25 TV group owned stations ................................ .... 99 4 16 Adopter category ................................ ................................ .............................. 101 4 17 Discriminant analysis: Early adopters and late adopters ................................ .. 102 4 18 Classification of early adopters and late adopters ................................ ............ 102 4 19 Summary of findings ................................ ................................ ......................... 103 5 1 Comparison of results of correlation and regression analyses ......................... 112 5 2 Determinants of the timi ng of digital TV adoption ................................ ............. 120 5 3 Common determinants of the timing of digital TV adoption .............................. 122
10 LIST OF FIGURES Figure page 1 1 Organization of the broadcast TV industry ................................ ......................... 19 1 2 Yearly hours of color broadcasting in major broadcast networks ........................ 24 2 1 Research model: Factors that determine the timing of digital TV adoption by broadcast stations ................................ ................................ .............................. 69 4 1 Timing of digital TV adoption by broadcast stations (199 7 2009) ....................... 83 4 2 Timing of digital TV adoption by affiliates and independents (1997 2009) ......... 84 4 3 Percentages of stations in adopte r category ................................ .................... 102 4 4 Importance of factors in the largest 25 TV group owned stations ..................... 109 4 5 Importance of factors in the largest 11 to 25 TV group owned stations ............ 109
11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DETERMINANTS OF THE TIMING OF DIGITAL TV ADOPTION IN THE U.S. BROADCAS T INDUSTRY By Heejung Kim D ecember 2011 Chair: David H. Ostroff Major: Mass Communication The digital TV transition provided business opportunities and challenges not only to broad cast TV stations, but also to all stakeholders in the Multichannel Video Programming Distribution ( MVPD ) industry. To explain the dynamics behind the digital TV transition this study investigate s the impacts of firm characteristics, market characteristics and market demographics on the timing of digital TV adoption by broadcast stations This study provides a n overview of the broad cast TV indust ry, summarizing the background information of the digital TV transition and the activities by digital leaders in the TV market. For empirical analysis, this study employed m ultiple regress ion independent samples t test s ANOVA, and discriminant analysis Multiple regression was conducted for stations owned by the largest 1 to 25 TV groups and stations owned by the largest 11 to 25 TV groups. The result s of multiple regression indicate that ownership, horizontal integration, vertical integration, competition, market size, and income were significant factors that influence d the timing of digital TV adoption for the l argest 25 TV group owned stations By contrast, vertical integration, competition, and market size were significant factors that influence d the timing of digital TV adoption f or the largest 11 to 25 TV group owned stations The
12 results of ANOVA indicate th at the mean differenc es of the timing of adoption were significant among ABC affiliates, CBS affiliates, NBC affiliates, Fox affiliates, other affiliates, and independents. The result s of two group discriminant analysis sugg est that early ad opters and late adopters were distinguished from each other in terms of ownership, horizontal integration, competition, income, and population density. The find ings suggest that market size was consisten tly the most influential determinant of the timing of digital TV ado ption.
13 CHAPTER 1 INTRODUCTION Digital TV in the Media Industry The digital TV transition is a unique technological innovation in the U.S. During recent decades, competition in the media industry has been increasing due to technological innovation s MVPD s (Multichannel Video Programming Distributors) have enter ed into the diverse media market s where they need to make dec isions on when to adopt particular innovation and which technolog ies they choose to market to consumers D igital technolog y plays an impo rtant role in promot ing the development of society and ec onomy as well as developing competitive strategies of industry sectors. The initial development of digital TV by b roadcast stations was slower than in the cable and satellite TV segments of the mark et Cable TV started digi tal cable service in the 1990s ; t he digital penetration rate of basic c able TV service amounted to 74.7% in 2010 (NCTA, 2011) By t aking a dvantage of digital technology satellite TV was able to o ffer full digital services to subsc ribers from the start. A lthough cable TV and satellite TV deploy ed digital technology ahead of broadcast TV stations, b roadcast TV stations played a key role in the digital TV transition in the U.S. since they had to imp lement the government mandated tran sition to digital TV. T he digital TV transition has provided a range of business and technical challenges for stakeholders in the media industr y including broadcast TV stations, cable TV satellite TV, and equipment manufacturers The tra nsitio n has chang ed the delivery methods of TV signals, content production, distribution, and exhibition, business models, and st rategies of the industry Researchers have examined several industry and policy issues in the process of the digital TV transition (Goodstadt, 2 008;
14 Hart, 2005; Waterman & Han, 2009). First, after the transition was complete, analog channels used by broadcast TV s tations were returned to the FCC to make t he 108 MHz of spectrum available for public sa fety an d wireless communication uses Since the r eturn of analog spectrum could increase federal revenue, the government encouraged rapid implementation of the transi tion (Hart, 2005) Second, all U.S. households could participate in the Converter Box Coupo n Program to apply for coupons redeemed for con verter boxes (NTIA, 2007). Third, b roadcast TV stations were able to deliver multiple digital channels of programming and i nteractive TV service through the digital TV transition, which enabled the stations to compete with new services available from cable and satellite TV systems. As of April 2008, 1,817 stations had received a digital TV license or construction permit, and 1,427 (78.5%) among these stations were on the air with digital broadcasting. By June 2009, 1,785 full power stations completed the tr ansition to digital TV (FCC, 2009). As of June 2006, the FCC reported that there were 88.3 million MVPD households in the U.S., which represent ed 86.4% of U.S. TV households ( FCC 2007 ). Since broadcast statio ns rel y on cable and satellite TV systems for t ransmitting TV channels, over 11% o f total TV households receive exclusively over the air TV signals without connection to cable or satellite TV service In December 2008, over the air TV households amounted to o ve r 11% of total TV households, 1 2 6 million out of 114 million TV households (FCC 2007 ). A t the same time, cable TV, a major distributor of broadcast TV channels had the largest share of 78% in the MVPD market (NCTA, 2009). Cable TV systems are also migrating to digital operation As of December 2 010, cable TV service passed 12 8 5 million homes among which the number of basic cable
15 customers was 59 8 million and the number of digital cable custom ers was 4 4.7 million (NCTA, 2010 ). Identifying driving forces behind the digital TV transition is an important area of research because it helps policymakers understand the conditions affecting broadcast stations adopting new technologies The digital TV transition brought new business opportunities to broadcast TV stations that had depended on cable and DBS systems in order to offer digital TV service Examining the dynamics behind digital TV adoption t his study identifies factors that influence d the timing of digital TV adoption. Thus, a fundamental question that guide s this study is : what factors dete rmine d the timing of digital TV adopti on in the broadcast TV industry? This study ex plores th is question focusing on firm characteristics, market characteristics, and market demographics of broadcast TV stations Significance of the Current Study Grounde d in the fields of economics and management, adoption studies have been conducted in various industries. The industries studied include agriculture (Griliches, 1957), electricity (Rose & Joskow, 1990; Knittel, 2006), and auto mobiles (Morgan & Daniels, 2001 ). The broadcast TV industry has unique characteristics distinguished from other industries, including the structure of production distribution exhibition and the two sided nature of the market. 1 In spite of this, little systematic study has been conducted on innovation adoption in the broadcasting industry. P revious 1 I n a two sided market, demand on one side of the market depends on demand on the other side. For example, broadca st TV offers programs to viewers in exchange for selling advertising Television viewers pay the cost of watching programs partially or entirely by viewing advertising, while advertisers usually pay broadcasters based o n total number of viewers (Wilbur, 20 08).
16 studies also lack integrated frameworks that lead researchers to investigate the dynamics behind the digital TV transition. Exploring the factors that influence the timing of digital adoption is an important for broadcast stations because a successful transition of stations would e nable them to gain competitive advantage over other broadcast stations Furthermore broadcast stations that adopt digital TV can compete with other MVPD service prov iders, including cable and satellite TV whose systems mostly deliver broadcast content to consumers. Since the digital TV transition in the U.S. had the goal to promote economic growth an d social development of t he nation the success of the transition wa s important for the industry and public policy s ectors. However, little research has focused on dynamics of the d igital TV transition leading to different corporate decisions on digital TV adoption Moreover, there is little empirical evidence indicating w hich factors influence d the timing of digital TV adoptio n in the broadcasting industry This gap may reflect the popular trend of adoption research in mass communicatio n, focusing on the micro individual level issues, such as user percept ion and attitude ( Atkin, Jefferies & Neuen dorf, 1998; Kang, 2003). Despite the advantages of the micro individual approach in understanding individual adoption behaviors the macro industry appro ach is useful for explaining firm level adoption decisions and thus can contr ibute to the understanding of the dynamics behind adoption decisions in the broadcast industry In the current study, different theories and approaches are incorporated to investigate digital TV adoption First, this study employs t he theories of i nnovatio n adoption and firm behavior to investigate the decisions of digital TV ado ption by broadcast stations with different firm and market characteristics. Second, this study
17 takes th e supply side approach to study innovation adoption while considering the con sumer demand component by including market demographics. This study can also contribu t e to the development of t he digital TV transition in other countries Although not the first nation to complete the digital TV transition, the U.S. is at the forefront of digital broadcasting through investing concerted efforts Whereas each country has different conditions in policy econo my, technology and politics, the digital TV trans ition in the U.S. can provide lesson s to broadcasters and policymakers in other count ries on what factors play a significant role in the process of the transition T his study also can contribute to broadening the scope of media research by investigating innovation adoption in the broadcasting industry. The digital TV transition has led to subsequent innovations that will introduce new business models and new delivery methods to reach consumers, such as OTT TV, mobile D TV and 3D TV. For example, as an innovation of transmission capabilities, mobile DTV allows broadcasters to use their digi tal subchannels to transmit TV programs to mobile devices. I t is also useful to conduct follow up studies of transition such as stat ions transition strategies and their relationship to the factors that affect the adoption of digital TV As others hav e emphasized, new technologies can be very disruptive to enterprise. Since this study provides the industry analysis of stations during the digital TV transition, it will serve as a guideline to investigate research issues in the post transition period. P urpose of the Study The purpose of this study is to invest igate selected factors that influence the timing of digital TV adoption in the broadcast TV industry Focusing on the unique structural characteristics of the broadcasting industry, this research se eks to find out which firm and market characteristics of broadcasting stations affect ed the timing of digital TV
18 adoption. Broadcast stations have unique characteristics compared to other media organization s and general non media businesses First, broadca st stations are heavily regulated and must get a license from the FCC before they operate. Consequently, they have to comply with the FCC rules and regulations, and also are expected to serve as public trustees in their communities (Carter et al., 2008). S econd, broadcast stations are largely involved in the production and distributio n of TV programs, which have public goods characteristics. The production cost of a public good is independent of the number of consumers and the consumption of a public good does not diminish its availability to other people. Third, broadcast stati ons have unique ownership patterns, such as network affiliation and group ownership. The structural conditions o f the broadcasting industry distinguish broadcast stations from other media firms and other non media firms in general industries. For this reason, this study focuses on the market structure of the broadcasting industry. Since the digital TV transition was an innovation adoption mandated by government, the t echnological an d institutional elements of the transition have largely been determined by government decisions In this respect the supply side approach can adequately explain the behaviors associated with digital TV adoption by broadcast stations. To investi gate the di gital TV transition, market and industry analysis is necessary to tease out a comprehensive understanding of the driving forces behi nd the digital TV adoption. Comprehensive market and industry analysis can offer structured suggestions for enhancing broadc itiveness among MVPD providers Such analysis of digital TV adoption also can contribute to future decision making by policymakers initiating innovation adoptions to bolster the national economy and
19 enhance market competition. To establ ish the context facing decision makers, t he following section examines the organization of the broadcast TV industry providing a general overview of industry operations. Broadcast TV Industry Te chnological developments including digital TV are affect ing the organization of the broadcast TV industry. The broadcast TV industry faces challenges from other MVPD service s that fractionalize audiences ; the platforms include cable TV, satellite TV the Internet, telephone and mobile services The market base d organization of the broadcast TV indus try is characterized by four different establishment s : (1) the market for equipment (equipment producers), (2) the market for program production (program producers), (3) the m arket for program distribution ( broadcast TV networks, broadcast stations, cable TV and satellite TV), and (4) the market for program exhibition (viewers). With the adoption of digital TV, traditional market boundaries are fading since programs are delivere d through multiple platforms including TV stations, cable systems, satellite systems, and the Internet (Bates, 2007). Evolving technologies also introduce n ew delivery system s such as Over the T op (O TT ) TV Figure 1 1 depicts the organization of the broadcast TV industry with simplified flow s of digital programs F igure 1 1. Organization of the b roadcast TV i ndustry Equipment Producers Viewers Broadcast Networks Cable Systems Satellite Systems O TT Systems Broadcast Stations Program Producers
20 Program P roduction TV programs are essential to numerous outlets in the TV industry to maintain qualit y broadcasts and appeal to audiences (Owen & Wildman, 1992). Program producers create prerecorded programs delivered to broadcast TV networks, TV stations, cable TV and satellite TV. L arge production companies su ch as Hollywood studios are primar y provider s of TV programs. In this industry, d igital technolog y lower s the b arriers to entry for individuals and firms that want to move into new markets As a result, small companies and individuals obtain the advantages of digital technology and utilize the advantages in producing programs Due to increasing c ompet ition among pr ogram produ cers, program producers a re becoming specializ ed, invest ing in advance d digital equipment (IBIS, 2010). Program Distribution The program distribution providers include broadcast TV networks, broadcast TV stations, cable systems and satellite systems. As a n advertiser supported medium, commercial broadcast TV has largely relied on t he network system that offers (Smith, Wr ight & Ostroff, 1998) to the partic ipants in the system The broadcast network system al lows broadcasters to achieve efficiencies through distributing programs to multiple station s at the same time (Sterling & Kitross, 2002). In the broadcast network system, n etworks and stations can p rovide each other what they need in that networks do not have local presenc e and stations have less access to regional and national programming and potential advertisers. In the U.S., c able TV systems are the most popular method of distribut ing TV programs to viewers Providing a great number of channels cable TV systems have been a challenge to the viability loc al broadcasters. In addition to cable TV s atellite TV systems have enable d broadcast stations to deliver TV programs to a
21 large number of audiences in wider geographical areas. A ddition al channel s gaine d from the digital TV transition allow broadcast stations to compete with cable and satel lite TV systems Digital techno logy contributes to expand ed program packaging and distribution alternatives to broadcast TV With a growing number of TV channels, the availability of new video services are likely to increase the demand for programming Program Exhibition TV p rograms are offere d to over the air TV viewers, cable TV subscribers, and satellite TV subscribers on their TV sets. Furthermore, the availabilit y of video programming over the Internet has increased the opportunity to expand the consumer base. For example, OTT TV is a group of Internet based TV systems connected to TV sets such as Google TV, Pandora, Hulu, and Netflix (Moyler & Hooper, 2009) OTT TV service s are transmitted directly from program provider s to viewer s using an Internet connection bypass ing over the air, cable, and satellite delivery of TV programming. The delivery of digital TV broadcasting via OTT TV systems can be a cost effective way of distributing TV programming to a large number of audiences The next section discusses the brief histo ry and the mandatory nature of the digital TV transition. The process of t he digit al TV transition was fraught with o verarching challenges and co ncerns. The government and the related industries exerted efforts to complete the dig ital TV transition This study emphasizes the mandatory nature of the digital TV transition focusing on the FCC s role of making the decision s for the transition. Digita l TV Transition The digital TV transition has been hailed as one of the most fundamental changes in the U.S. bro adcasting industry (Garperin, 2004; Waterman & Han, 2009). The digital
22 TV transition was initiated as a w ay to sustain the U.S. econom y through maintaining competitiveness in the global electronics markets. The transition came out of the decision to improve the performance of the U.S. consumer ele ctronics industry which compet ed with firms based in Europe and Asia. The motivation for the tr ansiti on came from the need to consolidate U. S. dominance in th e high tech and consumer electronics industries (Brinkley, 1997). The digital TV transition had a technological goal of advancing broadcast TV into the digital age and a policy goal of making the tra nsition as smooth as po ssible without disrupting viewers (FCC, 2009). T he transition enabled the government to receive federal earnings through reallocating spectrum. C onsumers also benefited from implementing the digital TV transition by receiving expande d programming in improved picture resolution and enhanced audio Adoption of Color TV During its history, the broadcasting system has adopted a number of technical innovations. The adoption of color TV is perhaps comparable to the digital TV transition in its scope and significance. During the sixty years prior to the digital TV transition the standards for U.S. television had not basically changed The FCC adopted the NTSC (National Television System Committee) standard for commercial television licenses in 1941 and approved the NTSC color standard which was compati ble with existing TV receivers in 1953. A number of previous studies have compared color TV adoption to digital TV adoption focusing on the role of government policy (Adda and Ottaviani, 2005; Hart, 2004; Leiva et al., 2006). As a major innovation adoption in the broadcast industry, the adoption of color TV had a similarity with that of digital TV in that color adoption had a formal technical standard de termined by the government. However, colo r TV broadcasts were transmittable and receivable on either
23 mo nochrome or color TV recei vers, and did not require the replacem ent of existing NTSC TV set s for reception of signals. The major difference between the adoption of digital TV and color TV is tha t digital TV was adopted in the marketplace under government mandates with specific timelines. The government mandates likely contributed to speed up the digital TV transition compared with color TV adoption that the adoption decision was left to the mark et without timeline mandates (CBO, 1999). Government plays an important role in developing and establishing technical standards. The adoption of color TV provides an example of the interplay between standardization by the government and the market driven process (Shapiro & V arian, 1999). While CBS and RCA -the owner of NBC -competed over the color TV standard, the FCC adopted the CBS system as an industry standard in 1950. However, the CBS system was incompatible with existing black and white TV sets w ithout a special attachment. The FCC's endorsement of the CBS system did not actually promote the development of the color TV industry using the standard. Later in 1953, the FCC reversed its earlier decision and adopted compatible RCA system as the color T V standard (Shapiro et al., 1999) In order to promote the adoption of digital TV in the market, RCA shared a great deal of color TV technology after the NTSC color standard was selected Also, NBC RCA s subsidiary -was the main provider of prime time color programming (Farrell, Shapiro, Nelson, & Noll, 1999). RCA invested $130 million in developing color TV by 1959, and NBC showed 4,000 hours of color programming by 1965. On the other hand, ABC and CBS had less incentive to broadcast in color; CBS offe red 800 hours of color, and ABC showed 6 00 by 1965 (Shapiro et al., 19 99 ). ABC
24 company RCA pioneered new color TV standard. While NBC increased its color programming to he lp RCA sell color TV receivers, the other networks were not as supportive of the adoption of color TV. In addition, n etwork affiliates did not adopt color TV immediate ly since the initial CBS system was not successful Figure 1 2 illustrates the adoption o f color TV broadcasting by network affiliates in terms of yearly hours Figure 1 2 Yearly h ours of c olor b roadcasting in m ajor b roadcast n etworks Digital TV Terminology Digital TV a general term for digitally transmitted TV services has more than one meaning First, digit al TV means HDTV (Noll, 2008) which h as higher resolution of pixels and offer s more advanced video and audio quality than that of the analog TV system. Alth ough early HDTV was developed from analog technology modern HDTV is standardiz ed by digital format s Thus, DTV is not ne cessar il y HDTV, which is a particular type of digital TV Second, advanced TV (ATV) is a t erm for digital TV employed by the FCC T he FCC regarded ATV as the digital TV system s that offer improved aud io and video q uality (FCC, 19 93 ) S ince ATV broadly 0 500 1000 1500 2000 2500 3000 3500 4000 Yearly Hours of Color Broadcasting NBC CBS ABC
25 refers to digital enhancement to the NTSC system ATV is a more inclusiv e term than HDTV Digital TV in this study refers to the broad sense of digital TV, which can be used interchangeably with ATV as envisioned by th e FCC Development of Digital TV Discussions of digital TV go back as early as the 1970s when the U.S. Department of Defense developed digital resolution systems during the Cold War (Bates, 2007). The interest in developing digital TV originated from glob al competiti on to establish an interna tional standard for HDTV. C arrying out rese arch under the leadership of individual governments, Europe and Japan were ahead of the U.S. in developing HDTV The analog based MUSE system for HDTV first developed by Japan s NHK used 1,125 lines of resolution (Brinkley, 1997). In the U.S., H DTV was first demonstrated in the U.S. in 1981 ( Dupagne & Seel, 1997 ). Later in 1983, CBS developed analog HDTV system s compatible with the NTSC system (Noll, 2001). As a reaction to HDT V initiatives from Europe and Japan in the 1980s, the U.S. established the development of d igital TV as national agenda to stimulate the electronics and equipment ma nufacturing industry. In 1982, the Advanced Television System s Committee (ATSC) was formed to coordinate the development of ATV standard s to substitute the NTSC standard the analog TV system used in the U.S The original ATSC membership include d representatives from the National Association of Broadcasters (NAB), the National Cable Television Association (NCTA), the Society for Motion Pictures and Television Engineers (SMPTE), broadcast, equipment, motion picture, consumer electronics, computer, cable, satellit e, and semiconductor industries (ATSC.org, 2011). In July 1987, the FCC formed the Ad visory Committee on Advanced Television Service (ACATS)
26 supervised by former FCC Chairman Richard Wiley. The purpose of the ACATS was to offer recommendations for ATV standard (FCC, 1987). The ACATS were represented by the members from broadcast, cable, el ectronics and equipment industr ies, government, and academics (Hart, 2004). O n November 17, 1987, at the request of 58 U.S. broadcasters t he FCC started rulemaking on advanced television (ATV) services by the ACATS The ACATS evaluated t he feasibility of switching to digital TV and competing systems and recommended that the ATSC review proposals on digital TV standards. The opinion s from t he ATSC argued that previous proposals on digita l TV were not appropriate The ATSC s conclusion on previous proposals on standardization encouraged ATV competitors to form the Grand Alliance for developing a new technical standard. The Grand Alliance was created in 1993 consisting of seven digital TV proponents involved in develop ing the digital TV system including AT& T, General Instruments, MIT, Phillips, RCA, Thompson, and Zenith. In 1996 the FCC formally adopted the Grand Alliance standard recommended by the ACATS. Instead of approving one single standard as a final standard for a scanning format, the FCC allowed br oadcasters to use the format that best satis fied their needs. This multi standard decision resulted from a disagreement over the scanning format between the consumer electronics industry and broadcast ing industr y and the comput er industr y The computer in dustry did not participate in the early stage of developing digital TV standards. As the computer industry realized the importance of digital TV standardization m ajor companies such as Microsoft, Compaq, Apple and Intel formed the CICATS (Computer Indus try Coalition on Advanc ed Television Service) to participate in the process of standard setting (Van Tass el, 2001). The computer industry
27 expected the FCC to approve a minimum standard based exclusively on progressive sc anning which would achieve a greate r compatibility with computer screens and improved perfor mance for displaying texts and still images. On the contrary the broadcast industry support ed interlaced scanning that used less bandwi dth and was suitable for displaying moving images. A s a comprom ise to mitigate the conflicting interests of different industries the FCC approved a number of standards including both interlaced scanning and progressive scanning (FCC, 1996). Mandated Transition The establishment of deadlines for moving to digital TV followed a similar multi participant process. First, the U.S. di gital TV transition is a gov ernment mandated transition. Congress structured the transition through l egislative actions and the FCC made ru les for the transition such as timeline s and digita l TV standard s The timeline and the standard setting are two components that formed the mandated nature of transition subject to legislative and regulatory decisions by Congress and the FCC Nevertheless, the mandated transition w as implemented with disc retion on the part of the stations as they met the need s in the marketplace. In this respect, t he process of the digital TV transition show s the interaction between technology, policy and industries. Both private and public entities participated in the pr ocess and the policymaking of the transition : the FCC Congress and the affected industries such as the broadcast computer and electronics manufacturin g This section examines t he key policymakers activities and decisions on the digital TV transition The broadcast policy model developed by Kra snow and Longley suggest s that the FCC, Congress, and regulated industries are key actors as a source of primary influence in broadcast policymaking (Krasnow, Longley, & Terry, 1982) Even though
28 the environment of broadcasting has changed, their approa ch provides a basic framework for analyz ing b roadcast policymaking ( Dupagne & S eel, 1998). Their model is built upon general systems theory, which has been used to e xplain complex phenomenon through describing inter relationship amo ng actors in a system in response to the environ ment. General s ystems theory assumes that an open system continuously interacts with other systems in the environment (Kast & Rosezweig, 1985). The completion of the digital TV transition too k more than a decade from its o nset in 1996. Over the following years the FCC had made rules and Congress had enacted legislation for the transition Since the digital TV transition would affect the future of the broadcast industry and the development of new markets, the comple tion of the transition involved the interests of all stakeholders Also, d igital TV polic y is a result of continued efforts to enforce the mandated transition by policymakers. This section reviews the decisions by the FCC and Congres s to bring about the digital TV transition FCC : Regulatory d ecisions making on digital TV began with its the F irst Notice of Inquiry (NOI) on ATV S ystems on how to foster ATV issued in July 1987 In this NOI, the FCC solicited comments o n digital TV related issues such as bandwidth and compatibility used for digital TV systems. In September 1990, the FCC declared that the new ATV standard must be more than an enhanced analog signal, but be able to provide a genuine HDTV signal in the Firs t Report and Order In addition, to ensure that viewers could continue to receive analog television broadcasts, the FCC dictated that the new ATV standard must be capable of multicasting (FCC, 1990). In May 1992, the F CC suggested a schedule for implementa tion of the transition in the Second Re port and Order. T he FCC adopted a multicast digital system which efficiently uses broadcast
29 spectrum to simplify the channel allocation and consumer choice of digital TV receivers (FCC, 199 2 ). In the Third Report and Order released in October 1992 the FCC revised the transition schedule and specified the requirements for broadcasters. Digital TV broadcasting was able to coexist with the NTSC system, utilizing 6 MHz bandwidths previously allotted for the NTSC system (F CC, 1992). In the 1996 Fourth Report and Orde r, the C ommission adopt ed the ATS standard as technical standard for digital TV After the ATSC standard was approved, the FCC adopted rules to set up the timeline to discontinue the analog TV service. In 1997, the Fifth Report and Order set a 15 year schedule for digital TV rollout The digita l TV schedule required gradual timelines according to the size of TV market the status of the network affiliation and the type of stations. While the Fifth Report and Order set the t ermination date of analog service as 2006, it also clarified that the transition might be left up to broadcast TV stations to meet the marketplace demand In addition, t h e Fifth Report and Order considered digital TV as a technological b reakthrough and reaffirmed the policy of free over the air TV (FCC, 1997) In 2000, t he FCC started to review the progress of the digital TV transition by iss uing periodic Report and Order s Until the cessation of analog service, the FCC promulgated thre e periodic reviews of FCC s DTV rules and policies. T he First DTV Periodic Report and Order in 2001 reduced build out requirements for broadcasters enabling stations to save construction and operating costs. T he FCC announced that broadcasters were to be given additional time to gain experience with digital operation (FCC, 2001). In the Second DTV Periodic Report and Order in 2004, the Commission established a three round channel electio n process through which relocated broadcast
30 stations within channels 2 51 (FCC, 2004). In the Third DTV Periodic Report and Order in 2007, the FCC reaffirmed Februa ry 17, 2009 as the deadline for discontinuing analog service and reported that 1,812 stations received digital TV channels (FCC, 2007). Congress: Legislative d ec isions Congress manifested its support for the development of digital TV when it held its first hearing on digital TV in 1981 ( Bates, 2009) Between 1989 and 1990, n ine bills were introduced to encourage funding and research for d igital TV although none o f them were enacted (Dupagne et al., 1997). Congress established t he legal foundation for the digital TV transition by authoriz ing the distribution of an additional broadcast channel to each station under the Telecommunications Act of 1996. New provisions the rules for granting licenses for digital TV service (47 U.S.C. § 336). The 1996 Act enabled broadcast stations to adopt digital TV by granting exclusive rights to existing broadcast st ations to offer terrestrial digital TV service, while these stations continu ed to offer analog TV service. In 1997, t s transition deadline Although the Act stipulated that stations were to broadcast in digital signa ls by the end of 2006, stations were able to extend the deadline depending on their market conditions. The DTV Transition and Public Safety Act of 2005 set February 17, 2009 as a deadline to terminate all analog TV signals How ever, Congress and the Obama administration had concerns over viewers who remained unprepared for the scheduled analog shut off. The 2009 Digital TV Delay Act was a response to th e con cerns over the lack of public preparedness. T he 2009 Digital TV Delay Act modified the final transiti on date from February 17, 2009 to June 12, 2009 However, stations could choose to comp lete their transition, discontinuing analog broadcasting before June 12 This time,
31 June 12, 2009 became a final deadline and full power broadcast s tations in the U.S. ceased to broadcast in analog signals In 2010, the FCC released the digital channel plan to reduc e TV channel bands f rom 51 to 45 in near future (FCC, 2010). Table 1 1 summarizes key legislative and regulatory measures by the FCC and Congress to impleme nt the digital TV transition. Table 1 1 Digital TV t ransition t imeline : Legislative and r egulatory m easures Measure Time line Summary FCC FCC DTV Review: 5 th Report & Order (1997) May 1, 1999 Required the major four network affiliates in the largest 10 markets to broadcast in digital (30% of viewers) November 1, 1999 Required top 4 network affiliates in the top 11 to 30 DMAs to broadcast in digital (50% of viewers) Congress FCC The Telecommunication Act of 1996 December 31, 2006 Established Decembe r 31, 2006 as the deadline for DTV transition Limited the eligibility for digital TV licenses to existing broadcast stations Analog spectrum are returned to government for auctions and public services FCC FCC DTV Review: 4 th Report & Order (1996) Decemb er 31 2006 Adopted the ATSC standard for digital TV Granted broadcasters $70 billion worth of addition 6MHz to each broadcast station alongside current analog signal until 2006. Mandated that in 2006 all broadcasts must be fully digital FCC The Balanced Budget Act of 1997 December 31, 2006 Approved the analog cut off date as February 17 FCC FCC DTV Tuner Report & Order (2005) March 1, 2007 Required TV reception devices sold in the US to install a digital tuner Congress The DTV Transition and P ublic Safety Act of 2005 (The Deficit Reduction Act of 2005) February 17, 2009 Approved the analog cut off date as February 17, 2009 Directed the FCC to start auctioning the spectrum gained from the DTV transition by January 28, 2008 Authorized the NT IA to establish the converter box subsidy program Congress The DTV Delay Act of 2009 June 12, 2009 Extended the proposed February 17 deadline to June 12, 2009
32 Time line The digital TV transi tion was planned and directed by the government. While differ ent timelines were required to build out digital facilities for broadcast stations working towards the transition, the FCC left the timing decision to broadcast stations so that they might be responsive to market deman d (FCC, 1998). This process is consist ent with digital TV transition being implemented bas ed on a combination of regulat ory and market based approach es The FCC established different timeframes for broadcast stations under different market conditions, requiring an early transition timeframe f or major network affiliates operating in large TV markets. The FCC took into account whether stations were the affiliates of four major networks and operated in top tier markets. This market staggered approach allowed gradual transition by requiring networ k affiliates with ABC, CBS, NBC, and Fox in the largest 10 markets to construct digital facilities by May 1, 1999. In the largest 30 markets, major four network affiliates which covered 53 percent of nationwide TV households were required to construct digi tal facilities by November 1, 1999. For all other full power TV stations, May 1, 2002 was established as a deadline (FCC, 1997). However, by the end of November, 2002, only 16% of the 1,196 stations in the category have full power digital TV facilities. A number of stations that failed to meet the deadline requested the FCC to change the May 1, 2002 deadline (FCC, 2002). The FCC maintained an open stance to grant construction deadline extensions during the early phase of the transition period, especially a fter the 9 11 attack in 2001. This openness allowed many stations to actually start digital broadcasting at the end of 2002. Carter, Franklin, & Wright (2008) a rgue that such postponement reflected the low demand for digital TV sets from consumers due to t he expensive price. In turn, there was a lack of
33 incentives to produce digital TV programming unless consu mers purchased digital TV sets Technical s tandards Digital TV provides better quality video and audio than an analog system Digital TV offers accur acy and efficiency for the same amount of bandwidth to broadcast stations and interoperability with other electronic media. When the need for the transition to digital was examined there was a question about whether the F CC would administer the choice of technical standards or leave the outcome to the m arket (Carter, et al. 1992). The standard setting for digital TV shows that the digital TV transition was a process affected by both government mandate s and market based dec ision s Although the digital TV standard was mandated broadcast stations had the latitude to choose among 18 different scanning formats. T his section reviews what specific digital TV formats were approved by the FCC and what formats are currently employed by broadcasters. In 1996, t he FCC chose the ATSC (Advanced Televis ion Systems Committee) standard based on the technology developed by the Grand Alliance The ATSC standard has 18 different scanning formats, but the formats primarily employed by broadcast stations were 480p, 7 20p, and 1080i scanning formats. The 'p' and 'i' resp ectively stand for progressive and interlaced scanning which differ in terms of the lines that make up each frame of TV video Offering progressively scanned digital images, 480p is the economical method for di gital broadcasting. Broadcasters that provide multicasting choose 480p as their digital TV format (Hart, 2010). 480p is allotted for SDTV while higher resolution formats of 720p and 1080i are designated for HDTV (Carter et al., 2008). 720p provides a highe r quality image than 480p and possibly as
34 high image quality as 1080i because it is progressive. ABC, NBC, and their affiliates employed 720p and made major investments in production facilities for broadcasting in this format. 1080i uses interla ced scannin g which provides a high quality pictures over the formats using progressive scanning CBS and their affiliates choose 1080i, which is costly to produce and display pictures (Hart, 2010) While broadcast stations receive a majority of primetime programming from networks syndicators and advertisers also supply TV progra mming for stations. Therefore, b roadcast s tatio ns should be able to receive programming of any digital format from these different programming sources to convert it to their transmission stand ard Table 1 2 displays popular digital TV formats employed by broadcast TV stations and networks. Table 1 2 Digital TV f ormats in the ATSC s tandard Format Active Lines Horizontal Pixels Aspect Ratio HDTV 1,080i 1,920 16:9 720p 1,280 16:9 SDTV 480p 7 04 16:9 or 4:3 640 4:3 (Source) ATSC (2011). For a network industry where total network size is an important factor in the success of a network standardization of technology particularly inter operability, is one of the important strategic decisio ns to gain competitive advantages and cost reduction (Varian 2004). Standards could be determined by stakeholders such as government, industry a lliances, the market, and industry leader s (Rosen, Schnaars, & Shani, 1988). Rosen et al. (1988) identify the b enefits of standard setting in terms of reducing the risks for consumers and firms, and also i ncreasing competitiveness in international marke ts. Standard setting becomes especially important when com plementary products utilize the new technology. For exam ple, TV programs should be produced in digital formats for broadcast stations to offer digital programming.
35 Th is situation can be characterized as en and caused by the interdependence between different service providers and produce rs (Gupta, Jain, & Sawhney, 1999). In the case of digital TV transition, th e incentives for digital TV adoption are likely to be affected by the choices of other stakeholders. Broadcast stations are willing to adopt digital TV when more consumers have digi tal TV receivers and more studios produce digital programs In turn consumers may want to wait until the pri ce of digital TV sets de crease s while studios are less willing to produce digital programming unless broadcast stat ions adopt digital TV Similarl y stu dios are more willing to invest in digital programming when an established base of consumers and broadcast stat ions adopt digital TV and pay for the programming In turn co nsumers and broadcast stations are likely to wait until there is sufficient d igital programming available so that they benefit from the ir adoption decision In this respect, Congress and the FCC acknowledged that most importantly, the digital TV transition should stimulate consumer demand (Graham, 2003 ). Clearly, s tandard setting is important to innovation adoption in terms of network users' expectations of the likely behavior of other users of the network. Standard ization reduces nce o f investing in an innovation and also assure s consumers that their purchase will not become technologically obsolete Consumer acce ptance is critical to technology adoption in network industries because adoption becomes more valuable as the number of adopters increases. Second, compatibility allows different systems to communicate or interface with competing suppliers of complementary products. As a core function of innovative products compatibility plays
36 an important role in the adopti on process For example, digital TV sets and receivers should share compatible standards for the digital TV transition to succeed (FCC, 2009). Also, Varian (2004) presents three types of competition related to standard setting describing the process for standard establi shment The three types of competition are standard war, standard negotiation, and standa rd leader A standard war refers to a situation where all firms compete to determine the industry standard. A standard negotiation occurs when firms compete within a s tandard, but pref ers the features of their own standard. A standard leader is typically a large, established firm that wants to maintain its proprietary standard, so other firms may want to interoper ate with the existing standard (Varian, 2004). With resp ect to developing standards, researchers have identified three types of standard setting methods (Gandel, 2002 ; Hemphill & Vonortas, 2005) First, an ex post or de facto sta ndard is determined in the market. The PC operating system i ndustry is a good examp le of de facto stan dards ; Microsoft has established de facto standards in software applications such as operating systems and word processors (Lee & Mendelson, 2007). Second, an ex ante or de jure standard, also known as industry standard, are developed an d approved by key participants such as industry groups and professional organizations in a market that seek consensus on standard setting For example, a group of DVD manufacturers led by Warner and Columbia worked together for a single standard and deve lop ed the open DVD standard. Third, a mandatory standard is established by national standards regulators or international standards organizat ions. Although a mandatory standard often makes reference to a de jure standard, a mandatory standard is distinguis hed from a de jure standard by its
37 compulsory natur e as required by l egisl ation (Hemphill & Vonortas, 2005). On the other hand, a de facto standard and a de jure standard are market mediated. A de facto standard is usually selected at a later stage of tec hnology development (Koski & Kretschmer, 2005 ), although a mandated standard has an advantage in that it generally takes less time to be establish ed as a standard The U S digi tal TV standard is a mandatory standard directed by the FCC, while it has a cha racteristic of the de facto standard developed by contending industries (FCC, 1996). Multicasting The digital TV transition offere d broadcast stations techn ical advances that were not possible with analog TV such as mult icasting and ancillary services Multicasting enables TV stations to broadcast simultaneously two or more channels on one digital signal. Ancillary or supplementary services refer to video and data services including subscription p rogramming and data transmission Since ancillary services are non broadcast service t h is study focuses on multicasting which is rel evant to the broadcasting area. Broadcasters expect to generate revenues and extend their position using multicasting. M ulticasting enable s broadcast TV stations to compete with cab le and satellite TV in the marketplace by allowing additional channels for 24 hour programming. As a result, m ulticasting is employed by broadcasters as a major digital strategy to reach niche audiences. Economic i mpact The digital transition was a large scale and complicated process. A major hurdle to digital TV adoption has been the cost issue. The National Association of Broadcasters (NAB) estimated that the TV industry spent approximately $16 billion in equipment and human resources by the end of the digital transition. However, the digital TV transition enabled broadcast stations to multicast at lower
38 resolutions or broadcast in HDTV. B roadcast stations had a choice of transmitting a maximum of five SDTV signals or one HDTV signal within t he 6MHz band widths in addition to existi ng analog channels. Thus, s tations were able to transmit multiple channels sim ultaneously in SDTV, reserving HDTV transmission for primetim e programming (Seel & Dupagne, 2008). Whereas broadcasting in HDTV has advantages in disp laying high quality pictures, multicasting in SDTV can benefit consumers, broadcasters, and advertisers by offering additional channels. Thus m ultic asting can positively impact the operation of TV stations in terms of competition and efficiency. It can e nhance the incentives to compete among stations by increasing program choices for t he same target audience. M ulticasting also can increase potential market supply and choi ce by, for example, expanding advertising slots. Advertisers can have more choices to promote their products on multicast channels at different times. Business m odel s Fo r broadcasters, multicasting had a potential to generate additional revenue streams since stations can operate multiple channels on a single ba ndwidth. Multicasting stati ons expect ed new busi ness opportunities that could expand their economic base and attract more viewers and advertisers. A recent survey from the Harris Corporation shows that 57% of broadcast stations indicate d that multicasting was one of the most viable business opportunitie s for them ( Romano 2008). However u ncertainties remain ed as to whether multicasting was a v iable strategy since broadcasters might need to build or renovate their digital fac ilities which could be costly. Creation of n etwork s The digital TV transition g ave local TV stations new opportunities to extract revenue using multicast channels and offering differentiated
39 progra mming Stations also can utilize digital multicast channels to overcome the insufficiency of major network programm ing. Multicasting enables stations to form new network affiliations specifically aimed at niche audiences in smaller markets (Miller & Prieger, 20 10 ). Repurposing old TV shows is a major programming strategy of multicast networks for attracting niche audie nces and creating a stream of revenue. Retro TV, which has 128 affiliates across the nation, is an example of multicast networks that feed digital subchannels to loca l stations ( Malone 2009). Another example is LATV, a national multicast network focusing on Spanish language programming. Started as a local channel in Los Angeles LATV is expand ing its business into growing markets. To illustrate, Hearst Argyle Television one of the largest TV station group s made affiliation deals to carry mult icast channe ls with LATV in Albuquerque, West Palm Beach, Orlando, and Tampa areas (Karpowicz, 2009) Ma jor networks also operate multicast channels to meet the needs of underserved or specialized audiences and inc rease overall audience numbers As a major broadcast network, NBC was the first in the industry to launch a digita l broadcasting network in 2004: NBC Weather Plus that discontinued the service in 2008. In the same year, NBC bought the Weather Channel that reached 97% of US cable subscribers, from Landmark Co mmunication, and began offering customized digital programming to local stations (AP, 2008). In 2009, ABC created a female targeted digital broadcasting network, Live Well HD Network, focusing on home, health and lifestyle. The network was later renamed as the Live Well Network, and became available on digital subchannels of non ABC affiliated stations own ed by Belo ( Malone 2009).
40 Local p rogramming Broadcast stations can transmit digital TV programming in two ways. First, stations can transmit digital pro gramming from national networks, known as network pass through or ne twork feed ( Romano 2008). Second, stations can transmit their locally originated digital programming. To broadcast their locally originated digital programming, broadcast stations should invest in digital facilities, such as studios and production equipment to create their digital content (Broadcasting & Cable, 2006). Multicasting also helps to create opportunities to expand homegrown local programming, especially for those stations whose local programming was limited to their existing analog channels. A survey conducted by the National Association of Broadcasters (NAB) showed that almost 90% of broadcast stations were inclined to produce local news, weather, and sports for multicas t local programming (NAB .com, 2011 ) Digital Strategies T his study examines digital TV strateg ies of broadcasters in the context of strategic management research. Strategic management research suggests choosing a strategy that can contribute to the firm perform ance (Barney, 1991). A major research area of strategic management focuses on identifying the source of competitive advantage ( Barney 1991 ; Porter, 1980 ). B usiness processes that exploit valuable and rare resources can be a source of sustained competitive advantage (Barney, 1991). Another focus on strategic management is diversification The purpose of thi s section is to give a theoretical contex t to the discussion of digital TV strateg ies for competitive advantage Competitive Advantage A st rategy is a coordinated set of actions designed to ga in competitive adva ntage for an organization ( Hitt, Ireland, & Hoskisson, 200 5 ). C ompetitive advantage is created
41 when a firm has a product or service that is perceived by customers as better than that o f its competitors. Porter (1980) identifies three strategies to gain competitive advantage: cost leadership, differentiation, and focus. A firm can maximize performance by being a low cost producer or by differentiating its products or services. Both the c ost leadership and the differentiation strategy can be accompanied by focusing on a given market segment. In particular, Porter s competitive strategy framework has received scholarly attention when applied to the technology re lated industries. A number of studies (Dess & Davis, 1984; Hambrick 1983) h ave tested Porter s conceptual framework, showing that the profitability of a firm is related to broad ening its market scope, differentiation, innovation, and cost leadership. The relationship between st rategy and t echnology adoption is also emphasized Porter and Miller (1985) suggest that information technology enables firms to advance differentiation or cost leadership. A firm can differentiate itself from others when it adopts a new technology by developing an ability to respond to market changes. In addition, a firm can achieve cost advantage by transforming the value chain w hen technology adoption takes place. Diversification The challenges to broadcast networks were discernible when the shares of network audiences declined. The combined market share from the three major broadcast TV networks, ABC, CBS, and NBC, dropped below 60 % in 1990 (Owen et al. 1994), and then to 44.8% in 2007 (Standard & Poor s, 2007). Competition from cable and satellite TV can lar enhance competitive advantages in the marketplace, broadcast networks develop and implement strategies. Diversification means that a firm operates in different related or unrelated markets which are sources of revenues and profits. In principle, a firm that
42 owns a diverse portfolio of businesses can spread the risk of its investment. Booz, Allen, and Hamilton (1985) define business to ac hieve improved performance and reduce overall risk The authors of t his report further explain that (1) diversification include all investments except those directly supporting the competitiveness of existing businesses; (2) diversification may take the f orm of investments that address new products, services, customer segments, or geographic markets; and (3) may be accomplished by different methods including internal development, acquisitions, joint ventures, licensing agreements (Booz et al., 1985). Sinc e the digital TV transition has completed participation in O TT service can be a strategy for revenue diversification Broadcast networks have more advantages over telecommunication service providers in OTT service since they have more access to diverse programming resources. Chapter Overview This dissertation, consisting of five chapters, proceeds as follows. Chapter 2 ex amines the scholarship on firm behaviors and innovation adoption focusing on factors that influence the timing of digital TV adoption in the broadcasting industry This chapter also reviews the relevant literature for researc h questi ons and conceptual model for the study Chapter 3 describes data collection, measurement of variables, and methods used to conduct the an alysis Multiple regression, independent samples t test, ANOVA, and discriminant analysis are performed and reported Chapter 4 discusses the empirical results of the analyses. Chapter 5 summarizes the findings of this study and discusses theoretical imp li cations, industry implications and policy implications from the findings This chapter also presents the lessons from the U.S. digital TV transition concluding remarks and suggestions for fu rther resear ch
43 CHAPTER 2 LITERATURE REVIEW AN D RESEARCH QUEST IONS Theories of the firm expounded in the studies of economics and strategic managem ent have developed conceptual frameworks for the investigation of innovation adoption. Innovation adoption theories provide a systematic explanation of how innovations are adopted over time at different levels. This chapter reviews prev ious studies relevant t o technology adoption such as digital TV, and identifies the factors employed in innovation adoption studies Using past literature as a foundation this chapter devel ops research questi ons. Theories of Firm Behavior Theories of firm behavior explain why certain firms carry out business activities more efficiently than others. The i ndus trial o rganization tradition and the strategic management tradition are two major re s earch streams for studying firm competitiveness. The industrial theor ies focus on industry characte ristics while the strategic management theories emphasize internal firm attributes that may lead to competitive advantage. The industrial organization theor y (Scherer, 1980) provide s the theoretical foundation for this study Industrial Organization T radition The industrial organization tradition has offered the conceptual basis of exploring firm behavior by em phasizing the structure of the industry in whi ch firms operate The main focus of the industrial organization analysis is that market condition s and market structure primari ly determine behaviors of firms (Scherer, 1980). The industrial organization analysis is based on the model that a given industry can be characterized by its market struc ture, conduct, and performance (S C P). The S C P model illustrates
44 e of the firms in the industry. Although mainstream industrial organizat ion study is founded o n the S C P model, the model has been criticized mainly because of its static framework. The modif ied version of S C P model has dynamic elements including feedback loops, added to the traditional model. The modified model dismiss es t he unidirectional flow of relationships among links among structure, conduct essential to the static framework (Tirole, 1988). The application of game theo ry and the recognition of factors affecting market c oncentration indicate that conduct and performanc e can affect market structure, as market structure on conduct (Audretsch, Baumol, & Burke, 2001). The industrial organization perspective offers a systematic framework for studying the behavior s of media firms. Wirth and Bloch (1995) review the media econ omics literature that uses the indu strial organization theory suggesting that media scholars apply the industrial organization theory to analyze the link between market structure and conduct, and performance of media firms. Moreover, t he industrial organi zation approach can provide a useful theoretical framework to investigate innovation adoption by media firms In studying adoption in media industries, factors associated with market structure, such as the growth of demand, market competition, product diff erenti ation, or vertical integration are examined in relation to adoption decisions (Albarran, 1996). Strategic Management Tradition S trategic management is a study of organizational processes: analysis of strategic goals and internal and external enviro nment, decisions on competition and operation, and actions of implementing strategies. The essence of these processes is to create sustainable competitive advantages in the marketplace (Dess, Gregory, Lumpkin
45 & Taylor, 2005). Within the tradition of strat egic management, two mai n approaches exist ; both are based on the source of competitive advantage. One o f the streams is the market based view derived from the in dustrial organization theories focuses on industry structure in the external environment of the fi rm. For example, Porter (1980) des cribe s market forces in terms of barriers to entry, buyer power, seller power, and availability of substitutes that determine competitive rivalry in a market Another major stream of research is t he resource based v iew (RBV) emphasizing the role of firm level resources as a source of competitive advantage and relates firm perfor mance to competitive advantage (Rumelt, 1991). The resource based view explores why a firm outperforms others and attempts to specify the con ditions under which firm resources can be used to create competitive advantage. Porter (1985) emphasizes focusing on which primary and supporting activities should be done internally and which should be outsourced or shared with oth er organizations for competitive advantage. Barney (2001) identifies firm resources as all tangible and intangible assets, capabilities, organizational processes and attributes a firm uses to choose and implement its strategy Resources include a number o f factors such as plant and equipment, geographic location, knowledge and skills of employees, corporate culture (Barney, 1986; Castanias & Helfat, 1991; Fiol, 1991; Rumelt, 1987). These firm resources can be largely classified into three categories: physi cal capital resources, human capital resources, and organizational capital resources (Barney, 1991; Barney, 2001). Physical resources include physical technology used in a firm such as plant and equipment, geographic location, and access to raw materials. Human capital resources are the
46 training, expertise, experience of individual employees and managers in a firm. systems as well as informal inter firm and intra firm rela tions (Barney, 1991). Heterogeneous resources are the basic concept for c ompetitive advantage of a f irm. The resource based theory suggests that f irms are heterog eneous in their capacities and each firm has unique tangible and intangible resources develop ed and expanded over time (Barney, 1991). A firm that has its own innovation strategy as competitive advantage can be more responsive to the needs of consumers and the market and thus gain competitiveness over its rivals. Porter (1985) supports this point by saying Aisbett (2002) examine Porter s view by testing the relationship between strategy and innovation adoption, showing that innov ation adoption occurs as a result of the employment of new competitive strategies. Theories of Innovation Adoption The tradition of innovation adoption and diffusion is one of widely employed theoretical foundation to study social and economic changes (Ro gers, 2003). This s ection reviews the conceptual def initions of adoption innovation, and diffusion, and literatures on the patterns of innovation adoption and diffusion. Then this section explores the characteristics of digital TV as innovation adoption b ased on its advanced technological features. This study also reviews different levels of approaches in innovation adoption studies : the individual level and organizational level approaches.
47 Concepts of Adoption Innovation and Diffusion Scholars have defi ned innovation and adoption in various ways. Rogers (2003) c Dasgupta, Granger and McGarry ( 2002) define technology the decision to accept, or invest in, a technology New technology adoption is associated with sunk costs and adoption decision by firms, according to Hall and Kahn (2003). First, adopting a new technology means giving up the fixed costs of adoption because most of the costs associated with adoption may not be recovered. Second, firms have a choice to delay adoption or adjust the level of adoption when they are unce rtain about the benefits of a new technology. To clarify the concept of adoption this study compares adopt ion and its closely related concepts such as diffusion and innovation. Schumpeter (1950 ) identifies that technological change involves five different phenomena: invention, development, entrepreneurial function, investment, and diffusion. He argues that al though innovation is a construction of new plants and equipment, introducti on of new firms, and the leader ship of new people, it is also a recombination of prior skills and resources Schumpeter (1950 ) underscores the importance of technology by differenti ating the concepts of invention, innovation, and diffusion. He explains that invention refers to the generation of new ideas, while innovation occurs when those new ideas are embedded in new commercial product Rogers (2003) defines pra ctice, or unit of adoption. D iffusion of innovation theory suggests that an innovation spreads members of a system, usi ng diverse channels over time. He suggests that a dopti on occurs at the decision phase of the innovation p rocess (Rogers, 2003).
48 On the other hand, d iffusion is a process by which an innovation becomes widespread across the market. Therefo re, innovation and diffusion occur at different stages of new technology adopt ion (Schumpeter, 1950 ). Innovation occurs wh en a new product becomes available on the market, while diffusion occurs when such innovation is available to individuals or firms for use. I nnovation refers to any practices that are new to systems in terms of equipment, structure, design, products, polic ies or processes ( Eveland 1979; Damanpour, 1991; Kimberly & Evanisko, 1980). King and his colleagues (1994) explain that i nnovation is the process which enables inventions to be used, while d iffusion is a social phenomenon that increases a capacity to use innovation A s a major catalyst to achieve competitive advantage, an innovation relies on its technological and industrial dimensions (Anderson & Ortinau, 1988; Swan & Newell, 1994). Braun and MacDonald (1994) conceptualize innovation as a chain process w hich consists of technology, entrepreneurship, management, a social need, and a n environment supportive to a chain reaction of innovation. Although the concepts of adoption, diffusion, and innovation have overlapping qualities and have been used interchang eably in previous studie s, this current study differentiates adoption and diffusion studies, focusing on differ ent time orientation in research. Adoption studies tend to seek the drivers of adoption at certain points in time while diffusion studies genera lly emphasize the behaviors of the diffusion process over time. Therefore, while the diffusion analysis us es aggregate adoption models, the adoption analysis focuses on apparently independent decisions at the consumer or the firm level. Since diffusion is considered a more continuous and aggregated process than adoption diffus ion studies are considered as an example of process theories, the
49 successions of events that lead to social phenomena. In contrast, adoption studies reflect the characteristics of the variance theory that examines variables affecting the choice by individuals and organizations (Thirtle & Ruttan, 1987). Patterns of Innovation Adoption and Diffusion Focusing on the differences between adoption and diffusion, Hoppe (2002) suggests an S sh ape d diffusion curve. Unlike the instantaneous nature of adoption which occurs immedi ately, the diffusion path of a firm takes an S shape d form that shows the presence of early adopters and late adopters. A diffusion curve show s that the rate of adoption s tarts slowly but the rate increases rapidly unti l it reaches a saturation level In most cases when a firm completes innovation adoption, the rate of adoption slows down (Stoneman, 2002). Emphasizing unequal adoption patterns among firms, Hoppe (2002) arg ues that innovation adoption is delayed at an initial stage because firms are uncertain about the value of adoption and a new technology. The theory of industry life cycle explains how firms react to innovation adoption and diffusion in the marketplace fro m an evolutionary perspective (Nelson & Winter, 2002). According to the industry life cycle theory, initially, it is uncertain how innovation can improve a product and what consumers want truly value When firms adopt or drop on innovation, a dominant desi gn emerges that can be conducive to innovations of certain firms but give potential entrants disadvantage s The industry life cycle suggest s that the emergence of a dominant design not only causes a small number of large firms to prosper but also eventuall y reduces entry of new competitors in an industry (Nelson et al., 2002).
50 Digital TV as Innovation Robertson (1971) produces a framework for classifying innovations focusing on their effects on established patterns. The three cate gories of innovations are: a continuous innovation, a dynamically continuous innovation, and a di scontinuous innovation. A continuous innovation is a b asic extension of an existing product from which consumers feel the need for only a slight change from their current behavior practi ces. In a continuous innovation, firm alters a product rather than estab lishing a new product. A dynamically continuous innovation occurs when consumers perceive an innovation as a new product but not as a major technological advance. Although a dynamicall generally chang e the existing market structure A discontinuous innovation occurs when an innovation is perceived by consumers as be ing a new product reflecting a major technological advance. A discontinuous innovation represents a major change in the benefits offered to consumers and in behaviors necessary for them to own and use the product. There fore, consumers discontinue their past behavior patterns in order to fit the new product into their lives. Digital TV may be considered to be a discontinuous i nnovation because of it s advanced technical features and capabilities to present new viewing experience, offering HDTV and interactive services (Anderson & Ortinau, 1988). Tushman and Anderson (1986 ) suggest that technology evolves through additive change punctuated by technological discontinuities that increase the env ironmental uncertainty facing firms They conceptualize patterns of technological change as competence enhancing and co mpetence destroying. A competence destroying innovation is drastically different from older technologies, requiring new skills, abilities, and knowledge that create new products or substitutes for old products. A competence
51 enhancing innovation is mainly i nvolved in product and performance improvement s which build on existing know how and does not necessarily give up prior skills required for older technologies. A competence enhancing innovation is initiated by established firms in most cases because such i nnovation tends to consolidate industry leadership ( Tushman & Anderson, 1986 ). This study looks at digital TV as both as a competence destroying and a competence enhancing innovation because the switch from digital to analog has a feature of both a substit ution for and an improvement of analog TV. While t he business of digital TV requires new operation and production skills, it is still built on skills and know how that have been used in analog TV broadcasts such as scriptwriting, filming, packaging and m arketing. Innovation Adoption and Diffusion Theories This study draws on innovation adoption and diffusion theories that provide a theoretical framework for research in several aspects. First, theories of innovation adoption are useful in pursuing interdis ciplinary studies, since the theories ha ve been applied to a vast array of disciplines, such as agriculture, marketing, education, economics, public health, and mass communication. Diffusion of innovations research goes bey ond a single discipline, requirin g researcher s to conduct comprehensive research on economic and social cha nges. Since this study lies at the intersection of mass communication, management, and technology, i nnovation adoption theories provide the foundation for this study. Second timing is an important element in innovation adoption and diffusion studies. Timing is a factor that differentiates macro level adoption studies from other micro level studies (Katz, Levin, & Hamilton, 1963). For example, the Ryan and Gross study highlig hts the s ignificance of time, confirming the presence of the S shaped curve (Ryan et al., 1943). In a typical S shaped curve, the
52 number of adopters increases at a fast rate at the early stage of diffusion, and rises incrementally until it reaches an inflection poi nt Since this study explores digital TV adoption over time, the S shaped curve in innovation adoption can capture the importance of time factor in digital TV adoption. Third theories of i nnovation adoption are useful in studying interactive innovation a ssociated with net work effects (Rogers, 2003). In interactive innovation, adoption decision in a network tends to be affected by the decisions by other potential adopters and interest groups since the increase in the number of adopters can increase the v a lue of the innovatio n Digital TV adoption is related to the context of interactive innovation in that the adoption of digital TV by broadcast stations and consumers do not have much value unless program providers produce and distribute sufficient digital TV programs in the market. F ourth this study uses the concepts of innovation to understand digital TV adoption. Digital TV is considered a type of in novation in this study and is consistent with innovation categories suggested by sch olars. Digital TV is a type of disruptive innovation (Robertson, 1971) because digital TV has improved features over analog TV, such as offering superior picture qualities and interactive TV services. Digital TV also has both competence enhancing and competence destroying (Tuc hman & Anderson, 1986) innovation ; although digital TV requires new skills to operate digital facilities and cameras, these new skills are more easily acquired if a station h as prior experience in operating analog TV systems Levels of Innovation Adoption Studies Innovation adoption studies can be based on different perspectives depending on the levels of analysis: the different perspectives are the demand side approach which explores in dividual adoption and the supply side approach which explores
53 organ i zational adoption (Corrocher, Fontana, & Parlanti, 2009; Damanpour, 1992). The individual level and the organizational level approaches focus on different types of determinants of innovation adoption. For the individual level approach, non economic factor s, such as persuasion and self confirmation, are taken into account in explaining individual adoption decisions (Chan Olmsted, 2006; Verkasalo, 2009). By contrast, the organizational level approach typically explores how firms adopt innovations, using macr o corporate factors including firm characteristics and market characteristics. Innovation adoption by firms can be influenced by interactions of individual level and organizational level factors (Hoppes, 2002). For example, innovation adoption in the pres ence of network externalities can be influenced by individual demand which depends on how many consumers have adopted innovation, as well as organizational factors (Farrell & Saloner, 1985). Since this study emphasizes digital TV adoption by broadcast stat ions, relevant studies on innovation adoption by firms are examined in the literature review. Innovation adoption decisions by firms are important in that an innovative technology has the potential to bring in additional revenue and enhance competitiveness of firms (Gatignon & Robertson, 1989). Individual level a pproach Inno vation adoption studies using the individual level approach emphasize consumer characteristics including demographics, socioeconomic status, and media consumption behaviors as predictors of adoption (Dupagne, 1999; Lin, 1998). Studies using the individual level approach rely on the diffusion of innovations theories (Rogers, 1983) and Technology Acceptance Model (TAM) to explain how and when consumers decide to adopt new technologies (Davi es, 1989). TAM is built upon the theories of reasoned action and planned behavior, explaining
54 perceptions determine the implementation of certain behaviors For example, Papies and Clement (2008) use the theory of planned behavior to explain consumer intention to adopt movie download services, employing non economic variables such as attitude, subjective norm, lack of volitional control, innovativeness, past behavior, planned usage, and price consci ousness They find that i nnovative consumers are influenced by social and technological factors, such as social influence and compatibility, in addition to attitudinal factors. Organizational level a pproach The i nvention of innovation has little impact on economic growth until suf ficient consumers and firms adopt the innovation. Although adoption decisions are made both on individual and organizational levels, innovation adoption at the firm level mostly determine s the benef its and cost s associated with adoption (Hall & Khan, 2003) Scholars use diffusion of innovation theories to explain innovation adoption by firms because the theories can offer the framework explaining how new technologies are adopted from the supply side perspective Adoption of new technologies at the firm leve l can be examined in two ways: intra firm diffusion and inter firm diffusion. Intra firm diffusion concerns the number of firms adopting new technologies or the information spillover within a firm. On the other hand, inter firm diffusion considers the inte nsity of innovation adoption, explains the differences among firms (Hollenstein, 2004; Karshenas & Stoneman, 2002). Studies in inter firm diffusions use the rank model, which is most commonly used in empirical studies on technology adoption (Karshenas et a l., 2002), and enable researchers to investigate the characteristics of the firm and market structure. Intra firm diffusions are supported by the epidemic model, which describes the possible influence of network externalities on technology adoption. In the
55 epidemic model, the current level of diffusion in the entire economy or the proportion of Feder, Just, and Zilberman (1985) suggest that individual adoption models assume that only one te chnology is considered for adoption decisions. Aggregate adoption is mea sured by the aggregate levels of new technology use within a giv en geographical area or a population. Thus, the level of adoption at the individual firm level is dichotomous, while the aggregate firm measure of adoption becomes continuous. Saha, Love, and Schwart (1994) emphasize the time factor in the aggregate nature of diffusion throughout an industry, as opposed to static behavior s of adoption at the individual firm level. Metcalfe (1995) differentiates the two traditions of studying technology adoption which comes from uncertainties about firm behavior : the equilibrium perspective and the evolutionary perspective, which depend on different economic rationales. The equilibrium persp ective relies on classic economic views profit maximization behavior In contrast, the evolutionary perspective posits that technology change depends on more t han th e behavior of each firm since it differentiates Majumdar and Venkataraman (1992) find two major research trends in adoption studies: the game theory approach and the determinants approach. In the game theo ry approach, scho lars are interested in adoption timing as to whether a firm should adopt now or later (Dasgupta, 1968; Reinganum, 1981). The determinants studies on the inter firm differences, generally explain firm size and profitability of new tech nologies as factors t hat influence the level of adoption (Mansfield, 1968). This study uses the perspectives from the determinant
56 studies because this study focuses on digital TV adoption decisions by broadcast stations unde r different economic conditions. Innovation Adoption by Firms When firms adopt innovation, the adoption consists of three different phases: (1) collecting information prior to adoption, (2) deciding on adoption versus non adoption, or whether or not to adopt the technology, and (3) deciding on the timing of adoption, or when to adopt innovation (Dong & Saha, 1998; Saha, Love & Schwart, 1994). Among the three phases in adoption decision, adoption timing decision is important in studying the digital TV transition because all full power stations were required to adopt digital T V within a specific time frame. The second phase of the decision on whether o r not to adopt is not significant for the digital TV transition, considering that all full power stations had no choice but to convert to keep their business I n the digital TV transition, g overnment exerted a regulatory power o n the transition process by establishing requirements and guidelines for the transition including the timeline s for the transition With respect to the timeline s b roadcast st ations were given flexibility to choose the timing of adoption Therefore, d eciding on the timing of adoption is an important strategic decision making for broadcast TV stations despite the mandatory timelines for digital TV adoption. Timing of Innovation Adoption S chumpeter (1934) the founder of innovation research, acknowledges the importance of th e timing of innovation adoption noting that a rational firm is unlikely to readily adopt new types of production because of capital losses. The ories of innovation adopt ion hold that technology based innovations are eventually adopted by everyone in the market (Rogers, 1967). Also, scholars emphasize timeframes of adoption when
57 identifying the process of innovation. Previous empirical studies ( Griliches, 1957; Stoneman, 1 983) suggest that when firms adopt new technologi es in a delayed manner, these technologies diffuse over time Researchers also explain that the timing of innovation adoption can rely on strategic considerations (Jensen, 1982; Reingaumn, 1981) and market s tructure (Kamien & Schwartz, 1972; Reingaumn, 1981). When two firms employ the same technology for production, new technology adoption can give a competitive advantage to the firm that adopts the technolo gy before its competitor. If a firm delays adoption, it can incur low technology costs due to economies of learning when they do eventually adopt the technology (Milliou & Petrakis, 2009). Early Adopter s Identifying the charac teristics of early adopter s at the firm level has been a major area of the timing of innovation adoption. Hollenstein (2004) explains the factors determining the timing of technology adoption based on inter firm differences in profitability potential. His study uses regression analysis to analyze survey data on the Swiss ICT (Informati on and Communications Technologies) sector. In examining the level of technology adoption Hollenstein uses the measure of overall ICT level defined as the number of ICT elements and the proportion of employees in the technology sectors. Ci arli and Rabell otti (2007) explore the relationship between the rate of adoption and ICT use in the textile enterprise in Italy to find out the influence of ICT on three types of adoption: product innovation process innovation and organizational innovation. Gatignon an d Robertson (1989) examine four categories of variables in the decision process of technology adoption in the high tech nology i ndustry. They suggest that firms most likely to be responsive to adoption are those in the industries with high co ncentration rat ios. In their study, firms are categorize d as adopter s rejecter s or
58 undecided In this category, adopters show a higher level of information process ing ability than non adopters. C hen (2005) studies the level of innovation adoption in the Taiwanese petr oleum refining indust ry from 1980 to 1989 He find s that both supply side and demand side factors, such as refinery size and consumption growth, tend to influence the different levels of adoption more than an institutional factor, which is regulatory subsi dies to refineries. Determinants of the Timing of Adoption Since the timing has precise measures an investigation of the timing of adoption remains a pertinent area of innovation adoption studies. Previous studies on innovation adoption focus on the condi tions of adoption which can lead to differen t decisions by firms This study examines the factors found to be relevant to innovation adoption stud ies including firm characteristics, market characteristics, and market demographics. The industr y organizatio n approach suggests that market structure chara cterized by various firm specific and market specific factors can explain market (Scherer, 1979). Dasgupta and Stigliz (1981) emphasize the importance of exogenous conditions for inn ovation in a study of the relationship between the structure of an industry, such as the number of firms and the degree of concentration, and the nature of innovative activity. Firm Characteristics It has been a cknowledged that adoption decisions by firms can be largely explained by firm characteristics and market characteristics (Gomez & Vargas, 2009). Firm characteristics include ownership, horizontal integration, and vertical integration according to previous literature In addition many i nn ova tion ado ption studies have examined the rel ationship between firm size and timing of adoption. L arge firms are
59 generally more success ful in early innovation adoption because they tend to have e conomies of scale and initiate competitive actions as a first mover (Th irtle et al., 1987). A greater resources and more skilled employ ees of large firms can also contribute to the timing of innovation adoption (Damanpour, 1992 ). On the other hand, H age (1980) and Dana ( 1994) find that small firms are in a better posi tion for innovation adopt ion because they tend to be more entrepreneurial and innovative in their business. T his study however, does not employ firm size as a factor in order to avoid an overla p between independent variables. This study uses DMA marke t rank and the number of TV households in a station s market to measure market size which embeds the characteristics associated with firm size Since this study assumes that the inclusion market siz e can explai n the variability of firm size only market s ize is used in this study. Ownership Concentrated ownership i s deemed important in supporting economies of scale Scholars (Hitt, Hoskisson & Ireland, 1994; Teece, 1996) suggest that concentrated ownership can provide firms with incentives to invest in and support innovation and the ability to cope with uncertainty. If th is proposition is applicable to the broadcast industry stations may largely benefit from concentrated ownership. For example, n etwork affiliates and TV group stations can receive TV pro gramming and expertise from their affiliated networks or their TV groups. Compared to independents, a ffiliates and stations owned by group owners can obta in progra ms purchas e equipment, or receiv e programming and engineering expertise O n the contrary Be sen & Johnson (1984) suggest that there is no evidence that group owned stations bargain more effectively in their program dealings with networks or program suppliers. In a study of digital radio adoption Borrell (1997) uses group ownership as an independ ent variable to predict the
60 types of radio stations that offer satellite delivered programming In this study, a group owner is defined as an entity that owns stations in multiple markets across the nation. This study finds that stations owned by corporati ons are more likely to adopt satellite delivered formats. Based on the importance of ownership in TV stations suggested by previous literature, this study poses the following research question. RQ1: What if any, influence d oes ownership have on the timin g of digital TV adoption by broadcast stations ? Horizontal i ntegration The effect of horizontal i ntegration on firm behavior has been an important discussion in innovation adoption studies Horizontally integrated TV group increase its viewership to comp ete for local advertising with other broadcast stations or cable systems in the market (Chipty, 1995). Robertson (1995) suggests that more horizontal integration is likely to increase the probability of adoption. GAO (2002 b ) reports that over 79% of curren t digital TV stations based primary funding for constructing digital TV facilities on their owners or parent firms Based on previous literature this study poses the following research question RQ 2 : What, if any, influence d oes horizontal integration h ave on the timing of digital TV adoption by broadcast stations ? Vertical Integration Vertical integration is defined as the ex pansion of a business by gaining control of operations fro m the acquisition of raw materials to the sale of the final product. Gr ossman and Hart (1986) define the essence of vertical integration from the ability to exercise contro l over decisions within the production process A vertical integrated firm control s related businesses which can help each other through conne ctions inclu ding
61 quality control, input supply, and complementary products I n the media industry, a fu lly vertically integrat ion would encompass the production, the distribution, and the exhibition of programs. V ertical integration in broadcast TV stations generally exists when they have an ownership of or an affiliation with program providers, such as production compan ies, distribution compan ies and broadcast network s Previous literature suggests that vertical inte gration can bring advantages to firms that adopt in novations. The motivations for or benefits from vertical integration, such as reducing transaction costs and providing economies of scale and scope can encourage innovation adoption (Blackman, 1998; Yoo, 2002). In addition to the impact s on individual fir ms, v ertical integration can have a welfare enhanci ng impact on the overall market such as lowering prices and improving product quality, an d benefit consumers (Chipty, 1997; Yoo, 2002). Previous studies (Anderson & Schmittlein, 1984; Yoo, 2002) indicate that additional distribution channel s created by vertical integration can offer an advantage to vertically integrat ed firm since such firms can maintain control over innovation adoption On the other hand, another stream of research (Oster 1982; Nadler & Tushman, 1987) suggest s that vertically integrat ed firms that do not have synergies may delay innovation adoption due to the uncertain ties associated with the adoption Boeker and Huo (1998) examine the effects of vertical integration on the rate of adopt io n in the computer industry find ing that the degree of vertical integration increase s the rate of adoption Armour & Teece (1980) suggest that vertical integration of a firm. Borrell (1997) uses the network affiliate relation as an independent factor in studying the adoption of satellite delivered programming by 503 radio stations He finds
62 that less vertically integrated stations are more likely to adopt satellite delivered programming Vertical integration includes vertical ownership and vertical contracts ( Mortimer 2008). Network affiliation, as a type of vertical contracts, is an essential content linkage to broadcast stations through which they have an access to network programming In network affiliation, a network contractually affiliate or franchise with a local station, with an agreement to provide the station with a schedule of television programs. Network affiliation has been frequently discussed in relat ion to vertical ownership between major broadcast networks and major content producers. Previous media studies (Goolsbee, 2007; Waterman, 2000) indicate the possibility that vertical integration can reduce competition in the market Since v ertically integr ated stations are more likely to carry their affiliated networks and more networks overall, but are also mor e likely to drop rival networks Network affiliates can be a part of vertically integrated stations when they are owned and operated by major networ ks, and such network affiliates may take advantage of benefits of vertical integration through contractual arrangements. Since network affiliates in this study are not confined to major network affiliated O&Os, network affiliation simply represents a stati on s access to the programming of affiliate ne tworks. Table 2 1 illustrate s the programming sources that a major network affiliate can receive from affiliate contracts. Table 2 1. Programming s ources of n etwork a ffiliates Owner Revenue Studio (Productio n) Network (Distribution) Disney $38.1 Walt Disney Pictures Touchstone Television Miramax Film Corp. Pixar Walt Disney Feature Animation Buena Vista ABC Network
63 Table 2 1. Continued Owner Revenue Studio (Production) Network (Distribution) Viacom $ 9.34 Paramount Pictures Paramount Television Group MTV Films MTV Productions Dreamworks Studios Nickelodeon Movies Nickelodeon Studios Wilshire Court Productions Spelling Films CBS Production Eye Productions King World Productions BET Pictures CBS Network GE $ 150.21 Universal Media Studios Universal Features Focus Features NBC Network News Corp. $32.8 20 th FOX Century Film Corp. FOX 2000 FOX Searchlight Pictures FOX Atomic 20 th FOX Animation Fox Network *Year 2010, in billions (Sources) Data collected from corporate websites Based on the previous studies confirming the importance of network affiliation as an essential element of vertical integration in the broadcasting industry this study examines network affiliation as a factor that influence s the t iming of digital TV adoption. RQ 3 What, if any, influence does vertical integration have on the timing of digital TV adoption by broadcast stations? RQ 3 1 Is there a significant difference in the timing of digital TV adoption between network affiliate s and independent s ? RQ 3 2 Is there a significant difference in the timing of digital TV adoption a mong ABC affiliates CBS affiliates, NBC a ffiliates Fox affiliates, other affiliates, and independents?
64 Market Characteristics Since innovation adoption ca n be a result of interactions of various factors in the market, market characteristics can determine the success of innovation adoption (St oneman & Diedren, 1994). Market characteristics including competition, market s ize, and growth are unlikely to be con troll ed by management of an individual firm (Frambach, 1993). Schumpeter (1947) suggest s that large firms with market power are more the Schumpeterian argument is supported, large scale firms would have more incentive to innovate because of the need to outperform rivals with new and improved technology. Scholars have provide d some reasons why market power might have a positive effect on innovation adoption First, m onopolists often have resources and capabilities sufficient to innovate efficien tly and increase profits T hey usually have economies of scale and ownership concentration so as to diminish unit costs of production. Second monopolists can build barriers to entry that protect their position in the future. Third, monopolists can invest additional profits to invest into new technologies to adopt innovation. Fourth the scale economies of large firms allow these firms to diversify the risks of adopting innovations and offer scope economies to be tter position themselves for R&D (Scherer, 1980). While research efforts show that mark et power and innovation are positively related other research ers f ind that large firms with market power have less incentive s t o innovate. Arrow (1962) finds that a co mpetitor can profit more than a monopolist from innovation. Geroski (1990) concludes that monopolists are not particularly innovative, and small firms seem to stimulate innovative activities. Similarly, Nelson (2002) argue s that a al structure may deter innovations
65 Competition The competitive environment of broadcast stations consists of the competition within the market a nd from competitors in other media markets one of which can be cable TV. Previous studies frequently use c om petition adoption. For example, Hoppe (2002) suggests that t he incentive s for adoption are likely to be higher for firms in competition because they have an incentive for preempting other competitors. In the digital TV transiti on, whether sufficient number of consumers exists in the market also need to considered because the level of technology adoption by stations also depends on consumers who are able to receive digital TV signals. The two sided features of the broadcasting ma rket are related to network externalities (Rochet & Tirole, 2006). Adoption literature widely supports the influence of c ompetition on technology adoption by firms Horsky and Simon (1983) assume that the existence of competition may affect firm behaviors because firms consider the effects of competition when they adopt new technology. They find that firms invest in advertising heavily i n earlier period of adoption and then decrease advertising expenditure over time. Hannan and McDowell (1987) find that t he competitive market environment influence d techno logy adoption decisions in the banking industry Milliou and Petrakis (2009) find that competition increases technology adoption when new products have sufficiently close substitutes. Reinganum (1981) sugg ests that a change in market concentration may change the rate of technology adoption if firms are fully commi tted to their final adoption dates. Based on previous discussions, this study formulates the following research question.
66 RQ 4 : What, if any inf luence, does competition have on the timing of digital TV adoption by broadcast stations? Market s ize Acemoglu & Linn (200 4 ) investigate the effects of market size for different types of drugs on entry of new drugs and innovation. In broadcast TV studies DMA (Designated Market Area) rank has been generally used as an e stablished measure of the TV market where broadcast services of similar stat ions reach the same population A total of 210 DMAs exist in the U.S. broadcast market; the largest DMA is New Yor k, which covers 6.5% of all TV households and the smallest is Glendive in Mo ntana with 0.003% of coverage ( N ielsen.com 2010 ). The FCC (1999) established the phase in transition policy based on the TV market size of broadcast stations. Since evidence in pr evious literature supports the relationship between market size and innovation adoption, this study examines the influence of market size on timing of digital TV adoption Therefore, this study poses the following research question. RQ 5 : What, if any, in fluence does market size have on the timing of digital TV adoption by broadcast stations? Market Demographics This study employs market demographics as a factor to examine what differences in income and population density affect timing of digital TV adopt ion. In addition to firm characteristics and market characteristics, scholars expect that local demand conditions such as income and population density are related to innovation adoption ( Kelly & Helper, 1999; Majumdar et al., 1998 ). B roadcast stations re venues largely depend on the s upport from advertisers whose main concern is household demograp hics of th e TV market. I n predicting digital TV adoption by individuals c onsumer d emographics have
67 been employed as an important factor (LaRose Neu endorf, & Jef fres 2003 ). Cerda (2003) investigates the effects of market demographics on innovation in the pharmaceutical industry Ford, Koutsky, and Spiwak (2007) find that br oadband adoption is more closely related to income than population density. The relationshi p of market demographics to timing of digital TV adoption may not be clear as it has shown in consumer adoption studies, because of the mandatory nature of the digi tal TV transition Nevertheless, i ncome and population density in a station s market can est ablish the baseline for a station that may show different timing decisions in digital TV adoption. Thus, t his study employs market demographic s to examine their influence on the timing of digital TV adoption RQ 6 : What, if any, influence d oes income have on the timing of digital TV adoption by broadcast stations? RQ 7 : What, if any, influence does population density have on the timing of digital TV adoption by broadcast stations? Early Adopters and Late Adopters This study examines the differences between early adopter s and late adopter s of digital TV Adoption usually follows the S shaped curve, which is characterized by slow adoption in the beginning, followed by an acceleration process, and a gradual slow down at the end (Geroski, 2000; Stoneman, 1984). Mansfield (1961) observes the patterns of diffusions in four different industries in discuss ing the S shaped innovation process. R ogers (1995) suggests that most innovations take an S shaped innovation pattern. As such the S curved process characterizes decision s to adopt an innovation or to delay its adoption. Based on the theories of innovation adoption this study poses the following question.
68 RQ 8 : What factors explain differences in the timing of digital TV adoption betw een two groups of stat ions: early adopters and late adopters? Table 2 2 Literature on determinants of innovation a doption Factor Author Measure Research D esign/Analysis Key Finding s Firm size Damanpour (1992) Personnel (Number of employees) Non personnel (capacity, volu me, finance) Meta analysis of 20 published innovation studies Firm size and innovation have a positive association. Firm size has been used as an effective moderator variable. Gomez & Vargas (2009) Number of employees Secondary data analysis using da tasets drawn from annual surveys. Large firms tend to adopt innovation more quickly than small firms. Firm type Damanpour (1992) Manufacturing or service Meta analysis of 20 published innovation studies Type of organization is significant as a moder ator in measuring relationship between firm size and innovation. Competition Gomez & Vargas (2009) Market share of the four largest firms Secondary analysis using datasets drawn from annual surveys. The role of competition is not confirmed. Kraft (1989 ) Number of competitors in the market where a firm operates Secondary data Multivariate regression Competition is likely to be a significant factor of innovation. Market size Acemoglu & Linn (2003) Total expenditure of consumers Multiple regression Sign ificant effects of market size on innovation Income Ford et al .(2007) Average household income Multiple regression Income has a positive effect on broadband adoption Population Density Majumdar & Vankataraman (1998) The share of urban population indicati ng the density of consumers Multiple regression Population density is positively related to the intensity of innovation adoption. Figure 2 1 illustrates the research model to investigate the timing of digital TV ado ption. Based on th e research model, thi s study explores the factors that influence the timing of digital TV adoption, as well as looking at the relative importance of the factors.
69 Figure 2 1 Res earch m odel: Factors that d etermine the timing of d igital TV a doption by broadca st s tati ons Population Density Market Size Income Competition Vertical Integration Horizontal Integration Ownership
70 CHAPTER 3 METHODS This chapter describes research methods and procedures employed in this study, including sampling, data collection, measurement of variables, and statistical analysis techniques. Multiple regression independent samples t tes t, ANOVA and discriminant analysis were used to analyze the data and investigate research questions Research Design The unit of analysis should be selected and specified to ensure appropriate measur es of variables and data collection procedures ( Hair et al., 2006 ). The broadcast TV station is used as the unit of analysis in this study since this study analyzes innovation ado ption decisions at the firm level. This study employs secondary data analysis to examine the factors that influence the timing of d igital TV adoption Secondary data analysis has been widely used in social sciences including media economics management, and policy studies (Beam, 2006; Vartanian, 2010). S econdary data analysis has an advantage of using a large spectrum of empirical dat a for research (Bouslaugh, 2007) 1 For the purpose of this study, second ary data analysis is useful since secondary data analysis is appropriate to anal yze data collected on a t emporal basis in order to investigate timing of digital TV adoption P revious r esearch on digital TV adoption (Oberg, 2000; GAO 2002 a ) mostly employed survey analysis for research Even though survey analysis i s helpful in identifying the factors in the questionnaires, it may not be useful for analyzing data covering extended geograp hic areas and time periods unless regularly administered on a large scale. Surveys may not be suitable to 1 Since secondary data analysis typically uses large datasets often collected for a long period of time, this analysis is usually appropriate for longitudinal studies (Vartanian, 2010).
71 represent industry characteristics and to offer accurate measures for study becau se survey data are typically based on a small number of firm employe es (Rogers, 2003). Compared to the limitations of sur vey analysis, secondary data analysis can provide more comprehensive analysis of digital TV adoption since researchers can utilize a l arge set of data gathered from multiple sources. Sample Selection B roadcast stations can be classified based on business models and power levels: (1) commercial and noncommercial stations and (2) low power and f ull power stations. O nly full power stations were required to switch to digital in the dig ital TV transition co mpleted by June 12, 2009. Compared with commercial stations, non commercial stations have the business model that is not most responsive to market and likely aim to serve underrepresented audiences and local communities. Thus, t he sample of this study was s elected from full power and commercial broadcast TV stations excluding low power and noncommercial stations The stations had their year of adoption (the year that the stations adopted digital TV) and adoption period (the time span between the specific da te of adoption date and the completion date of digital TV adoption). T he station s in the sample were required to complete the adoption of digital TV by June 12, 2009, the deadline for the transition and to discontinue analog broadcasting. In the sample, t he adoption year spreads over 13 years, from 1997 to 2009 and the adoption period spans from one day to 3,053 days. Th e sample of this study consist s of the stations owned by the largest 25 TV groups. The stations owned by the largest 25 TV groups a ccount for more than 50% of all commercial TV stations in the U.S. Thus, these
72 s tations are considered appropriately r epresent the characteristics of the US broadcast stations. 2 The sample data for this study used 641 stations owned by the largest 25 commercial TV groups. From 641 stations owned by the largest 25 TV groups, 8 observations were eliminated from the data set for analysis in a screening process because these cases included incomplete data. As a result, the remaining 633 observations were used for an a lysis. In the sample stations (N = 633), the number of stations owned by the largest 10 TV groups were 318, which amounted to 50.2% of the entire sample stations. 3 Since more than a half of the sample stations were the largest 10 group owned stations, a s ubset of sample was created by excluding the largest 10 TV group owned stations. Thus, this study used two different samples; Sample 1 consists of stations owned by the largest 25 TV groups, and sample 2 consists of stations owned by the largest 11 to 25 T V groups. Table 3 1 shows the number and proportion of grou p owned stations compared to commercial TV stations This table illustrate s that g roup ownership of TV stations continued to expand throughout 1990 From 1990, the proportion of TV group owned stat ions slowed down (Roberts, Frenette, & Sternes, 2002). The dominance of group owned stations in the broadcast TV indust ry can be explained that the industry seeks for economies of scale since operating a collection of stations cost cheaper rather than a s ingle station. 2 The rank ing s of the largest 25 TV g roups are determined based on the total TV household reach of full power stations owned by the group Reach is obtained by adding up the percentages of the total TV homes of each market in which the group has at least one station ( Albinial, 2010 ). 3 Standard & Poor s (2011) report s that more than a quarter of commercial TV stations are owned by the largest 10 TV groups.
73 Table 3 1. Group o wnership in U S c ommercial TV s tations Year Total Commercial TV Stations Number of Group owned TV Stations % of Group owned TV Stations 1950 108 40 37 1960 523 294 56 1970 677 440 65 1980 741 506 68 1990 1,213 781 64 (Source) FCC (2008). Table 3 2 presents t he largest 25 TV groups with ownership identification and national coverage The s tations owned by the largest 25 TV group s include network affiliates and independent s For example, the stations owned by Belo Corps consist of 18 network affiliates an d two independent s Table 3 2 The l argest 25 TV g roup s Rank TV Group Ownership # of Stations Coverage 1 Ion Avenue Capital, Black Diamond Capital, Trilogy Capital 59 63.9 2 Univision Broadcasting Media Partners Inc 38 42 .0 3 CBS CBS Corp. 29 38.4 4 Fox News Corp. 2 9 37.1 5 NBC General Electric 25 36.6 6 Trinity Paul Crouch & Jan Crouch 28 35.9 7 Tribune Tribune Co. 24 35.1 8 ABC The Walt Disney Co. 10 23.3 9 Sinclair Sinclair Broadcast Group 56 21.6 10 G annet Gannet Co. 23 18.2 11 Hearst Hearst Corp. 35 18 .0 12 Multicultural TV Arthur Liu 5 16.7 13 Belo Belo Corp. 20 14.5 14 Entravision Entravision Communication 19 12.5 15 Raycom Media Employee owned but funded by Retirement Systems of Alabama 40 1 2.1 16 Nexstar Broadcasting Nexstar Broadcasting Group 51 11.1 17 Local TV Oak Hill Capital Partners 18 10.8 18 Cox Media Cox Enterprises 14 10.4 19 Newport TV Providence Equity Partners 35 10 .0 20 E.W. Scripps E.W. Scripps Co. 10 9.9
74 Table 3 2 Continued Rank TV Group Ownership # of Stations Coverage 21 Liberman Jose Liberman and Lenard Liberman 4 9.7 22 Meredith Meredith Corp. 11 9.1 23 Lin LIN TV Corp. 32 8.5 24 Media General Media General 20 8.3 25 Post Newsweek The Washington Post Co. 6 7.4 Total 641 *% of US TV Household (Source) Albinial (2010). Data Collection Station data were collected from multiple data source s. The m ain data source was Television and Cable Factbook which p rovides information on the beginning date of digit al TV adoption, the analog shut off date, ownership, netwo rk affiliation DMA rank and TV households of each station and TV market The station information was cross checked for accuracy in the TV station directory on T VnewsCheck.com, a broadcast industry news website. The directory on this website offers the profiles of all U.S. broadcast stations and the link to the FCC data base, TV Station Query In addition, t he Broadcasting and Cable Yearbook was a complementary source to profile station data The li st of sample stations was obtained from the largest 25 TV group list compiled in Broadcasting & Cable a trade journal of the broadcasting industry ( Albinia l, 2010). Market demographic data on DMA areas were collected from government public records, includ ing the County and City Data Book and the data base on the U.S. Census Bureau website Definitions and data sou rces for all variables are provided in Table 3 3 Measurement Dependent Variable The dependen t variable in this study is the timing of digital TV adoption. Rogers (1995) suggests that time is an es sential dimension in the process o f innovation
75 adoption He defines time as the degree to which a unit of adoption adopts innovation relatively earlier th an other units in a system (Rogers, 1995) Therefo re, the timing of innovation adoption at the inter firm level means relative timing that enables a firm s timing decision to be compared with other firms Previous studies measure the timing of adoption largely in two ways. First, the date of adoption is u sed to measure the timing of adoption (Collier & Messick, 1975 ; Saloner & Shepard, 1995 ). Second the timing of adoption was measured by the t ime period taken between the time of adoption and the time of finaliz ation of the adoption ( Boeker & Huo, 1998; Ne o, Khoo, & Ang, 1994). This study uses the two measures of timing as previous studies First, the year of adoption is employed to measure the timing of digital TV adoption. Second, the time period from the date when stations adopted digital TV to the date when stations completed the digital TV transition was used as a measure of the timing of digital TV adoption. Independent Variables There are two fundamental explanations for employ ing an independent variable in a causal model (Allison, 1999) F irst, an i ndependent variable is relevant to the study and researchers want to know its causal effect on the dependent variable. Second, an independent variable is the variable of interest t hat researchers want to control Employing the supply side approach t his st udy ex amines firm characteristics, market characteristics and market demographics as the factors that influence the timing of digital TV adoption. Firm characteristics Pre vious adoption studies have discussed the decisions of innovation adoption at the f irm level. Station related variables include ownership characteristics, horizontal integration and vertical integration
76 Ownership Station o wnership in the TV market is an important element that shapes firm characteristics of a station. This study employ s ownership as a factor to examine how the ownership of a station in fluences the timing of adoption. Yanich (2010) measures the ownership characteristics of a station as the number of stations owned by the group owner. Ford and Jackson (1997) measure owner ship characteristic s of a cable TV o perator as the number of cable systems owned by the MSO (Multiple System Operators). Borrell (1997) uses the presence of group ownership as an independent variable to predict the type of radio stations in the study of ra satellite programming Since a TV group usually owns more than one station in multiple markets, group owned stations can gain economies of scale in terms of management, programming, and advertising. Based on previous studies, owne rship in this study is Horizontal integration Horizontal integration takes place when an organization acquires companies in the same business (Straubhaar, LaRose, & Davenport, 2010). In the U.S. broadcasting industry the degree of h orizontal integration has been used as a standard to regulate ownership of media outlets to achieve a certain level of competition. The national TV ownership caps measure horizontal integration as th e national coverage of TV viewers by station group owner. 4 Regarding the cable TV industry, Ford percentage of total cable TV subscribers. Based on previous studies and the FCC rules, horizontal integration in this study is measured by the p ercentage of TV households tha t 4 The FCC permits that a single TV group can own any number of television stations nationally as long as the group reaches no more than 39% of the national TV audienc e (FCC, 2011).
77 Vertical integration Vertical integration in the broadcast industry refers to the integration between program production and di stribution facilities within conglomerates (Waterman, 2000 ) Under v ertical integration the two areas are considered for vertical ownership and vertical contract. Expanding the meaning of vertical integration, this study uses vertical integration as verti cal contracts between networks and stations N etwork affiliates have contractual arrangements w ith major networks such as ABC, NBC, CBS, and Fox as well as CW, MyNetwork, Ion, Telemundo, and Tribune. I ndependent s are the stations that are not directly aff iliated with any network While network affil iates are supplied with original primetime programming from networks independent s tend to depend most of the ir programming on syndication and local production of programming In this study vertical integration i s measured by the number of affiliated networks of a station; whether a station is single affiliat e station ( a station affiliated with one network ) multiple affiliat e station ( a station aff iliated with more than one network) and no affiliat e station (a n independent). Market c haracteristics Competition Competition is mea sured by the number of commercial statio ns in market Previous studies measure c ompetition as the number of stations broadcasting in the market (Lacy, Atwater, Qin, & Pow ers, 1989). The measure r epresents the number of commercial TV stations both VHF and UH F, excluding public stations and low power broadcast stations Competition may provide incentives to stations for adoption because having more stations in market can de crease the market share of a station This study assumes that s tations in more competitive market likely adopt digital TV early in order t o gain competitive advantage s over their competitors.
78 Market size Although the definition of market size can vary m arket size is measured in terms of population i n media studies (Berry, 1999) To measure market size in this study, two different variables were used. F irst m arket size is measured by the DMA ( Designated Market Area ) rank of TV stations established by Nie lson Media Research ( Yanich, 2010 ) Second, the number of TV households in market is employed as a measure of market size. 5 Market demographic s To assess the impacts of local demand potentials on digital TV adoption market demographics are used as a set of i ndependent variables. Since i ncome and population density have been employed as representative factors for market demographi cs in previous literature this s tudy includes these two factors in order to explain the timing of digital TV adoption. Income The growth of income generally represents an increase in consumer spending. Previous studies (Berry, 1999; Demirguc Kunt et al., 2008) measure i ncome by p er capita income in a station s market For example, B eilock and Dimitrova (2003) identify that inco me is a key determinant of I nternet adoption in a country. These studies suggest that broadcast stations operating in the market with higher income may adopt digital TV earlier th an the stations in the market with lower income. Population density P opulat ion density is defined as the number of people per square mi le (Baldridge & Burnham, 1975) Population density is used as a benchmark for est imating audience contribution s in the market. Sparsely populated a reas with low 5 T he number of TV households is more proper for empirical analysis, since the number of TV households has a higher level of measurement than the DMA rank. Since t he higher the leve l of measurement of a variable the more powerful are the s tatistical techniques used to analyze it.
79 population densities can create hig h cost c onditions for media and telecommunications firms For example, i n the adoption of high speed Internet, it is uncertain whether consumers in low population density areas are willing to pay a rate that would cover the full cost of broadband Internet access (Compaine, 2003). Based on previous research, t his study examines whether population density affects timing of digital TV adoption Table 3 3 describes t he measureme nt of variables and source s of data used in this study. Table 3 3 Measurement of v ariables and data source for the m odel Variable Measurement Source of Data (Dependent Variable) Timing of Digital TV Adoption (1) Adoption Y ear of digital TV (2) Number of days bet ween t he adoption date and the completion date of digi tal TV adoption Broadcasting & Cable Yearbook (2010) T elevision & Cable Factbook (2010) (Independent Variables) Firm Characteristics Ownership N umber of TV stations owned by Broadcasting & Cable Yearbook (2010), T elevision & Cable Factbook (2010) Horizontal Integration Percentage of TV households reaches Broadcasting & Cable Yearbook (2010), T elevision & Cable Factbook (2010) Vertical Integration Number of affiliated networks of a station Broadcasting & Cable Yearbook (2010) T elevision & Cable Factbook (2010) Market Characteristics Competition Number of commercial TV market Broadcasting & Cable Yearbook (2010) T elevision & Cable Factbook (2010) Market Size (1) DMA rank of a station (2) Number of TV households in a station s market Broadcasting & Cable Yearbook (2010) T elevision & Cable Factbook (2010) Market Demographics Income Per capita income in a station s market U.S. Census Bureau ( 2007) Popula tion Density The ratio of population to square miles in a station s market U.S. Census Bureau (2007)
80 Data Analysis Method s This study employs m ultiple regression as main method of data analysis Multiple regression is a widely applied statistical techni que which explore s the relationship between a single dependent variable and several independent variables (Hair, Anderson, Black & Tathum, 2006) Using multiple regression is appropriate for this study for several reasons Multiple regression can provide a robust statistical procedure that reveals the relationship s between the independent variables of firm characteristics, market characteristics, and market demographics, and the dependent variable of timing of digital TV adoption M ultiple regression analys is is especially u seful in providing a cau sal analysis of data (Allison, 1999; Osborne, 2000). Multiple regression can separate the effects of independent variables by examin ing the unique contribution of each independent variable on the dependent variable controlling for other variables (Allison, 1999). T his study focuses on exploring firm characteristics, market characteristics, and market demographics that influence the timing of digital TV adoption. Since multiple regression can e stimate the magnitude of eac h independent variable employing multiple regressi on in this study can identify which independent variable best explains the timing of digital TV adoption. All independent and dependent variables in this study are continuous variables, which is appr opriate for regression analysis Before conducting multiple regression assumption tests for multiple regression (Hair et al, 2006) were performed First, histograms and normal probability plots were used to check normality of data. Second, a correlation matrix was used to assess muliticollinear ity among independent variables Third residual plot s were used to check the assumption of equal variance of errors Fourth the linearity assumption was checked by producing scatter plots For the variables that d id not meet the normality
81 and equal variance assumption s logarithmic transformation s were applied because the data were positively skewed (Hair et al., 2006). This study models the year of adoption and adoption period of digital TV adoption as a function of firm characteristics, market characteristics, and market demographics to examine the influence of independent variables on the timing of digital TV adoption. In general t he types of data for the empirical analysis consist of longitudinal, cross secti onal, and pooled data (Gujarati & Porter, 2 0 09). The model in this study uses pooled or combined data, which have elements of cross sectional data. Also, t he data set s used in this study are panel data because the data are collected for the same TV station s over the years. The empirical model draws on the factors identified from previous adoption studies. T he dependent variable is the timing of d igital TV adoption, measured by (1) the year of digital TV adoption, and (2) the period between the adoption date and the completion date of digital TV adoption The independent variables used in this model were ownership, horizontal integration, vertical integration, competition, market size, income, and population density. In addition, ANOVA and independent samples t tests are conducted to investigate the role of vertical integration in the timing of digital TV adoption. ANOVA is used to compare mean differences in the timing of adoption among six groups of stations, while i ndependent samples t tests are employed to compare such difference s between two groups of stations, network affiliates and independents. Also, d iscriminant analysis is performed to identify the two groups of station, early adopters and late adopters, and determine which variables best predict diff erences between e a rly adopters and late adopters.
82 CHAPTER 4 RESULTS This chapter provi des the result of data analysis a description of the samples and descriptive s tatistic s, t he result of correlation analysis among the independent variables, the resul t of regression analysis the result of independent samples t tests and ANOVA (Analysis of Variance ) and discriminant analysis. Measurement of the Dependent Variable The dependent variable, t he t iming of adoption was measured in two ways : the year of a doption and adoption period First, the year of digital TV adoption span s from 1997 to 2009. Among the 6 33 s tations owned by the largest 25 groups the earliest adoption occurred on October 1, 1997 by KIVO in Honolulu owned by the Hearst Argyle TV Group. 1 Second, the adoption period span s from one day to 3,053 days. The adoption period was obtained by counting the numb er of days, from the date when each station adopted digital TV to the date when each station completed digital TV adoption This study expec ts that using two measures of the timing will result in the opp osite direction s of relationship between variables since the year of adoption has an inverse relationship with adoption period. In order that using the year of adoption will produce the result s that have the same directional relationship with using the adoption period, values range d from 1 to 13 w ere given to each year of digital TV adoption 2 Table 4 1 shows the number of stations and Figure 4 2 illustrates the number of affiliates and indepe ndents that adopted digital T V from 1997 to 2009. As can be seen, largest 1 After the FCC assigned digital TV channels on April 2, 1997 the Hearst Argyle TV Group received the permission to build digital transmitters, becoming the first to adopt digital TV in the U.S. market (Duggin, 19 98). 2 In the analysis, the year of 2009 was set as a reference year and values are added up. To illustrate the year of 2009 was coded as 1, the year of 2008 as 2, the year of 2007 as 3 and the year of 1997 as 13.
83 number of stations ( 20 5 32.3%) adopted digital TV in 2002 while the smallest number of station ( 1 1.6%) adopted digital TV in 1997 Figure 4 2 illustrates the year of digital TV a doption by netwo rk affiliates and independents In the sample, i ndependents adopted digital TV later than affiliates with the year of ado ption star ting from 2002 By contrast, ABC affiliates adopted digital TV earlier than other affiliates and independent s with the year of adoption starting from 1997. Figure 4 1. Timing of d igital TV a doption by b roadcast s tations ( 1997 2009 ) Table 4 1. Timing of d igital TV a doption by a ffiliates and i ndependents (1997 2009) Year of Adoption ABC Affiliates CBS Affiliat es NBC Affiliates Fox Affiliates Other Affiliates Independents 1997 1 1998 11 8 7 3 2 1999 8 6 11 12 5 2000 1 8 7 10 1 2001 4 7 4 1 10 2002 2 4 30 30 2 7 8 4 13 2003 13 9 16 7 42 3 2004 2 4 3 7 16 2005 8 7 6 10 16 2 2006 2 5 5 7 33 3 2007 2 3 8 11 1 2008 1 1 4 2 7 2 2009 1 3 5 2 17 2 Total: 76 Total: 90 Total: 101 Total: 96 Total: 244 Total: 26 *The sample contains 26 multiple affiliate stations. Since all multiple affiliate stations include one of major network affiliates, for example, CBS/CW, multiple affiliate stations are coded as one of four major network affiliates. 0 50 100 150 200 250 Number of Stations Year of Adoption Number of Stations
84 Figure 4 2 Timing of digital TV adoption by affiliates and i ndependents (1997 2009) Descriptive Statistics The u nit of analysis in this study is a station Table 4 2 presents descriptive statistics for the largest 25 TV group stations. Mean, median and standard deviation describe the basic features of the data in this study. Although the mean is often used to represent the central tendency of the data, the m edian is appropriate when describing skewed data or data with extreme values For the datase t in this study, the median is particularly useful to describe market size 2 and population density whose data are likely influenced by extreme values in large TV markets. Table 4 2 presents descriptive statistics for the largest 25 TV group owned stations. For the largest 25 TV groups, t he mean of the year of adoption was 2002. 9 5 (SD = 2. 69 N = 633) The mean of adoption period was 1564.29 (SD = 688.12). The mea n of ownership was 34.41 (SD = 14.99). H oriz ontal integration and vertical integration showed mean scores of 24 83 (SD=1 6.88 ) and .98 (SD=.26) respectively. T he mean of competition was 10. 39 (SD=5.13) The means of DMA rank and the 0 10 20 30 40 50 60 70 80 90 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Number of Stations Year of Adoption ABC affiliates CBS affiliates NBC affiliates Fox affiliates Other affiliates Independents
85 number of TV households were 49.46 (SD = 38.00) and 1 298 796.70 (SD = 1 394 262.04), respectively. T he means of income and population densit y were 35 832 05 (SD=11,272.33) and 4798.48 (SD = 4 525.56 ) respectively. Compared to the na tional per capita income of $ 27,041 in 2005 2009 (U.S. Census Bureau, 2011) t he mean of income used in this study was higher than that of national income Also, m edian s are presented for each variable. The median of ownership was 32.00. The medians of horizontal integration and vertical integration were 32 .00 and 1 .00 respectively. The median s of competition and market size w ere 9.00 and 5.89, respectively. The median s of income an d populati on density were 33,542 and 3 407.50 respectively. Table 4 2 Descriptive s tatistics : The largest 25 TV g rou p owned s tations Variable (Measure) Mean Median SD Timing ( Year o f Adoption ) 2002.9 5 2002 2. 69 Timing (Adoption Period) 1564.29 1720 688.12 Ownership 34.41 32.00 14.99 Horizontal Integration 24.83 18.20 16.88 Vertical Integration .98 1.00 .26 Com petition 10. 39 9.00 5.1 3 Market Size 1 (DMA Rank) 49.46 38.00 44.80 Market Size 2 ( # of TV Households) 1298796.70 776080.00 1394262.04 Income 35832 05 33542 .00 11 272.3 3 Population Density 4798.48 3407.50 4525.56 Table 4 3 presents descriptive statisti cs for the largest 11 to 25 TV group owned stations. T he mean of year of adoption was 2003.03 (SD = 2. 68 N = 315 ) and t he mean of adoption period was 15 33.77 (SD = 694.78 ). The mean of ownership was 30.16 (SD=13.55). H orizontal integration and vertical i ntegration showed mean scores of 11.54 (SD= 2.86 ) and .98 (SD=. 34 ) respectively. The mean s of competition DMA rank, and the number of TV households were 9.15 (SD= 5.50 ) 68.06 (SD = 49.33), and
86 805 973.56 (SD = 849 160.39 ), respectively. T he means of incom e and population density were 68.06 (SD= 49.33 ) and 805 973.56 (SD = 8 49,160 39 ) respectively. Also, m edian values were presented for each variable. The median of ownership was 35.00. The medians of horizontal integration and vertical integration were 11.1 0 and 1 .00 respectively. The medians of competition DMA rank, and then the number of TV households were 8.00 53.00, and 619,610.00, respectively. The median s of income and population density were 32 525.00 and 2 736.00 respectively. Table 4 3 Descrip tive statistics: The largest 11 25 TV g roup owned s tations Variables (Measure) Mean Median SD Timing ( Year of Adoption ) 200 3.03 2002 2. 68 Timing (Adoption Period) 15 33 7 7 17 09 68 4 23 Ownership 30.16 35.00 13.55 Horizontal Integration 11.54 11.10 2. 86 Vertical Integration .98 1.00 34 Competition 9.15 8.00 5.50 Market Size 1 (DMA Rank) 68.06 53.00 49.33 Market Size 2 ( # of TV H H s) 805973.56 619610.00 849160.39 Income 33431.56 32525.00 7669.08 Population Density 3519.15 2736.00 2848.84 Corr elation Analysis C orrelation analysis was conducted before the regression analysis in order to describe the relationship between the dependent and independent variables. Co rrelation analysis shows several bivariate association s among continuous independen t variables (Hair et al., 2006) The main purpose of correlation analysis is to examine for evid ence of violation assumptions in multiple regression, in particular, to check multicollinearity Mul ti collineari t y becomes a problem when two independent variab les are highly correlated. Researchers generally use a correlation greater than .65 for evidence of multicollinearity (Agresti & Finlay, 2009). In correlation analysis, a strong correlation was found between DMA ran k (Market Size 1) and the number of TV
87 ho u seholds (Market Size 2) To correct the potential problem with multicollinearity M arket S ize 1 was removed from the initial regression model Since Market S ize 2 showed a higher correlation with the dependent variable t han M arket S ize 1 Market Size 2 was retained in the model and renamed Market Size (Hair, Celsi, Money, Samouel, & Page, 2011). The variance inflation factors (VIF) also measure multicollinearity in the regression model. Large VIF values, such as 5.0 or higher, express a high degree of multic ollinearity (Hair et al, 2011). The VIFs for all independent variables in the model were reasonably low, ranging from 1.1 to 3.3 which did not indicate concerns about multicollinear ity Thus t here was no strong correlation among independent variables tha t could make it inefficient to conduct regression analysis. Table 4 4 displays correlations among independent variables and the timing of adoption, measured by year of adoption and adoption period The timing of digital TV adoption had significant associa tions with all independent variables with the exception of horizontal integration and vertical integration. The correlation matrix shows that m arket siz e had highest correlation with the year of adoption ( r = .376 ) The strengths of association with the y ear of adoption showed the order of ownership ( r = .276), population density ( r =.234 ), competition ( r = .187), and income ( r = .117). Also, m arket size ( r = .373) showed the highest correlation with adoption period. The strengths of association with adop tion period showed the order of ownership ( r = .268), population density ( r = .222), competition ( r = .187 ), and income ( r =. 110). Horizontal integration and vertical integration had no significant correla tion with adoption period respectively. Among in dependent variables, t here were overall significant associations with the
88 exception of the association between (1) horizontal integr ation and vertical integration, and (2) ownership and income. Table 4 4 Correlatio n m atrix : The l argest 25 TV g roup owned s tations VARIABLE 1 YEAR ( PERIOD) 2 OWN 3 HI 4 VI 5 COMP 6 MKT_ SIZE 7 INC 8 POP_ DEN 1. YEAR ( PERIOD) .276** ( .268 ** ) 003 ( 006 ) .061 (.056) .187 ** (.187 ** ) 376 ** (. 373 ** ) .117** (.110 ** ) 234 ** (. 222 ** ) 2. OWN .276** ( .268 ** ) .360** .146** .183** .245** .060 .118** 3 HI (log) 003 ( 006 ) .360** .019 .290 ** .442 ** .224** .334 ** 4 VI .061 (.056) .146** .019 .199 ** .18 3 ** .095** .10 6 ** 5. COMP .187** (.187 ** ) .183** 290 ** .199 ** 669 ** .2 94** 416 ** 6 MKT_SIZE (log) .376** (. 373 ** ) .245** 442 ** .183** .669** 495 ** 646 ** 7 INC .117 ** (.110 ** ) .060 224 ** .095 ** .294 ** 495 ** 548 ** 8 POP_DEN (log) .234** (. 222 ** ) .118** .334 ** .106* .41 6 ** 646 ** 548 ** p < .05, ** p < .01 Table 4 5 presents the results of correlation analysis for the largest 11 to 25 TV group stations. The timing of digital TV adoption had significant associations with all independent variables with the exception of horizontal integ ration and vertical integration. Among the independent variables, market siz e had the highest correlation ( r = 444 ) with year of adoption The strength of association with the year of adoption showed the order of competition ( r = .233), population density (r = .220) ownership ( r = .215), and income ( r = 208 ) Market size also showed th e highest cor relation ( r = 4 25 ) with adoption period. The strength of association with adoption period showed the order of competition ( r = .2 20 ), ownership ( r = .202), i ncome ( r =. 193 ), and population density
89 ( r = 192 ) Among independent variables, t here were overall significant associations, with the exception of the association between horizontal integration and market size. Market size showed the highest association w ith the timing of adoption both for the largest 25 gr oup stations and the largest 11 to 25 group stations. Table 4 5 Correlatio n m atrix: The l argest 11 to 25 TV g roup owned s tations Variable 1 YEAR ( PERIOD) 2 OWN 3 HI 4 VI 5 COMP 6 MKT_ SIZE 7 INC 8 POP_ DEN 1. YEAR ( PERIOD) .2 15 ** ( .2 02 ** ) 0 25 ( 022 ) .0 76 (.083 ) 2 33 ** (.220 ** ) 444 ** ( 425 ** ) 208 ** (.193 **) 2 20 ** (.192 **) 2. OWN .2 15 ** ( .2 02 **) 169 ** .1 90 ** 324 ** 523 ** 169 ** 1 90 ** 3 HI (lo g) 0 25 ( 0 2 2 ) 169 ** 150 ** 102 079 119 124 4 VI .0 76 (.083 ) 1 90 ** 150 ** 255 290 ** 193 ** 1 88 ** 5 COMP (log) 2 33 ** (.220 ** ) 324 ** 102 255 715 ** 3 22 ** 324 ** 6 MKT_ SIZE (log) 444 ** (. 425 ** ) 523 ** 079 290 ** 715 ** 4 62 ** 499 ** 7 INC (log) 208 ** (.193 ** ) 169 119 ** 193 ** 322 ** 4 62 ** 423 ** 8 POP_ DEN (log) 2 20 ** (.192 ** ) .1 90 ** 124 188 ** 324 ** 499 ** 423 ** p < .05; ** p < .01 Determinan ts of the Timing of Digital TV Adoption This study emp loyed multiple regression analysis to explore the impact of firm characteristics, market characteristics, and market demographics on the timing of digital TV adoption The relationship s of independent v ariables to dependent variable (RQ1 to RQ 7 ) were examined As stated previousl y, regression analyses were performed for two samples; the largest 25 TV group stations (N = 633) and the largest 11 to 25 TV group stations (N = 315)
90 The Largest 25 TV Group owned Stations For the largest 25 TV group stations, regression analysis was conducted using the year of adoption as a measure of the dependent variable The regression model was significant in terms of model fit (F= 25.646 p <.01 ). The model explained ap proximately 2 2 % of total variance ( R 2 = .2 23 ) Examining the adjusted R 2 (adjusted R 2 = .2 1 4) shows little loss in prediction when compared to the R 2 value. In addition, multiple regress i on was conducted using adoption period. The regression model was sign ificant in terms of model fit (F= 2 4 954 p <.01 ). The model explained approximately 22 % of total variance s ( R 2 = .2 1 8 ) Examining the adjusted R 2 (adjusted R 2 = 2 10 ) shows little loss in prediction compared to the R 2 value The Largest 11 to 25 TV Group owned Stations For the largest 11 to 25 TV group stations, r egression analysis was conduct ed for the largest 11 to 25 TV group stations when the timing was measured by the year of adoption The regression model was significant in terms of model fit (F= 1 5 035 p <.01 ). The model explained approximately 2 6 % of total variance ( R 2 = .25 5 ) Examining the adjusted R 2 value (adjusted R 2 = .2 38 ) shows little loss in prediction when compared to the R 2 value. In addition, regression analysis was conducted when the timing of digital TV adoption was measured by adoption period. T he regression model was significant in terms of model fit (F= 1 3.880 p <.01 ). The model explained approximately 24 % of total variance ( R 2 = .240 ) Examining the adjusted R 2 value (adjusted R 2 = .223 ) shows little loss in prediction when compared to the R 2 value. Firm C haracteristics This study has three research questions exploring the influence of firm characteristics on the timing of digital TV adoption; R Q 1 ( What, if any, influence
91 own ership have on the timing of digital TV adoption by broadcast stations ? ) RQ 2 ( What, if any, influence does horizontal integration have on the timing of digital TV adoption by broadcast stations? ) and RQ 3 ( What, if any, influence does horizontal integra tion have on the timing of digital TV adoption by broadcast stations? ). (1) For t he l argest 25 TV Group owned Stat ions : When the timing of digital TV adoption was measured by the year of adoption ownership (p <.01 ), h orizont al integration ( p <.01 ) and verti cal integration (p <.01 ) were all found to be significant factors Using adoption period ownership (p <.01 ), horizontal integration (p <.01 ) and vertical integration (p <.01 ) were all found to be significant factors. (2) For t he Largest 11 to 25 TV Group ow ned Stations: When the timing of digital TV adoption was measured by the year of adoption vertical integration (p <.01 ) were found to be a significant factor while ownership and horizontal integration had no impact on the year of adoption Using adoption period, vertical integration (p <.01 ) were found to be a significant factor while ownership and horizontal integration had no impact on the year of adoption. Market Characteristics R esearch question 4 ( What, if any, influence d oes competition have on th e timing of digital TV adoption by broadcast stations ? ) and r esearch question 5 (What, if any, influence d oes market size have on the timing of digital TV adoption by broadcast stations? ) focus on the influence of market characteristics on the timing of di gital TV adoption. (1) For t he Largest 25 TV Group owned Stations: B oth competition ( p <.05 ) and market size ( p <.01 ) were found to be significant factors on the year of digital TV adoption. When the timing was me asured by adoption period, both competition ( p <.05 ) and market size ( p <.01 ) were found to be significant factors. (2) For t he Largest 11 to 25 TV Group owned Stations: B oth market size ( p <.01 ) and competition ( p <.05 ) were found to be significant factors on the year of adoption When timing was measur ed by adoption period, both market size ( p <.01 ) and competition ( p <.05 ) were found to be significant
92 Market Demographics R esearch question 6 ( What, if any, influence does income have on the timing of digital TV adoption by broadcast stations ? ) and r es earch question 7 ( What, if any, influence does population density have on the timing of digital TV adoption by broadcast stations ? ) focus es on the influence of market demographics on the timing of digital TV adoption. (1) For the Largest 25 TV Group owned Sta tions: When timing was measured by the year of adoption only income ( p <.01 ) was found to be a significant factor o n the year of adoption. By contrast, p opulation density had no influence on the year of adoption W hen timing w as measured by adoption peri od, only income ( p <.05 ) was found to be a significant factor P opulation density had no significant effect on adoption period. (2) For the Largest 11 to 25 TV Group owned Stations: When timi ng was measured by th e year of adoption both income and populat ion density had no significant impact on the year of adoption. Likewise both income and population density had no significant impact on adoption period Tables from 4 6 to 4 9 present the results of regression analyses Table 4 6 reports the results of regres sion analysis for the largest 25 TV group owned stations when timing was measured by the year of adoption Table 4 7 shows the results of re gression analysis for the largest 25 TV group owned stations when timing was measured by adoption period. Table 4 8 shows the results of regression analysis for the largest 11 to 25 TV group owned stations when timing was measured by year of adoption Table 4 9 shows the results of regression analysis for the largest 11 to 25 TV gr oup owned stations when t iming was meas ured by adoption period The results indi cate that regression analyse s conducted for the largest 11 to 25 TV group owned station s provide d a slightly better fit than th e largest 11 to 25 TV group owned stations Also, regression analyses using the year of adoption produce d a slightly better fit than using adoption period.
93 Table 4 6 Regression a nalysis for the l argest 25 TV g roup owned s tations: Year of a doption Variable B SE t p Ownership .025 .008 3.188 .002 ** Horizontal Integration (log) 1.346 .442 3.048 .00 2 ** Vertical Integration 1.549 .379 4.083 .000** Competition .056 .025 2.237 .0 26 Market Size (log) 3.364 .409 8. 233 000** Income 2.263 .000 2. 184 029 Population Density (log) .300 .410 733 464 F (633) R 2 Adjusted R 2 2 5.646 ** .2 23 2 14 *p < .05; **p < .0 1 Table 4 7 Regression a nalysis for the l argest 25 TV g roup owned s tations: Adoption p eriod Variable B SE t p Ownership 6.716 2. 029 3.310 .00 1 ** Horizontal Integration (log) 3 27.961 11 4.936 2.853 .00 4 ** Vertical Integration 3 99.170 9 8.688 4.045 .000** Competition 1 4 406 6. 523 2. 208 .0 28 Market Size (log) 8 17 576 10 6.324 8. 197 000** Income .006 .00 3 2. 2 71 02 3 Population Density (log) 46.2 78 10 6.618 434 664 F (633) R 2 Adjusted R 2 2 4. 954 ** .2 1 8 2 10 *p < .05; **p < .0 1 Table 4 8 Regression a nalysis for the l argest 11 25 g roup owned s tations: Year of a doption Variable B SE t p Ownership .006 .012 .497 619 Horizontal Integration (log) .905 1.390 .651 701 Vertical Integration 1.6 44 .4 10 4.012 000** Competition(log) 1.6 23 76 2 2.130 .031* Market Size (log) 4.2 33 60 4 7.005 000** Income (log) .5 91 1.74 7 .338 759 Population Density (log) .0 11 57 4 .0 20 953 F ( 315 ) R 2 Adjusted R 2 1 5 .035 ** .2 55 .2 38 *p < .05; **p < .0 1
94 Table 4 9 Regression a nalysis for the l argest 11 25 g roup owned s tations: Adoption p eriod Variable B SE t p Ownership 1.866 3. 124 597 551 Horizontal Integration (log) 2 00 539 3 63.833 551 5 82 Vertical Integration 4 27 938 10 7 307 3.9 88 .00 0 ** Competition (log) 429 637 19 9 442 2. 1 54 .0 3 2 Market Size (log) 1 099 574 15 8 200 6.951 000 ** Income (log) 1 27 847 4 57 329 280 7 80 Population Density (log) 67 157 1 50 205 447 655 F ( 315 ) R 2 Adjusted R 2 1 3.880 ** .2 40 2 23 *p < .05; **p < .0 1 Comparison of Independent Variables The standardized coefficient ( ) shows relative assessment of each indepe ndent variable s importance in the regression model. In order to show the relative contribution of independent variables, this study compares the standardized coefficients of independent variables. Table 4 10 compares standardized coefficients for the larg est 25 TV group owned stations Starting from the largest, t he magnitude s of relative contribution for the year of adoption showed the order of m arket size vertical inte gration, horizontal integration, ownership, competition and income. Starting from the largest, t he relative contribution of independent variables adoption peri od showed the order of market size, vertical integration, ownership, horizontal integration, competition, and income. Table 4 10 Comparison among independent v ariables : The l argest 25 TV g roup owned s tations Regression Model IV ( Year of Adoption ) ( Adoption Period ) Ownership .138 .1 44 Horizontal Integration (log) 143 1 34 Vertical Integration .14 8 .1 4 7 Competition .107 1 0 6 Market Size (log) 527 5 26 Income .0 95 .0 9 9 Population Density (log) .0 36 .0 2 1
95 Table 4 11 shows the result of comparison of standardized coefficients for the largest 1 1 to 25 TV group owned stations The relative contribution of independent variables showed the same order for the year of adoption and adoption period Starting from the largest, t he magnitud es of relative contribution showed the order of market size, vertical integration, and competition. Table 4 11. Comparison among independent v ariables : The l argest 11 to 25 TV g roup owned s tations Regression Model IV ( Year of Adoption ) ( Adoption Period) Ownership .030 .0 36 Horizontal Integration (log) .034 .0 29 Vertical Integration 21 0 2 11 Competition (log) 1 5 2 1 5 5 Market Size (log) 62 3 6 24 Income (log) .0 20 .0 16 Population Density (log) .00 1 .0 2 7 Additionally, t his study conducted independent samples t test s and ANOVA to compare the differences in the timing of digital TV adoption for different categories of vertical integration: affiliates and independents. The purpose of conducting independe nt samples t test and ANOVA is to assess group mean differences in the timing of adoption. The independent t test examines statistical difference s in two groups, while ANOVA is used to test differences in three or more groups (Hair et al., 2006; Hair et al ., 2011). Independent Samples T test s Research q uestion 3 1 asked whether there is a significant difference in the timing of digital TV adoption b etween network affiliates and independent s. To investigate this research question, independent samples t test was conducted to compare the mean differences between two different station groups: network affiliates and independents. Based on the independent variable, vertical integration, stations
96 were grouped into two categories: network affiliates and independent s .The dependent variable, the timing of digital TV adoption was measured by the year of adoption and adoption period. Table 4 1 2 shows th e result of independent samples t test s for the largest 25 TV group stations including sample size, mean, standard de viation, t test statistics and p value. When timing was measured by the year of adoption M=2003. 85) was later than of adoption (M=2002.91). However, there was no statistical significan ce in year o f adoption between affiliates and independents When timing was measured by adoption period, (M= 1354.31 ) was shorter than mean adoption period (M= 1573.29 ). However, there was no statistical ly significant diffe rence in adoption period betw een network affiliates and independents Table 4 1 2 Independent s amples t t est : The l argest 25 TV g roup owned s tations Network Affiliate s (N = 607) Independents (N = 26) t p Mean SD Mean SD Year of Adoption 2002.91 2.700 2003.85 2.327 1.745 .082 Adoption Period 1573.29 689.528 1354.31 630.428 1.591 .112 Table 4 1 3 shows th e result of independent samples t test s for the largest 11 to 25 TV group stations including sample size, mean, standard deviation, t test statistics and p value. When the timin g of digital TV adoption was measured by the year of adoption (M = 2002.97 ) was earlier than the mean f or network affiliates (M = 2003.86 ), but no statistical significance was foun d in the mean year of adoption betw een the two groups of TV stations. When the timing was measured by adoption period, (M = 2002.97 ) was shorte r than (M = 2003.86 ), but no statistical signif icance was found betw een the two groups of TV stations These results suggest that affiliates and
97 independents do not show significant mean difference in the timing of digital TV adoption Table 4 1 3 Independent s amples t t est : The l argest 11 to 25 TV g r oup owned s tations Network Affiliates (N=607) Independents (N=26) t p Mean SD Mean SD Year of Adoption 2003.86 2.764 2002.97 2.167 1.818 .080 Adoption Period 1563.71 702.590 1336.77 581.011 1.734 .095 *p < .05; **p < .0 1 ANOVA Research Question 3 2 a sked whether there is a significant difference i n timing of digital TV adoption a mong six groups of stations: ABC affiliates CBS affiliates NBC a ffiliates Fox affiliates other affiliates and independents Analysis of Variance (ANOVA) wa s employed to investigate inter station difference s in the timing of digital TV adoption due to vertical integration. The Largest 25 TV Group owned Stations In the sa mple of the largest 25 TV group owned stations t he stations were categorized in to six g roups : i ndependents (N=26) ABC affiliates (N=76) CBS affiliates (N=90) NBC affiliates (N=101) Fox affiliates (N=96) o ther affiliates (N=244) Table 4 14 summarizes the results of ANOVA which showed significant mean differences among the six vertical integration categories S ix groups of stations showed significant mean differences on the year of adoption (F=10 347 p <.01 ), and on adoption period (F= 9.985 p <.01 ) The results indicate that statistical differences exist amo ng the six groups of station s in the timing of adoption. The mean scores show that ABC affiliates had the earliest mean year of adoption while i ndependents had the latest mean year of adoption. Starting from the earliest, t he order in the mean year of adoption was ABC affiliates, CB S affiliates, NBC affiliates, Fox affiliates, other affiliates and finally independents Conversely, independents ha d the
98 shortest mean adoption period, while ABC affiliates had the longest mean adoption period. Starting from the shortest, t he order in th e mean of adoption period was independents, other affiliates, Fox affiliates, NBC affiliates, CBS affiliates, a nd finally ABC affiliates which appeared in reverse order from using the year of adoption Table 4 14 ANOVA : The l argest 25 TV g roup owned s tat ions V ertical Integration N Mean S D F p Timing (Year of Adoption ) Independents 26 2003.846 2.327 10. 347 .00 0 ** ABC affiliates 76 2001.778 2.606 CBS affiliates 90 2002.211 2.612 NBC affiliates 101 2002.436 2.865 Fox affiliates 96 2002. 875 2.807 Other affiliates 244 2003.725 2.690 Timing (Adoption Period) Independents 26 1354.308 630.428 9.985 .00 0 ** ABC affiliates 76 1859.658 651.875 CBS affiliates 90 1756.078 665.503 NBC affiliates 101 1687.772 727.501 Fox affil iates 96 1175.854 6 21.594 Other affiliates 244 1368.262 688.115 *p < .05; **p < .0 1 The Largest 11 to 25 TV Group owned Stations In the sample o f the largest 11 to 25 TV group owned stations t he stations were categorized in to six groups: independe nts (N=22 ), ABC affiliates (N= 53 ), CBS affiliates (N=64 ), NBC affiliates (N=78 ), Fox affiliates (N= 51 ), o ther affiliates (N=47 ). Table 4 15 presents the results of ANOVA for s tations owned by the largest 11 to 25 TV groups, which showed significant mean di fferences among the six groups of stations categorized by vertical integration types Six groups of stations showed significant mean diff erences on the year of adoption (F=4.731, p <.01 ) and on adoption period (F= 4.615 p <.01 ). The results indicate that statistical differences exist am ong the six groups of stations i n the timing of adoption for the largest 11 to 25 TV group stations The mean scores show that ABC affiliates had the earliest year of adoption while i ndependents had the latest mean year of adoption. Starting from the earliest, t he order in the mean year of adoption was: ABC
99 affiliates, CBS a ffiliates, NBC affiliates, Fox affiliates, other affiliates, and finally independents By contrast, independents had the shortest mean adoption period, while ABC affiliates had the longest mean adoption period. Starting from the shortest, t he order in the mean adoption period was independents, other affiliates, Fox affiliates, NBC affiliates, CBS affiliates, and ABC affiliates which appeared in reverse o rder from using the year of adoption. Table 4 15 ANOVA : The l argest 1 1 to 25 TV g roup owned s tations DV V ertical Integration N Mean S D F p Timing ( Year of Adoption ) Independents 22 2003.864 2.167 4.731 .00 ** ABC affiliates 53 2001.75 5 2.731 CBS affiliates 64 2002.813 2.416 NBC affiliates 78 2003.013 2.817 Fox affiliates 51 2003.294 2.693 Other affiliates 47 2004.085 2.412 Timing (Adoption Period) Independents 22 1336.773 581.011 4.615 .00 ** ABC affiliates 53 1865.132 683.559 CBS affiliates 64 1610.828 619.606 NBC affiliates 78 1554.154 717.922 Fox affiliates 51 1480.608 675.863 Other affiliates 47 1278.383 637.435 *p < .05; **p < .0 1 Discriminant Analysis RQ 8 ( What factor s explain the differences between the two group s of stations, early adopters and late adopters of digital TV ?) is formulated to examine what factors discriminate two groups of stations, early adopters and later adopters. A two group discriminant analysis i s performed to determine which variables explain the differences between early adopters and late adopters The main objective of discriminant analysis is to examine group differences, predict ing to which particular gro up an individual will belong based on independent varia bles. Discriminant analysis reveals independent variables that maximize the differences between groups defined by the dependent variable (Hair et al., 2011). This study develops the categorizati on of three group, early
100 adopters, majority, and late adopters, based on the innovation adoption the ories and the re search purpose. This categorization reflects natural groupings in year of adoption over 1997 2009. While this category does not exactly match the percentages in the diffusion of innovat ions theory, 3 this classification provides a logical basis for the research question in two aspects. First, the classification uses a natural breakpoint for each adopter category, as well as approximately corresponding with the theoretical model Second, t his classification can logicall y fit the data in predictin g station groups for empirical analysis. In the adopter category for this study, early adopters and late adopters are used for a two group discriminant analysis in order to examine the differences b etween early adopters and late adopters. 4 103 stations (16%), which have 1997, 1998, 1999, and 2000 as the year of adoption, fall into the category of early adopters, whereas 123 stations (19%) which have 2006, 2007, 2008, and 200 9 as the year of adoption fall into the category of late adopters. Table 4 16 presents the adopter category used in this study Figure 4 3 illustrates the proportions of stations based on the adopter category Table 4 17 presents the results of discriminant analysis, including de scriptive statistics and predictive power of independent variables. Based on the group mean, early adopters had higher scores in vertical integration, competition, market size, income, and population density than late adopters. The canonical correlation wa s .722. 3 Rodgers (1995) classifies adopters i nto five categories, stemmed from the characteristics shown in the S shaped diffusion process: innovators (2.5%), early adopters (13.5%), early majority (34%), late majority (34%), and laggards (16%). This study uses approximately upper 16% of stations in terms of beginning year of digital TV adoption as early adopters, incorporating the categories of innovators and early adopters. The beginning years of early adopters include 1997, 1998, 1999, and 2000. Likewise, about lower 16% of stations in terms of beg inning year of digital TV adoption falls into late adopters, which corresponds to the category of laggards. 4 In discriminant analysis, the polar extreme approach can be used to compare extreme two groups, excluding the middle group (Hair et al., 2006).
101 Group centroids were located at 1.135 for early adopters and .950 for late adopters. The sig =.48, p <.01) indicates that there were overall significant differences between early adopters and late adopters. Among the significant independent variables, market size shows the most predictive power in distinguishing early adopters from late adopters The structure matrix from the analysis shows that early adopters were significantly different from late adopters in terms of market size (r = .646), ownership (r = .478), population density (r = .448), competition (r = .301), and income (r = .186). The usefulness of discriminant analysis is generally measured by its accuracy rate, the ability to classify observations accurat ely into the categories of the dependent variable. Table 4 18 displays the classification matrix that predicts group membership of the discriminant function. The discriminant function was more accurate in predicting early adopters (91.3%) than in predictin g late adopters (82.1%). The predictive outcome of the discriminant function had a hit ratio of 86.3%, showing a 36.3% improvement over chance probability. The result indicates that variables used in this study were effective for identifying influences on the timing of digital TV adoption. Table 4 16 Adopter c ategory Adopter Category Year of Adoption Number (%) This Study Rogers (1995) Early Adopters (N=103, 16%) Innovators (2.5%) 1997, 1998 32 (5%) Early Adopters (13.5%) 1999, 2000 71 (11%) Majority (N=417, 65%) Early Majority (34.0%) 2001, 2002 232 (37%) Late Majority (34.0%) 2003 2005 175 (28%) Late Adopters (N=123, 19%) Laggards (16%) 2006 2009 123 (19%) Total: 633 Total:633
102 Figure 4 3. Percentages of s tation s in a dopter c ategory Table 4 17 Discriminant analysis: Early adopters and late a dopters Mean (SD) Variable Early Adopters (N=103) Late Adopters (N=123) F Ownership .453 24.22 (9.25) 37.75(16.34) .801 55.75** Horizontal Integratio n 359 23.22 (12.42) 25.30 (17.25) .995 1.04 Vertical Integration 702 1.00 (.00) .94(.25) .969 7.10* Competition .132 12.72 (4.85) 9.52(5.29) .910 22.11* Market Size (log) 998 6.25 (.26) 5.73(.46) .688 101.74* Income (log) .236 4.57(.12) 4 .52(.11) .964 8.45* Pop Density (log) .205 3.75(.26) 3.47(.32) .821 48.83* Note = .48 x 2 (7, N=226) = 162.33 p <.01 *p < .05; **p < .0 1 Table 4 18 Classification of early adopters and late a dopters Predicted Group Early Adopters Late Adopters Observed Group N N % Correct Early Adopters (N=103) 94 9 91 .3 Late Adopters (N= 123) 22 101 82.1 Note. H it ratio (o ve rall prediction accuracy) = 86.3 % 5% 11% 37% 28% 19% 0% 5% 10% 15% 20% 25% 30% 35% 40% 1997-1998 1999-2000 2001-2002 2003-2005 2006-2009 % of Stations Year of Adoption
103 Summary The results of regression analyses showed that ownership, horizontal integration, vertical integration competition, market size and income had influence on the timing of digi tal TV adoption for the largest 25 TV group owned stations On the other hand, vertical integration, competition and market size had influence on the timing of digital TV adoption for the largest 11 to 25 TV group owned stations. Table 4 1 9 summarizes the findings of regression analyses focusing on the impacts of factors on the timing of digital TV adoption as well as the findings of independent samples t tests, ANOVA, and discriminant analysis. Figure 4 4 visually depicts t he importance of factors in reg ression analyses for the largest 25 TV group owned stations Figure 4 5 illustrates the importance of factors in regression analyses for the largest 11 to 25 TV group owned stations. Table 4 19 Summary of f indings Sample Research Question Influence / Diffe rence Direction Summary of Findings The Largest 25 TV Group owned Stations (N=633) RQ1. What, if any, influence d oes ownership have on the timing of digital TV adoption by broadcast stations ? Yes (Year of Adoption) A station whose TV group own er owns a fewer number of stations is more likely to have an earlier year of digital TV adoption. A station whose TV group owner owns a greater number of stations is more likely to have a later year of digital TV adoption. Yes (Adoption Period) A station w hose TV group owner owns a fewer number of stations is more likely to have a longer adoption period of digital TV. A station whose TV group owner owns a greater number of stations is more likely to have a shorter adoption period of digital TV.
104 Tab le 4 19. Continued Sample Research Question Influence/ Difference Direction Summary of Findings RQ2 What, if any, influence d oes horizontal integration have on the timing of digital TV adoption by broadcast stations ? Yes (Year of Adoption ) A station w hose TV group own er has less coverag e is more likely to have an earlier year of digital TV adoption. A station whose TV group own er has more coverage i s more likely to have a later year of digital TV adoption. Yes (Adoption Period) A station w hose TV group owner has less coverage is more likely to have a longer adoption period of digital TV A station whose TV group owner has more coverage is more likely to have shorter adoption period of digital TV. RQ3. What, if any, influence does ve rtical integration have on the timing of digital TV adoption by broadcast stations? Yes (Year of Adoption) + A station that is affiliated with a fewer number of networks is likely t o have a later year of digital TV adoption. A station that is affiliated w ith a greater number of networks is likely to have an earlier year of digital TV adoption. Yes ( Adoption Period) + A station that is affiliated with a fewer number of networks is likely to have a shorter period of digital TV. A station that is affiliat ed with a greater number of networks is likely to have a longer adoption period of digital TV. RQ 3 1. Is there a significant difference in the timing of digital TV adoption between network affiliates and independent s? No (Year & Period) N/A There is no significant mean difference in the year of adoption and adoption period between network affiliates and independents RQ 3 2. Is there a significant difference in the timing of digital TV adoption a mong ABC affiliates CBS affiliates NBC affiliates Fo x affiliates, other affiliates, and independents? Yes (Year & Period) N/A There are significant mean difference s in the year of adoption and adoption period among ABC affiliates, CBS affiliates, NBC affiliates, Fox affiliates, other affiliates, and indep endents.
105 Table 4 19. Continued Sample Research Question Influence/ Difference Direction Summary of Findings ABC affiliates are likely to have the earliest mean year of digital TV adoption, followed by CBS affiliates, NBC affiliates, Fox affiliates other network affiliates, and finally independents Independents are likely to have the shortest mean adoption period, followed by other network affiliates, Fox affiliates, NBC affiliates, and CBS affiliates, and finally ABC affiliates. RQ4. What, if a ny, influence does competition have on the timing of digital TV adoption by broadcast stations? Yes ( Year of Adoption) A station with a fewer number of stations in the market is lik ely to have an earlier year of digital TV adoption. A station with a g reater number of stations in the market is likely to have a later year of digital TV adoption. Yes ( Adoption Period) A station with a fewer number of stations in market is likely to have longer adoption period of digital TV. A station with a greater number of stations in market is likely to have shorter adoption period of digital TV. RQ5 What, if any, influence does market size have on the timing of digital TV adoption by broadcast stations? Yes (Year of Adoption ) + A station operating in small er market is likely to have a later year of digital TV adoption. A station operating in large r market is likely to have an early year of digital TV adoption. Yes ( Adoption Period) + A station operating in small er market is likely to have a shorter adopti on period of digital TV. A station operating in larger market is likely to have a longer adoption period of digital TV RQ 6 What, if any, influence does income have on the timing of digital TV adoption by broadcast stations? Yes (Year of Adoption ) A station operating in market with less I ncome is likely to have an earlier year of digital TV adoption. A station operating in market with more I ncome is likely to have a later year of digital TV adoption.
106 Table 4 19. Continued Sample Research Question Influence/ Difference Direction Summary of Findings RQ 6 What, if any, influence does income have on the timing of digital TV adoption by broadcast stations? Yes ( Adoption Period) A station operating in market with less I ncome is likely to have a long er adoption period of digital TV. A station operating in market with more I ncome is likely to have a shorter adoption perio d of digital TV. RQ 7 What, if any, influence does population density have on the timing of digital TV adoption by broadcast stati ons? No (Year of Adoption ) N/A Population density has no influence on the year of digital TV adoption No ( Adoption Period) N/A P opulation density has no influence on adoption period of digital TV. The Largest 11 to 25 TV Group owned Stations (N=315) RQ1. What, if any, influence d oes ownership have on the timing of digital TV adoption by broadcast stations ? No (Year of Adoption) N/A O wnership has no influence on the year of digital TV adoption No (Adoption Period) N/A Ownership has no influence on adoption period of digital TV. RQ2 What, if any, influence d oes horizontal integration have on the timing of digital TV adoption by broadcast stations ? No (Year of Adoption ) N/A TV group own er has no significant influence on the year of digital TV adoption. No ( Adoption Period) N/A TV group own er has no significant influence on adoption period of digital TV. RQ3. What, if any, influence does vertical integration have on the timing of digit al TV adoption by broadcast stations? Yes (Year of Adoption ) + A station affiliated with a fewer number of networks is likely to have a later year of digital TV adoption. A station affiliated with a greater number of networks is likely to have an earlier y ear of digital TV adoption Yes ( Adoption Period) + A station affiliated with a fewer number of networks is likely to have shorter digital TV adoption period. A station affiliated with a greater number of networks is likely to have a longer digital TV a doption period. RQ 3 1. Is there a significant difference in the timing of digital TV adoption between network affiliates and independent s? No (Year & Period) N/A There is no significant mean difference in the year of adoption and adoption period between network affiliates and independents
107 Table 4 19. Continued Sample Research Question Influence/ Difference Direction Summary of Findings RQ 3 2. Is there a significant difference in the timing of digital TV adoption a mong ABC affiliates, C BC affiliates, NBC affiliates Fox affiliates other affiliates, and independents? Yes (Year & Period) N/A There are significant mean difference s in the year of adoption and adoption period among ABC affiliates, CBS affiliates, NBC affiliates, Fox affiliates, other aff iliates, and finally independents. ABC affiliates are likely to have the earliest mean year of digital TV adoption, followed by CBS affiliates, NBC affiliates, Fox affiliates, other network affiliates, and finally independents Independents are likely to have the shortest mean adoption period, followed by other network affiliates, Fox affiliates, NBC affiliates, CBS affiliates and finally ABC affiliates RQ4. What, if any, influence does competition have on the timing of digital TV adoption by broadcast stations? Yes (Year of Adoption ) A station with a fewer number of in market is likely to have an earlier year of digital TV adoption. A station with a greater number of stations in market is lik ely to have a later year of digital TV adoption. Yes ( Adoption Period) A station with a fewer number of stations in the market is likely to have longer adoption period of digital TV. A station with a greater number of stations in the market is likely to have shorter adoption period of digital TV. RQ 5 What, if any, influence does market size have on the timing of digital TV adoption by broadcast stations? Yes (Year of Adoption ) + A station operating in small market is likely to adopt digital TV in a later year of digital TV adoption A station operating in large market is likely to adopt digital TV in an earlier year of digital TV adoption. Yes ( Adoption Period) + A station operating in small market is likely to have shorter digital TV adoption period A station operating in large market is likely to have longer digital TV adoption period.
108 Table 4 19. Continued Sample Research Question Influence/ Difference Direction Summary of Findings RQ6 What, if any, influence does income have on the timing of digital TV adoption by broadcast stations? No (Y ear of Adoption ) N/A Income has no influence on the year of digital TV adoption. No ( Adoption Period) N/A Income has no influence on adoption period of digital TV. RQ 7 What, if any, influence does population density have on the timing of digital TV adoption by broadcast stations? No (Year of Adoption ) N/A Population density has no influence on the year of digital TV adoption. No ( Adoption Period) N/A Population density has no influence on adoption period of digital TV. Early Adopters and Late Ad opters (N = 226) RQ8 What factors, if any, explain differences in the timing of digital TV adoption between two groups of stations: early adopters and late adopters? Yes (Year of Adoption) Early adopters are found to have significantly less ownership, l e ss horizontal integration, less competition, larger market size, lower income, and higher population density than late adopters.
109 Figure 4 4. Importance of f actors in the l argest 25 TV g roup owned s tations Figure 4 5. Importance of f actors in the l ar gest 11 to 25 TV g roup owned s tation s
110 CHAPTER 5 DISCUS SION AND CONCLUSION Digital technologies are bringing comprehensive change s t o the organization of the broadcast TV industry consisting of the production, the distribution, and the exhibition of prog rams T he full deployment of terrestrial digital TV broadcasts was accomplished through the digital TV transition. Since t he digital TV transition will bring more advanced TV services and cross pl atform applications, b roadcast stations may need to seek new business opportunities, focus ing on competitive advantag es The implementation of digital broadcasts. Since little empirical research has been conducted on digital TV adoption by broadcast stations, this study can contribute to the understanding of factors associated with the timing of digital TV adoption. T his chapter summarizes the findings in terms of the relationships among independent variables and the broadcast TV industr y as well as the contribution of each variable to the timing of digital TV adoption This chapter also discusses the i mplications of this study to academia, industry, and policymakers. Lastly, this chapter provides concluding remarks and the guideline for furthe r res earch Summary of Major Findings R esearch ques tions in this study investigate the influence of firm characteristics ( ownership, horizontal integration and vertical integration), market characteristics (competition and market size), and market demographics (income and population density) on the timing of digital TV ad option Multiple regression independent samples t tests ANOVA, and discriminant analysis were used as empirical analys e s to answer research questions. According to regression ana lyses for the largest 25 TV group
111 stations (Sample 1 ), t he independent variables in R Q 1 ( ownership ), R Q 2 ( horizontal integration), RQ 3 ( vertical integration ) R Q 4 ( competition ) RQ 5 ( market size ) and RQ 6 (income) were found to be significant in expl aining the year of adoption and adoption period Only po pulation density in RQ 7 had no significant impact on the year of adoption and adoption By contrast, t he results of regression analys e s for the largest 11 to 25 TV group stations (Sample 2) showed th at independent variables in RQ 3 (vertical integration), RQ 4 ( c ompetition) and RQ 5 ( market size ) had influences on the year of adoption and adoption period of digital TV On the other hand independent variables employed in RQ 1 ( ownership ) RQ 2 (horiz ontal integration), RQ 6 ( income ) and RQ 7 ( p opulation density ) had no significant impact on the year of adoption and adoption period of digital TV. Table 5 1 summarizes the impacts of inde pendent variables on the year of adoption and adoption period in S ample 1 and Sample 2 based on the results of correlation and regression analyses If a n independent variable was found significant in regression analysis but not significant in correlation analysis, the impact of th e independent variable on dependent vari able may be less powerful For example, vertical integration was not correlated wi th the timing of adoption but found to be significant in multiple regression When a n independent variable was found to be significant in both analyses the variable was con sidered to have a strong impact on the dependent variable. If a n independent variable was not significant in regression analys is, the variable was considered to have no impact regardless of th e results of correlation analysi s. For example, although populat ion density was highly correlated with the timing
112 of digital TV adoption in correlat ion analysi s it showed no significant impact in regre ssion analys i s. Table 5 1. Comparison of r esults of c orrelation and r egression a nalyses Sample DV IV Co rrelation Analysis Regression Model Impact (Sample 1) The Largest 25 TV Group Stations Timing of Digital TV Adoption Ownership Yes Horizontal Integration X Yes Vertical Integration X Yes Competition O Yes Market Size Yes Inc ome O O Yes Population Density O X No (Sample 2) The Largest 11 to 25 TV Group Stations Timing of Digital TV Adoption Ownership O X No Horizontal Integration X X No Vertical Integration X Yes Competition O Yes Market Size O O Yes I ncome O X No Population Density O X No Firm Characteristics Firms are distinguished by their characteristics that allow researchers to observe differences in firm behavior. The current study employed ownership, horizontal integration, and network aff iliation as firm characteristics to investigate their influence on the timing of digital TV adoption. The results of regression analyses indicate that firm characteristics generally had significant impacts on the timing of digital TV adoption, with the exc eption of ownership and horizontal integration for the largest 11 to 25 TV group owned stations. The findings are broadly cons istent with the findings of previous litera ture, which has provided mixed evidence on the influence of firm characteristics on the timing of adoption. This study found that ownership and horizontal integration had negative influence on the timing of adoption. A stream of adoption research suggests that leading firms can lag behind non leading firms in adopting innovation when leading firms are faced with ra dically different technologies Christensen (1997) suggests that a
113 number of leading firms tend to wait until a new market is grown enough, when they are challenged by disruptive innovations. Ownership As discussed in Chapter 2, ow nership has been frequently explored in previous station level studies to understand the behaviors of broadcast stations (Napoli, 2001; Yanich, 2010). The results of this study suggest that a station whose TV group had a fewer number of stations was likely to adopt digital TV earlier. One possible explanation is that a group that has a larger number of station s may not have access to more secure programming res ources for digital TV adoption For example, Ion, Sinclair, and Nexstar are the three groups that have largest number of stations. However, all of them have no ownership interests by major media conglomerates, as ABC, CBS, NBC, and Fox TV groups do. Since the availa bility of digital programming is likely to be one of the important incentives for broadc asters in adopting digital TV it is inferred that the lack of li nk between the TV groups that own more stations and major media conglomerates may contribute to the negative direction of ownership on the timing of digital TV adoption. The result also sugge sts that ownership is a significant factor in the timing of digital TV adoption for the largest 25 TV group owned stations but not for the largest 11 to 25 TV group owned stations (Table 5 2). Horizontal i ntegration Horizontal integration, measured by t he percentage of TV households that a level of penetration into national TV market A firm s horizontal integration is often associated with a larger business with a strong position in the market in ter ms of operation and marketing. The result of analysis indicates that a station with less horizontal integration is likely to adopt digital
114 TV earlier. Although seemingly surprising, this result likely supports previous literature of innov ation adoption by first movers. From standpoint of the RBV approach, the existing resources and capabiliti es of large incumbents may impede innovation adoption. Christensen (1993) found a common pattern of late adoption of new generations of computer disk drives by incum ben ts with large market shares. As a discontinuous innovation and c ompetence destroying innovation, digital TV is the new generation TV service that requires radical changes in technology and business. Thus, it is likely that e xisting resource s and capabiliti es gained from horizontal integration may retain less continuing value with digital TV adoption. In addition integrated firms likely have complex bureaucracy that may slow down the decision making of innovation adoption (Hall & Khan, 2003). Vertical i nte gration As a strategic choice, vertical integration has been considered as on e of the factors that influence innovation adoption. A major discussion on the role of vertical integration is that vertically integrated firms can increase market power in releva nt industries (Chipty, 2001) When vertical integration is combined with market power, firms can maintain strategic flexibility (Harrigan, 1985 ) In the broadcast TV industry, vertical integration provides opportunities for network programming to expand co nsumers (Waterman, 2000 ) The results of multiple regression show ed that v ertical integration is a n overall significant factor in the timing of digital TV adoption (Table 5 2). The results suggest that the more stations vertical ly integrat ed the earlier t hey tend to adopt digital TV. This is consistent with previo us studies supporting the positive role of vertical integration in the ti ming of adoption (Boeker & Huo 1998 ; Teece, 1996 ) In assessing the position of independents in the TV ma rket compared to affiliate, Gomery
115 (1984) suggests that are not equivalent to those of affiliates since independents tend to have less advertising revenue. different composition of viewers and less established revenue structure may lead independents to c hoose to adopt at a later time than affiliates. N etwork affiliat ion is one of the important structural elements of broadcast TV stations T he widespread appeal of network programming due to network affiliation can help affiliates to provide a large number of viewers to advertisers Focusing on the importance of network affiliation to stations this study conducted additional analyses on the differen ces in timing of adoption using vertical integration. The results of ANOV A revealed statistical differences in the timing of digital TV adoption among the group means of ABC affiliates, CBS affiliates, NBC affiliates, Fox affiliates, other affiliates, and independents. ABC affiliates had the earliest mean year of digital TV ado ption, while independents had the latest mean year of digital TV adoption. The results that ABC had the earliest timing of digital TV adoption than other network affiliates appear to reflect ABC has shown a continued interest in o nline diversification of their programs, actively participat ing in new digital distribution channels to generate new revenue streams (Disney, 2010). To illustrate, ABC was the first major network to distribute full length program episodes online using its own ABC.com. ABC also sells their shows through Apple's iTunes store. In 2009, ABC became a partner of Hul u along with NBC and Fox In the deal with Hulu, venture and provided ABC broadcast service thus might contribute to early adoptio n of digital TV by ABC affiliates.
116 T he results are also k competition in HD programming in 2001. First, GAO (2002 a; 2002b ) identifies that HD programming as a generally accepted factor that encourages consumer adoption of digital TV. Second, GAO (2002 a, 2002b ) finds that ABC and CBS were digital leaders by prov iding almost all their primetime programming in HDTV. For example, ABC provid ed almost all of their primetime programs in HD during the 2001 2002 season (GAO, 2002a). Also, CBS reported in the 2001 survey that the ir affiliates provid ed more HDTV progr ammi ng than those of other major networks which was an average of 33 ho urs per week (GAO, 2002b) On the other hand, NBC and Fox were found to be lagged behind ABC and CBS in terms of the amount of HD programming. NBC had relatively little and Fox had no HDTV primetime programming during the 2001 2002 S easo n (GAO, 2002a) In addition to ANOVA, t he independent samples t test s were conducted to compare the mean differences between affiliat es and independents The results of the independent samples t test s did no t show significant difference s in the t iming of adoption between the two groups of stations O ne possible explanation for this outcome could be that the sample sizes of affiliates and i ndependents were incomparable in the analysis ; th e number of independen ts (N = 26 ) was too small when compared to that of affiliates (N = 607). Market Characteristics Since innovation adoption can be a result of interactions of various factors in the market, market characteristics can determine the success of innovation ado ption. Market characteristics, usually uncontrolled by management of an individual firm, are perhaps most commonly explored variables in predicting innovation adoption by firms (Frambach, 1993; Stoneman & Diedren, 1994). This study examines the role of
117 com petition and market size in the timing of digital TV adoption. The results of regression analysis indicate that competition and market size were significant factors that determine the timing of digital TV adoption. Competition The results of regression an alysis suggest that competition significantly contributes to the timing of digital TV adoption (Table 5 2). A station with less competition in market was more likely to choose to adopt digital TV earlier. This finding suggests that, as Geroski (2000) point s out, excessive competition can slow down the adoption of innovation because more competition tends to lower the returns to adoption. It is inferred that decreased number of stations in a market might increase the returns with digital TV adoption since a ll stations had to adopt digital TV by mandated decision on the timing because stations knew that all others would eventually adopt digital TV. Thus, it is possible that the mainstream firms want to delay their decision until they ensure that regulations and technologies do not significantly change. Without mandates Hoppe (2002) suggests that the incentives for adoption are likely to be higher for firms in competition be cause they have an incentive for preempting other competitors. Copeland and Shapiro (2011) also found that that the rate of computer market. Market s ize In this stud y, market size is measured by the number of TV households in the TV market. The results of regression analyses show that market size was the most powerful predictor of the timing of adoption (Table 5 2). The findings suggest that
118 stations operating in larg e markets are likely to adopt digital TV earlier. The findings are consistent with previous literature indicating that firms in large markets have the potential det erminant for adopting innovation (Frambach, 1993; Hannan et al., 1987; Milliou & Petrakis, 2 009). This study supports that the size of the market positively contributes to the timing of digital TV adoption decision by broadcast sta tions. The FCC also acknowledged the importance of market size when the Commission established the timelines of digit al TV adoption for broadcast stations. The FCC required the major four affiliates in the largest 10 TV markets, which represented more than 30 percent of TV households, to broadcast in digital signal by May 1, 1999. In addition, th e FCC required that the s tations in the largest 30 TV markets which represented more than 50 percent of TV households, must build out digital facilities by 2002. The FCC identified that a majority of digital TV adoption had occurred as of 2003, the following year after the 2002 t in policy (FCC, 2003). Market Demographics This study employed income and population density as independent variables to explore the role of market demographics in the timing of digital TV adoption. For the largest 25 TV group sta tions, only income has a significant influence, while population density has no influence on the timing of digital TV adoption. For the largest 11 to 25 TV group owned stations, income and population density do not contribute to timing of digital TV adopti on. One stream of research argues that innovation adoption is more likely available in locations with high income and population density (Forman, Goldfarb, & Greenstein, 2003 ; Garcia Murillo, 2005).
119 Income The results of regression analysis demonstrate th at income has an influence on timing of digital TV for the largest 25 TV group stations. The results indicate that a station in market with less income likely adopted digital TV earlier. Different from cable or satellite TV, broadcast TV do not directly de pend on subscription fees for revenues. In this case, local demand may be less re sponsive to income in a market. Consumers in the market with less income may not afford to subscribe cable or satellite TV service, or they are more likely to drop such subsc ription service. In previous studies, Income generally has shown a positive impact on innovation adoption at the individual level (Rogers, 1995) and in telecommunication services (Beilock et al., 2003). By contrast, it is argued that technology adoption is less constrained by demographics; income is less a predictor of early adopter behavior because technologies are available for almost free or at low prices. Since receiving over the air digital TV technically cost little, it would lessen the positive influ ence of income on digital TV adoption. This study gives support to the idea that differences in income across markets may explain different timing of adoption. Population d ensity Population density had no influence on the timing of digital TV adoption (Ta ble 5 2). The results do not support previous studies on the role of population density in the adoption of the Internet (Forman et al., 2003; Garcia Murillo, 2005). A possible reason for the results may be found in the business model of broadcast TV statio ns. Since broadcast TV does not directly depend on subscription fees for their revenue, local demand of digital TV may be less responsive to population density compared to pay TV or telecommunication services that require wired networks. Another reason can be
120 found in the mandatory nature of the digital TV transition. Since stations must adopt digital TV to meet the FCC timeline for the transition, population density, which can be an index of local demand, less li kely affects the decision of timing by stati ons. With respect to non mandated adoption of the Internet researchers (Forman et al., 2003; Garcia Murillo, 2005) find out that innovation adoption is more likely available in locations with high income and population density. In particular, Forman et al (2003) suggest that adoption will be most extensive in large cities with high population densities because such areas are more likely to pooling resources, which can decrease the costs of adoption. Table 5 2 compares the results of regression analyses, s howing their influence and the level of significance. Table 5 2 Determinants of the t iming of d igital TV a doption Owner ship Horizontal Integration Network Affiliation Competition Market Size Income Population Density L argest 25 TV Group owned Stati ons (N= 633); Year of Adoption Yes ( p <.01) Yes ( p <.01) Yes ( p <.01) Yes ( p <.0 5 ) Yes ( p <.01) Yes ( p <.0 5 ) No Largest 25 TV Group owned Stations (N= 633); Adoption Period Ye s ( p <.01) Yes ( p <.01) Yes ( p <.01) Yes ( p <.0 5 ) Yes ( p <.01) Yes ( p <.0 5 ) No Largest 11 to 25 TV Group owned Stations (N= 315); Year of Adoption No Yes ( p <.01) Yes ( p <.0 5 ) Yes ( p <.01) No No Largest 11 to 25 TV Group owned Stations (N= 315); Adoption Period No Yes ( p <.01) Yes ( p <.0 5 ) Yes ( p <.01) No No
121 Measures of the Timing of Adoption T his study determined that using the year of adoption as a single measure of the timing of adoption would not b e sufficient for precise analysis For example, the two stations in the sample WDCW in Washington D.C. and KAAH in Honolulu adopted digital TV i n 200 2 and 2003, respectively. H owever, the time difference in adoption between th e two stations i s 12 days while the stations have two different year of adoption For a more accurate assessment this study include d adoption period, the number of days be tween the beginning date and the completion date of digital TV adoption as a second measure of the timing of digital TV adoption Using adoption period as a measure makes it possible to track any difference between two measures that might exist during the entire timing digital TV adoption However, the findings of this study show that the results d id not vary greatly by the different measures of the timing The significant factors in the results of analysis using the year of adoption were also found to be significant in the results of analysis using adoption period, and vice versa. U sing year of adoption as a measure of the timing of adoption gave a slightly better fit [ R 2 = .2 23 (Sample 1) and R 2 = .25 5 (Sample 2)] in the regression model than using adopt ion period [ R 2 = .2 1 8 (Sample 1) and R 2 = .240 (Sample 2 ) ] Sample Compositions The s ample of this study initially consisted of the stations ow ned by the largest 25 TV groups In the sample, the proportion of the largest 10 gr oup owned stations (N = 318) was 51% of entire sample stations (N = 633). Since the proportion of the largest 10 group owned stations was more than the half of the sample, this study expected that that t he inclusion of the largest 10 group owned stations would likely affect the resul ts of analysis Thus, a s ample subset was created by excluding the large st 10 group owned
122 stations Th is study thus ex amined and compared the factors that influence the timing of digital TV adoption in two different samples : t he largest 25 group owned stat ions (Sample 1) and the largest 11 to 25 group owned stations (Sample 2) For the larg est 25 group owned stations, all independent variables were statistically significant with the exception of population density For the largest 11 to 25 group owned stat ions, vertical integration, competition, and market size were statistically significant factors. Table 5 3 shows factors that influence the timing of adoption in resul ts of analysis for two samples. V ertical integration, competition, and market size were c ommon significant factors for both samples. Conversely, ownership, h orizontal integration and income were significant factors only for the largest 25 group owned statio ns, but not for the largest 11 to 25 g roup owned stations (Table 5 3 ). Table 5 3 Commo n d eterminants of the timing of digital TV a doption DV Measure Common Determinants for the Largest 25 and t he Largest 11 to 25 TV Group owned Stations (Sample 1 & Sample 2) Determinants for t he Largest 25 TV Group owned Stations Only (Sample 1) Determinan ts for t he Largest 11 to 25 TV Group owned Stations Only (Sample 2) Year of Adoption Vertical Integration Competition Market Size Ownership Horizontal Integration Income N/A Adoption Period Vertical Integration Competition Market Size Ownership Horizon tal Integration Income N/A Early Adopters and Late Adopters This study conducted discriminant analysis to explore the differences in early adopters and late adopters in terms of independent variables. Ownership, horizontal integration, competition, mark et size, income, and population density are found to be significant factors that explain the differences between early adopters and late adopters. The results of discriminant analysis indicate that early adopters ha ve the pattern of
123 fewer number s of statio ns own ed by their group owner less horizontal integration, less competition, larger market size, lower income, and higher population density. Although population density was not a significant factor in regression analysis, the results of discriminant anal ysis show that population density is a significant factor that differentiate early adopters and late adopters. Since discriminant analysis in this study compared two extreme groups in the sample, it appears that popularity density is found to be a signific antly predictor that differentiate early adopters from late adopters. The results of the analysis suggest that e arly adopters tend to have higher population density in their market than late adopters. In addition, m arket size is consistently found to be th e strongest predictor as in regression an alysis The findings support previous research that emphasizes the importance of market size in innovation adoption (Acemoglu et al., 2003). When deciding to become an early adopter or a late adopter, stations invol ved with digital TV adoption would consider the current technology level is superior to alternative technologies which might be available in the future, the supply infrastructure such as programming and equipment premises available for stations, the base o f consumers who would a dopt digital TV. Thus, the timing of digital TV adoption by stations is likely to depend on the size of digital TV consumers and anticipations about market growth and technology development (Geroski, 2000). These corpo rate level cons iderations may also explain the differences between the timing decisions by early adopters and late adopters. The findings from discriminant analysis clea rly demonstrated the difference between the early adopters and late adopters. The reason for different timing choice is explained by the trade off between early adoption and late adoption. Identifying the characteristics of early adopter firms has been a major
124 research area in the timing of innovation adoption. The results of discriminant analysis are cons istent with previous studies that identified early adopters and later adopter. For example, Hollenstein (2004) explains the factors that determine the timing of technology adoption in information technology. W hile early adoption can give competitive advant age of increasing market share to early adopters they need to deal with high costs for technology investment and uncertainty (Hoppe, 2000). By contrast, late adopters can reduce cost s associated with adoption, although they may be placed in a disadvantage ous position i n the market (Hoppe, 2000). Theoretical Implications Since the digital TV transition has scarcely been examined from the industry perspective l ittle has been revealed about the contribution of factors in the timing of digital TV adoption. T h e findings of this study emphasize the importance of theoretical frameworks in st udying the timing of digital TV adoption since the results generally validated the applicability of firm characteristics and market characteristics to digital TV adoption C onsidering the mixed evidence on the contribution of factors, t he findings of this study are largely consistent wit h previous research, suggesting the influence of ownership, horizontal integration, vertical integration, competition market size, and incom e on the timing of digital TV adoption. T h is study indicate s that less horizontal integration is likely related to earlier adoption of digital TV This finding confirms the argument that less integrated firms may adopt innovation more readily due to their responsiveness to the technology c hanges (Cohen & Levin, 1989). While some studies (Teece, 1996) suggest that horizontally integrated firms may readily take on innovation adoption based on greater pool of resources Roberts et al. (1994) suggest that hori zontal integration may not work in itself.
125 H orizontal integration of a firm may need a larger firm size and structure with central control in order to ensure its viability. However, this increase in firm size may not work with digital TV adoption. I n the b roadcast industry, large TV groups are increasingly looking for cost effective technologies and computerized workflows without increasing their expenditures and employe es. Digital technology likely affects the operation in large TV groups in a way that the y can share expertise between stations or outsource their works ( Winslow 2011). In examining the influence of vertical integration this study found that vertical integration had a significant influence on the timing of adoption among independents ( no aff iliate ) stati ons affiliated with one network ( single affiliate ) and stations a ffiliated with more than one network ( multiple affiliate ) T he finding s imply that a station with more affiliated networks is likely related to earl ier adoption of digital TV Most f irm behavior theories explain that adopting innovation in ver tically integrated firms often depends on other fir m specific factors such as firm size (Astebro, 2002 ; Teece, 1996 ). Yoo (2008) also notes that the conditions under which a network affilia te operate do not match those of its competitors The result s of this st udy suggest that a station w ith less competition in its market was likely to adopt digital TV earlier. T his finding is consi stent with previous literature in other media studies. Hami lton and McManus (2005) observe that an increase in the number of firms in the market delays adoption in the U.S. newspaper industry Additionally, Saloner and Shepard (1995) provide evidence that innovation adoption occurs faster in more concentrated mar kets The use of marke t demographics in this study is relatively new, since little media adoption research has examined the im pacts of market demographics on innovation adoption. With the exception of income in
126 regression analyses for the largest 25 TV gro up owned stations income and population density had little impact on the timing of digital TV adoption Despite their insignificance the inclusion of market demographics would be appropriate since the variables represent market potential s and local deman d associated with d igital TV adoption. In this respect, the analysis of market demographics would be necessary for more refined research. T he results of this study suggest that the timing of digital TV adoption may not be necessarily motivated by market p ower and a great er pool of resourc es Since the transition to digital TV would be an expensive decision for broadcast TV stations, a number of stations might cho o se to delay the transition. The disadvanta ges of becoming an early adopter are discussed in th e study by Lieberman and Montgomery (1988) ; e arly adopters typically be ar large investment expenses to adopt technology innovation. E arly adopters also likely adopt technology innovation with uncerta inty, until the adoption is found to be beneficial to the m. Similarly, early adopters may face difficulties in ad justing new environmental changes since technologies and regulations are likely to change after their adoption S tations equipped with such r esources may choose to be early adopter s of digital TV rega rdless of the uncertainties, such as the risk of huge sp ending and little revenue gain ( Lieberman et al., 1988 ). s discussion may be applicable to exp lain gain and loss when they decide the timing of d igital TV adoption Industry Implications In discussing the implications for the broadcast TV industry, it is important to note that t he structure of the indus try is largely affected by federal policies including the grant ing and renewal of licenses (Ca rter et al., 2008) Although local broadcast TV
127 markets vary according to their economic and structural conditions the unique market structure of the broadcast TV industry plays a greater role in structure than in other general businesses A broadcast TV market is often comprised of a relatively small number of stations in the same market with high entry barriers to newcomers such as high investment, licensing requirem ents and regulations. In the regression analysis in which the largest 10 TV group owne d stations were exclude d from the sample, horizontal integration had little influence on the timing of digital TV adoption Also, income had little significant impact on the timing of digit al adoption in the analysis of the largest 11 to 25 TV group owned stations The se results impl y that stations owned by largest 10 TV groups are likely associated with thes e factors in the decision as to when they should adopt digital TV. With regard to the impact of market size and vertical integration t he dataset in th is study illustrates that independents operate mostly in large markets. For example, the smallest market in which independents operated is Kansas City, the 32 nd largest TV market, in the sample of this study. More populated c ities in large TV markets have a greater number of independents, while smaller markets have few, if any (Gomery, 1984). In comparing large market size might not contribute to the timing of digital TV adoption because the number of indep endents was much small er than affiliates acros s the country. The industry implications for broadcast TV stations in this study are twofold. First, broadcast TV stations need to develop business models that are economically viable and attractive to consume rs. Offering value added digital service s can be a choice to adapt in changing digital environment. The change in the broadcast TV market after the
128 transition would be beyond offering digital TV broadcasts. Digital technology enables consumers to download and watch broadcast TV programs with no restriction of time and place Thus, as illustrated in the cases of digital leaders in the broadcast TV market ( s ee Chapter 1), broadca st TV stations can add multimedia features and applications to their service usin g the Internet or wireless networks In this respect m obile digi tal TV is a promising area in which broadcast TV stations can gain competitive advantage s over their competitors including MVPD providers Mobile digital TV can create new revenue stream, as well as a means to reach new consumers, since mobile digital TV can deliver streaming and multimedia content at lower cost per viewer 1 Second, broadcast TV stations need to build relationships and adjust their position with network providers, program prov ider s, consumers, and advertisers. In particular, s tations need to seek strategic partnership to reach untapped niche markets. Building relationship with Internet service providers, wireless c arriers, and cable operators can enable broadcast stations to us e multiple platform s in programming delivery. B roadcast stations can find additional value to their traditional business from the opportunities created by digital TV adoption Policy Implications A major objective of the U.S. di gital transition was to en hance the international competitiveness of industries through providing advanced digital broadcasting service to the public (Hart, 2004). The t iming of digit al TV adoption was a n important issue for relevant industries and regulators since digital TV adop tion is a primary means to achieve th e national goal of economic growth From a policy standpoint, t he digital TV 1 Unlike one to one data transmission, mobile digital TV uses to that has relatively low implementation costs (Harris Corporation, 2011).
129 transition in the U.S ha d two major areas of concerns One area was how to ensure broadcast TV stations and consumers to adopt digit al TV. 2 T o make a smooth transition from analog to digital TV the FCC allocated broadcast stations an additional 6 MHz channel with which to begin digital broadcasts Another policy area was related to spectrum management and allocation. The public can benefit fro m the digital TV transition that offers increased spectrum efficiency. The more efficient use of spectrum allows the reallocation of broadcast spectrum for industries and the public 3 Although this study focuses on investigating digital TV adoption at the firm level policy implications of digital TV can be provided by examining what happened in the progress of the digital TV transition. The first point is associated with the FCC s decision on the timing of digit al TV adoption Since the digital TV transiti on is part of the national spectrum arrangement plan d eciding the deadline for digital TV transition was largely dependent on political negotiation and administrative convenience (Hart, 2004) It seems that the needs of the industry were neg lected when th e FCC advanced toward the digital TV transition If the FCC was more concerned about stations readiness in the process of transition the FCC would have set forth a more realistic deadline reflecting industry needs and capabilities by reopening the discus sion with Congress. As such policymaking activities played an important role in determining timing of digital TV adoption. 2 The FCC (1997) stated several goals to encourage the development of digital TV : 1) to ensure confidence and certainty in the digital TV transition; (2) to increase the availability of new products and services to consumers; (3) to encourage technological innovation and competition; and (4) to minimize regulation 3 The FCC passed m easures to improve spectrum availability and efficiency as well as spurring innovation in November 2010. First, the FCC expects to reallocate 120 MHz of spectrum from broadcast stations to wireless broadband service. Also, spectrum will be available for ex perimental programs to advancing new technologies.
130 The second p oint is the role of market size in de termining the timeline for the transition. In the the FCC employed market size and affiliation with established networks as the standard for graduated transition. This tiered adoption rules required that stations affiliated with ABC, CBS, NBC, and Fox operating in the largest 30 markets had to adopt dig ital TV earlier than those in smaller markets. The F CC clarified that the reason for th is regulation was in the largest television markets can be expected to lead the transition to DTV and that these stations are better situated t o invest the capital necessary to establish the first DTV stations (FCC, 1997 ; FCC, 2001 ) The FCC also recognized stations produce less revenue than those in large markets, adversely affecting their ( FCC, 1995). Thus, the FCC expected that market stations will find it easier to begin DTV service after learning from the experience gained by the larger market stations (FCC, 1997 ; FCC 2001 ). The purpose of the FCC s staggered approach was to inc rease the penetration level of digital TV into TV households (FCC, 2001). the pertinent predictors that this study particularly identified as having a positive influence in the timing of digital TV adoption Thus, t he FC is evaluated as an effort to adjust digital TV ad option according to the different firm and market characteristics of stations. An important implication of this study is t hat the digital TV transition was planned and coordin ated by the gover nment, and such mand atory nature was a unique quality of the digital TV transition differentiating the transition from other innovation adoption A s a regulatory authority t he FCC has played an important role in initiating innovation policies Particular ly in the digital TV transition, t he FCC had established t he technical
131 standards for digital TV before stations adopted digital TV In other cases, technical standards were achieved through competition in the market. For example, the FCC allowed the marke tplace to set the standard amon g competing technical standards in adopting AM stere o ( Berg, 1989; Ducey & Fratrick, 1989 ). It appears that l ess government involveme nt in adopting AM stereo resulted in different paths to adoption and standardization from th ose in digital TV adoption ; the market approach tended to slow the adoption of AM stereo b y the indu str ies and consumers. Lessons from U.S. Digital TV Transition The digital TV transition has a potential to benefit the national economy, relevant industrie s and consumers. While the process of and requirements for digital TV transition w ould vary across each country depending upon its market condition and policy, t he U.S. experience would provide a lesson to countries working on the digital TV transition 4 A major obstacle in implementing the digital TV transition was the delay in the transition deadline. After t he initial transition deadline was scheduled on Dece mber 31, 2006, it was pushed back to February 17, 2009, and delayed again to June 12, 2009. The delay not only invoke d disagreements among stakeholders but also cause d confusion for consumers due to insufficient consumer education about the transition In this respect, Adda and Ottaviani (2005) suggest that substantial c oordination among stakeholder s is needed for digital TV transition to take place. Also, t he GAO (2007) notes that lack of consumer educa tion was critical in causing delays to the original transition schedule T he delay in the digital TV transition reflect s that the 4 Goldstein (2008) notes three issues that some U.S. stations have to resolve. These issues are; (1) technical issues, such as the relocation of digital an tenna s ; (2) coordination issues with foreign broa dcasting authorities in Canada or Mexico, with cable pr oviders and satellite companies in the U.S. or (3) other issues, such as the construction of digital towers or financial constraints that might hinder the progress of the tra nsition.
132 digital TV transiti on is a set of complicated decision s that involve market technology, politics, economics, and regulation. Another issue with the digital TV transition was to provide incentive s f or related industries and consumers. As discussed in Chapter 1, in order to s uccessfully implement the digital TV transition, the chicken and egg coordination problem needs to be solved through creating incentives for related stakeholders ( Ducey et al, 1989; GAO 2002 a ) To c reate incentiv es f or industries and consumers to adopt d igital TV the government can offer subsidies or tax exemption s to encourage the transition In the U.S, 6 MHz of additional spectrum was allocated to broadc ast stations prior to the full digital broadcasting. Also, a subsidy program for consumers was admi nistered to give consumers incentives in purchasing digital TV equipment. Additionally, regulators need to pay more attention to increase the provision and consumption of advanced digital TV service The benefits of HDTV to consumers, including advanced v ideo qualities and aspect ratio, have been largely promoted by broadcasters and equipment manufacturers. Nevertheless, a recent survey in 201 1 shows that 63.3 % of TV households were able to receive HDTV progra mming while 6 7 1 % had HDTV sets (Televi sion Bur eau of Advertising, 201 1 ). Another recent survey in 2011 reveals that many broadcast TV s tations have just started considering program production in HDTV Only 38 % of surveyed stations have accomplished in h ouse HDTV production fully. Also, only 45% of the stations have HD control rooms the basic facility necessary for HDTV production (prweb.com, 2011). Th e results indicate that more industry and government efforts need to be made to produce and deliver HDTV programmin g I n the midst of technical and econo mic shifts countries in the other parts
133 of the world are facing challenges in completing the digital TV transition 5 The U. S. experience would present key lessons for a successful transition to other countries. Efforts for coordination consumer education and creating incentives f or related industries and consumers would be necessary to ensure the complet ion of digital TV transition in a timely manner. For future technology adoption whether government mandates the adoption or leave s it to the market sho uld be determined and evaluated based on the nature of the technology and its potential to the market and consumers. D igital TV is a powerful medium that has overarching impact on broadcasters, consumers, and a number of related industries providing inter faces with other communication systems and advanced TV services to consumers. Considering the broad impact of digital TV transition on a number of was a necessary policy choice. Mo reover, t he digital TV transition was a government led program that had the goal to promote economic growth and social development of the nation. Both broadcast TV stations and consumers had to adopt digital TV for the transition to be successful. The prod uction, distribution, transmission and reception of TV programs experienced overall changeover. Therefore, it was necessary to provide stations with guidelines in order to implement the transition in timely and orderly fashion. G overnment mandates also mi ght be a cost effective policy option for stations and the related industries to expedite the transition The completion of digital TV transition not 5 A number of coun tries are implementing the digital TV transition, including Canada (August 31, 2011), France (November 30, 2011), Italy (December 12, 2012), Korea (December 31, 2012), and Australia (December 31, 2013). Some western European countries such as Germany (Nov ember 25, 2008), Denmark (November 1, 2009), Belgium (March 1, 2010), and Spain (April 3, 2010), have completed digital TV transitio n (The date in parenthesis is the final transition date).
134 only enabled s tations to save the cost of offer ing both digital and analog signals but also the analog sp ectrum to be converted to more efficient uses Conclu ding Remarks This study noted a gap in previous adoption studies that suggest t he need for industry analysis and then art iculated the factors that influence timing of digital TV adoption based on innova tion adoption theories and firm behavior theories. This study suggests that there exists a trade off between the choice of early adoption and late adoption. Early adopter s can incur high cost s for digital investment and uncertainty Hoppe (2000) indicates that innovation adoption can give competitive advantage s to the firm that adopts early by increasing its market share. By contras t, late adopters can reduce cost s associated with adoption although they may be placed in a disadvantageous positi on i n the ma rket compared to early adopters In explaining the differences in the timing of adoption by early adopters and late adopters, this study suggests that early adopters tend to operate in large markets with low income and high population density. On the other hand, late adopters tend to operate in small markets with high income and low population density. The purpose of this study was to explore the influence of firm characteristics, market characteristics, and market demographics on timing of digital TV adop tion by broadcast TV stations An instructive finding of this st udy was that m arket size is the most significant factor in predicting the timing of digital TV adoption. Also, this study found common factors that influence the timing of digital TV adoption as vertical integration, competition, and market size. Conversely ownership, hori zontal integration, and income are the factors that influence the timing of digital TV adoption when the analyses included the largest 10 TV group owned stations. These resul ts suggest that
135 firm characteristics and mar ket characteristics were influencing the timing of digital TV adoption T he contribution of independ ent variables to the timing of adoption suggests the importance of structural characteristics of broadcast TV st ations as well as the continuing validity of existing variables generally used in innovation adoption studies Ano ther important finding is that th is study identifies the difference in the timing across five different affiliates and independents S ignific ant differences are demonstrated across five groups of network affiliates and independents. Limitations and Sugge sted Fu rther Studies While this study identified and explain ed factors that influence the timing of digital TV adoption, t h ere are several li mitations to this study in relation to further studies First, t his study needs to include additional variables that may explain more dimensions of digital TV adoption. E xamining separate effects of firm characteristics and market characteristics may not b e sufficient to explain complex decisions associated with digital TV adoption. It is recommended that future research develop more diverse factors based on theor ies For example, the strategies of cable and satellite systems, as major carriers that provide broadcast programming, may affect adoption decisions by broadcasters. Thus, environmental factors including cable and satellite systems can be incorporated into investigating the timing of digital TV adoption Second, it is also important to evaluate p ost transition programming and business strategies for broadcasters now that the transition has been completed One of the major post transition challenge s would be the programming issue As with the case of basic cable TV, digital TV programming can be d elivered using two different types of channels: general mass appeal channels and niche oriented channels. Future studies can focus on how to fill out the increased channels and creat e consumer
136 demand for new channels. Another challenge to stations is how t o improve performance capabilities. Firms can develop competitive advantages and outperform other competitors through adopting innovations that can lead to a higher market performance. Future studies can explore the changes in firm performance by broadcast stations after the adoption o f digital TV. Third while t he digital TV transition presents opportunities for the future of broadcasting the question of how stakeholder s in broadcasting will adapt to the changing digital environment remains. This study s uggests that multiple platform strategy is appropriate for broadcasters to maintain competitive in the digital environment. The popularity of internet streaming TV services, such as Hulu .com and Netflix illust rates that broadcasters may need to focus on t heir online presence and other platforms for program delivery Besides examining broadcasters, researchers could investigate the factors that influence strategic and managerial choice for cable and satellite TV systems Although broadcasting has been devel oped and organized by local broadcast stations in the U.S., the broadcast TV stations are facing competition from cable and satellite TV. Since the majority of con sumers receive broadcast TV programs via cable or satellite systems, their strategic choice w ould be important to broadcasters. This research also can be extended to examine strategic choice for other MVPD providers facing digital challenges and opportunities
137 LIST OF REFERENCES Acemoglu, D. & Linn, J. (2004) Market s ize in i nnovation: Theory and e vidence from the p harmaceutical i ndustry Quarterly Journal of Economics 119(3), 1049 1090 Adda, J ., & Ottaviani M (2005) T he t r ansition to digital television. Economic Policy 41 160 209. Albarran, A. (1998). Media economics: Research paradigms, issues, and contributions to mass communication theory. Mass Communication & Society 1(3/4) 117 129. Albinial, P. (2010). B&C's top 25 station groups 2010. Broadcasting & Cable 140(15), 14 20. Retrieved from EBSCOhost Allison, P. D. (1982). Discrete time methods for the analysis of event histories. Sociological Method ol ogy 13 61 98. Allison, P. D. (1999). Multiple regression: A primer Thousand Oaks, CA: Pine Forge Press. Atkin, D., Jefferies, L., & Neuendorf, K. (1998). Understanding internet adoption. Journal of Broadcasting & Electronic Media, 42 (2), 475 490. Atkin, D., Neuendorf, K. Jefferies, L. ( 2003 ). Predictors of audience interest in adopting digital Television Journal of Media Economics 16 ( 3 ), 159 173 Audretsch, D., Baumol, W., & Burke, A. (2001). Competition policy in dynamic markets. International Journal of Industrial Organization 19 (5), 613 634. Battisti, G., & Alfonsina, I. (2009). The intra firm diffusion of c omplementary innovations: Evidence from the adoption of management practices by British establishments. Research Policy 38, 1326 1339. Beatty, R. C., Shim, J. P., & Jones, M. C. (2001). Factors influencin g corporate web site adoption: A time based asse ssment. Information and Management 38, 337 354 Becker, L. (2003). Secondary analysis. In G. Stempel, D. Weaver & G. Wilhoit (Eds.), Mass communication research and theory (pp. 252 266). Boston, MA: Allyn & Bacon. Beilock, R., & Dimitrova, D. V. (2003). An exploratory model of inter country internet diffusion. Telecommunications Policy 27 237 252. Berg, S. V. (1989). Technical standards as public goods: Demand incentives for cooperative behavior. Public Finance Review 17(1), 29 54.
138 Berg, S. V. (1991 ). Implications of new technologies and competitive entry for telecommunications policy Research Paper 91 27, Public Utility Research Center, Univer sity of Florida. Boeker, W. & Huo, Y. (1998). Innovation adoption by established firms: Unsolved issues. The Journal of High Technology Management Research 9(1), 115 130. Booz, A llen & Hamilton (1985). Diversification: A survey of European chief executives New York: Booz, Allen & Hamilton, Inc. Borrell, A. J. (1997). Radio station characteristics and th e adoption of satellite delivered radio programming. Journal of Media Economics 10 (1), 17 28. Brinkley, J. (1997). Defining vision: The battle for the future of television New York : Harcourt Brace & Company. Broadcasting and c able y earbook (2010). Was hington, DC: R.R. Bowker Carter, T. B., Franklin, M. A., & Wright, J. B. (2008). The First Amendment and the fifth estate: Regulation of electronic mass media New York: Foundation Press. Chandy, R. K. & Tellis, G. umbency, size, and radical product innovation. Journal of Marketing 63 (3), 1 17. Chen, M. (2005). Technology adoption in response to changes in market conditions: Evidence from the US petroleum refining industry. Economics of Innovation and New Technolog y 14 (8), 735 756. Chipty, T. (2001). Vertical i ntegration, m arket f oreclosure, and c onsumer w elfare in the c able t elevision i ndustry The American Economic Review 91(3), 428 453. Christensen, C. M. (1993). The rigid disk drive industry: a history of com mercial and technological turbulence. Business History Rev iew 67 ( 4 ), 531 588. Christensen, C. M., Surez F. F., & Utterback J. M. (1998). Strategies for s urvival in f ast c hanging i ndustries Management Science 44 ( 12 ) 207 220. Collier, D., & Messick R. E. (1975). Prerequisites versus diffusion: Testing alternative explanations of social security adoption. The American Political Science Review 69 (4), 1299 1315. Compaine, B (2003). Revisiting cost and affordability assumptions for high speed data services in l ow population density l ocations Retrieved from http://web.si.umich.edu/tprc/papers/2003/236/LowDensityBroadband2.pdf
139 Copeland, A., & Shapiro, A. H. (2011 ). The impact of competition on technology adoption: An App les to PCs analysis Retrieved from http://www.copeland.marginalq.com/res_doc/copeland_shapiro.pdf Damanpour, F. (19 91). Organizational i nnovation: A m eta a nalysis of e ffects of d eterminants and m oderators The Academy of Management Journal 34 ( 3 ), 555 590 Damanpour, F. & Schneider, M. (2006). Phases of the adoption of innovations in organizations: Effects of organiz ations, organization, and top managers. British Journal of Management 17, 215 236. Dasgupta, P. (1986). The theory of technological competition. In J. Stigliz & G. Mathewson (Eds.), New development in the analysis of market structure (pp. 519 547). Ca mbr idge, MA: MIT Press. Dasgupta, S., Granger, M. & McGarry, N. (2002) User acceptance of e collaboration technology: An extension of the technology acceptance model. Group Decision and Negotiation 11 87 100. Davies, F. (1989). Perceived usefulness, per ceived ease of use and user acceptance of information technology. Management Information Systems Quarterly 13 (3). 319 3 40. Ducey, R V., & Fratrick M. R. (1989). Broadcasting i ndustry r esponse to n ew. t echnologies Journal of Media Economics 2(2) 67 86 Dupagne, M., & Seel, P. B. (1998). High definition television: A global perspective Ames: Iowa State University Press. Eveland, J. D. (1979). Issues in using the concept of adoption of innovations. Journal of Technology Transfer 4 ( 1). 1 13. Farrell, J ., & Shapiro, C. (1992). Standard s etting in h igh d efinition t elevision Brookings Papers on Economic Activity. Microeconomics 1 93 Farrell, J. & Saloner, G. (1985). Standardization compatibility, and innovation. Rand Journal of Economics 16 70 83. Federal Communications Commission (199 5 ). Advanced t elevision s ystems and t heir i mpact u pon the e xisting t elevision b roadcast s ervice F ourth F urther N otice of P roposed R ulemaking and T hird N otice of I nquiry Washington, DC: FCC F ederal C ommunications C ommission (1997). Advanced t elevision s ystems and t heir i mpact upon the e xisting t elevision b roadcast s ervice Fifth Report and O rder Washington, DC: FCC.
140 Federal Communications Commission (200 1 ). In the m atter of r equests f or f urther e xtension of th e Nov ember 1, 2000, d igital t elevisio n c onstruction d eadline Washington, DC: FCC Federal Communications Commission (2003). DTV build out, requests for extension of digital television construction deadline, commercial television stations with May 1, 2002 dead line Washington, DC: FCC. Fisher, F. M., McGowan, J. J., & Evans, D. S. (1980). The audience revenue relationship for local television stations. The Bell Journal of Economics 11 (2), 694 708. Ford, G. S., & Jackson, J. D. (1997) Horizontal concentratio n and vertical integration in the cable television industry Review of Industrial Organization 12 ( 4 ), 501 5 18. Ford, G. S., Koutsky, T. M., & Spiwak,L.J.(2007). The broadband performance index: A policy relevant method of comparing broadband adoption amo ng countries. Retrieved from http://www.phoenix center.org/pcpp/PCPP29Final.pdf Forman, C., Goldfarb, A., & Greenstein, S. (2003). How did location affect adoption of the commercial Inter net? Global village, urban density, and industry composition NBER Working Paper No. 9979 Retrieved from http://www.nber.org/papers/w9979 Frambach, R. T. (1993). An i ntegrated m odel of o rganizational a do ption and d iffusion of i nnovations European Journal of Marketing 27 ( 5 ) 22 41 Gan dal, N. (2002). Compatibility, s tandardization, and network effects: Some policy implications. Oxford Review of Economic Policy 18 (1), 80 91. Garcia Murillo, M (2005) International b roadband d eployment: The i mpact of u nbundling. Retrieved from http://ssrn.com/abstract=977405 Gatignon, H., & Robertson, T. S. (1989). Technology diffusion: An empirical test of competitive effects. Journal of Marketing 53 35 49. General Accounting Office ( 2002a ). Additional federal efforts could help advance digital television transition Washington, DC: GAO. General Accounting Office (2002 b ). Many broadcasters will not meet May 2002 digital transition deadline Washington, DC: GAO. General Accounting Office (2004 ). German DTV transition differs from US transition, but certain key challenges are similar Washi ngton, DC: GAO. Geroski, P. (2000). Models of technology dif fusion. Researc h Policy 29, 603 625.
141 Gibbs, J. & Kraemer, K. (2004). A c ross c ountry i nvestigation of the d eterminants of s cope of e commerce u se: An i nstitutional a pproach E lectronic M arkets 14(2). 124 137. Goldstein, M. L (2008). Digital television transition: B roadcasters' transition status, low power station issues, and information on consumer awareness of the DTV transition: T estimony before the House Subcommittee on Telecommunications and the Internet Washington, D.C.: Government Accountability Office. Gom ery, D. (1984). Economic change in the US television industry Screen 25(2). 62 67. Goolsbee A. (2007). Vertical i ntegration and the m arket for b roadcast and c able t elevision p rogramming Retrieved from http://fjallfoss.fcc.gov/edocs_public/attachmatch/DA 07 3470A10.pdf Graham, D. P. (2003). Public Interest Regulation in the Digital Age C ommunication Law C onspectus 97, 116 Griliches, Z. (1957). Hybrid corn: An expl oration in the economics of technology change. Econometrica 25 (4), 502 522. Grober, V., Goslar, M. & Segar, A. (1995). Adopters of telecommunications initiatives: A profile of progressive US Corporation. International Journal of Information Management 15 (1), 33 46. Hair, J. F., Anderson, R. E., Black, B., & Tatham, R. L. (2006). Multivariate data analysis Upper Saddle River, NJ: Prentice Hall. Hair, J. F., Celsi, M. W., Money, A. H., Samouel, P., & Page, M. J. (2011). Essential of Business Research Met hods Armonk, NY: M. E. Sharpe. Hall, B. H. & Khan, B. (2003). Adoption of new technology NBER Working Paper No. 9730 Retrieved from http://www.nber.org/papers/w9730 Hanly, D & Cheung K. ( 1998 ) Mar ket structure of multi product firms under free entry. Economics Letters 6 159 63. Hansen, J. A (1992). Innovation, f irm s ize, and f irm a ge Small Business Economics 4, 37 44. Harrigan K. R. (1985). Vertical integration and corporate s trategy The A cademy of Management Journal 28( 2 ), 397 425 Harris Corporation (2011). ATSC mobile digital television Retrieved from http://www.broadcast.harris.com/media/ATSC MobileDigitalTelevision_25 5573.pdf
142 Hart, J. A. (2004). Technology, television, and competition: The politics of digital TV Cambridge, MA : Cambridge University Press. Hart, J. A. (2005). Politics of transition to digital TV. The International Journal of Technology, Knowledge, and Society 1 (4), 123 130. Hart, J. A. (2010). The transition to digital TV transition in the United States: The endgame. International Journal of Digital Television 1(1), 7 29. Hitt, M., Hoskisson, R., & Ireland R. (1994 ). A m id r ange t heory of the i nteractive e ffects of i nternational and p roduct d iversification on i nnovation and p erformance Journal of Management 20(2), 297 326. Hollenstein, H. (2004). Determinants of the adoption of information and communication technolog ies: An empirical analysis based on firm level data for the Swiss business sector. Structural Change and Economic Dynamics 15 315 342. Hoppe, H. (2000). Second mover advantages in the strategic adoption of new technology under uncertainty. International Journal of Industrial Organization 18, 315 338. Hoppe, H. (2002). The timing of new technology adoption: theoretical models and empirical evidence. The Manchester School 70(1), 56 76. Huck, S. (2007). Reading statistics and research Boston, MA: Allyn & Bacon. Kahane, L. H. (2008). Regression basics Thousand Oaks, CA: Sage Publications. K ang, M. (2002). Digital cable: E xploring factors associated with early adoption. Journal of Media Economics 15 193 207. Kast, F. E., & Rosenzweig, J. E. (1985). Organization and management: A systems and contingency approach New York: McGraw Hill. Kim, H. (20 08, May ) A l ongitudinal a nalysis of v ertical i ntegration and s ource d iversi ty in b roadcast TV p rogramming, 1993 2006 Paper presented at the meeting of t he Internatio nal Communications Association Montreal, Canada. King, J. Gurbaxani, V., Kraemer, K., McFarlan, F., Raman, K., & Yap, C. (1994). Institutional factors in information technology innovation. Information Systems Research 5 (2), 139 169. Knitt el, C. R. (2006). The adoption of state electricity regulation: The role of interest groups. The Journal of Industrial Economics 54 (2), 201 222.
143 Kuan K., & Chau, P. (2001). A perception based model of EDI adoption in sm all businesses using technology or ganization environment framework. Information and Management 38, 507 5 52 Levin, S. G., Levin S. L., & Meisel J. B. (1987). Dynamic a nalysis of the a doption of a n ew t echnology: The Case of Optical Scanners The Review of Economics and Statistics 69 ( 1 ), 12 17 Lieberman, M. B. & Montgomery D.B ( 1998 ) First mover (dis)advantages: Retrospective and link with resource based view. Strategic Management Journal 19 ( 12 ) 1111 1125. Lin, C. A. (1998). Exploring personal computer adoption dynamics. Journa l of Broadcasting and Electronic Media 42 ( 1), 95 112. Majumdar, S. & Venkataraman, S. (1993). New technology adoption in US telecommunications: The role of competitive pressures and firm level inducements. Research Policy 22 (5/ 6), 521 536. Ma lone, M. (2009). Digi c hannels e njoy b rave n ew p ost DTV w orld. Broadcasting & Cable 139(25) 16 Mansfield, E. (1961 ). Technical change and the rate of imitation. Econometrica 29(4), 741 766. Metcalfe, S. (1985). The economic foundation of technology policy: Eq uilibrium and evolutionary perspectives, In P. Stoneman (E d.). Handbook of the economics of innovation and technological change (pp. 409 512). O xford, UK: Basil Blackwell. Miles, R. E. & Snow, C. C. (1978). Orga nizational strategy, structure, and process New York: McGraw Hill. Miller J. E., & Prieger J. (2010). The b t ransition d ate r oulette: Strategic a spects of the DTV t ransition Retrieved from http://works.bepress.com/james_priege r/2 Milliou C. & Petrakis E. (2007). Upstream horizontal mergers, vertical contracts, and bargaining. International Journal of Industrial Organization 25 (5), 963 987. Milliou, C. & Petrakis, E. (2010). Timing of technology adoption and product ma rket competition. International Journal of Industrial Organization 29 (5), 513 523. Morgan, L. O., & Daniels, R. L. (2001). Integrating product mix and technology adoption decisions: A portfolio approach for evaluating advanced technologies in the automo bile industry. Journal of Operations Management 19 (2), 219 238.
144 Mortimer J. ( 2008 ). Vertical c ontracts in the v ideo r ental i ndustry Review of Economic Studies 75(1) 165 199 Moyler, A. & Hooper, M. (2009). Over the top TV (OTT) platform technologie s Retrieved March 16, 2011, from http://www.iptv news.com/__data/assets/file/0003/162948/OTT_TV_White_Paper_Rel_1_ver1.pdf Napoli, P. (2001). Marke t conditions and public affairs programming: Implications for digital television policy. Press and Politics 6 (2 ), 15 29. Nelson, R. R. & Winter, S. G. (2002). Evolutionary theorizing in economics. Journal of Economic Perspective 16 (2), 23 46. Neo, B ., Khoo, P., & Ang, S. (1994). The adoption of TradeNet by the trading community: An empirical analysis Retrieved from http://aisel.aisnet.org/cgi/viewcontent.cgi?art icle=1086&context=icis1994 Nooteboom, B. (2000). Learning and innovation in organizations and economies New York: Oxford University Press. Owen, B. (1999). The internet challenge to television Cambridge, MA: Harvard University Press. Papies, D. & C lement, M. (2008). Adoption of new movie distribution services on the Internet. Journal of Media Economics 21 131 157. Pires, G. D., & Aisbett, J. (2002). The relationship between technology adoption and strategy in business to business markets: The cas es of e commerce. Industrial Marketing Management 32 (4) 291 300. Porter, M. E. ( 1980). Competitive strategy New York: Free Press. Porter, M. E. & Millar, V., E. (1985). How i nformation gives you competitive advantage. Harvard Business Review 63(4) 149 160. PRWeb (2011). HD and d igital t ransition s till in i nfancy at m ost U.S. TV s tations Retrieved from http://www.prweb.com/releases/2011/04/prweb5231144.htm Reinganum, J. (1981 ). On the diffusion of new technology: a game theoretic approach. Review of Economic Studies 48 395 405. Riemenschneider, C. K., Harrison D. A., & Mykytyn P. P. (2003). Understanding IT adoption decisions in small business: integrating current theorie s. Information and Management 40, 269 285.
1 45 Roberts S Frenette J & Stearns D (2002) A c omparison of m edia o utlets and o wners for t en s elected m arkets: 1960, 1980, 2000 Washington, DC: Federal Communications Commis sion Rogers, E. M. (2003). Diff usion of innovations New York : Free Press Romano, A. (2008). Local s t ations m ultiply. Broadcasting & Cable 138(10), 16 18 Rose, N. L., & Joskow, P. L. (1990). The diffusion of new technologies: Evidence from the electric utility industry. The Rand Jo urnal of Economics 21 (3 ), 354 373. Saha, A. Love, H & Schwart, R. (1994). Adoption of emerging technologies under output uncertainties. American Journal of Agricultural Economics 76 836 846. Saloner, G., & Shepard, A. (1995). Adoption of technologi es with network effects: An empirical examination of the adoption of automated teller machines. The RAND Journal of Economics 26 (3), 479 501. Sartorius, C., & Zundel, S. (2005). Time Strategie s, Innovation and Environmental Policy Cheltenham, U.K.: Edwa rd Elgar Publishing Schumpeter, J. (1950 ). Capitalism, socialism, and d emocracy New York : Harper. Scupola, A. (2003). The adoption of Internet commerce by SMEs in the South of Italy: An environmental, technological, and organizational perspective. Jou rnal of Global Information and Technology Management 6(1), 52 71 Shapiro, C. & Varian, H. R. (1999). The art of standard wars. California Management Review 41(2), 8 32. Shimizu, K. & Hitt M. (2004) Strategic f lexibility: Organizational p reparedne ss to r everse i neffective s trate gic d ecisions, Academy of Management Executive 18(4), 44 59. Sinkula, J. M. (1994). Market Information Processing and Organizational Learning. Journal of Marketing 58(1), 35 45 Smith, L. F., Wright, J., & Ostroff, D. (1 998). Perspectives on radio and television: Telecommunication in the United States Mahwah, NJ: Lawrence Erlbaum Associates. Snider, J. H. (2005). Speak softly and carry a big stick: How local broadcasters exert political power. Lincoln, NE: iUniverse. S terling, C. H., & Kittross, J. M. (2002). Stay tuned: A history of American broadcasting Mahwah, NJ: Lawrence Erlbaum Associates.
146 Straubhaar, J., LaRose, R., & Davenport, L. (2010). Media now: Understanding media, culture, and technology Boston, MA: Wad sworth. Teece, D. J. (1996). Firm organization, industry structure, and technological innovation. Journal of Economic Behavior and Organization 31, 193 224. Television Bureau of Advertising (2011). TV basics: A report on the growth and scope of televis ion Retrieved from http://www.tvb.org/media/file/TV_Basics.pdf Televis ion and cable f actobook ( 2010 ). Washington, DC: Warren Publishing Thirtle, C. G., & Ruttan, V. W. (1987). The role of dem and and supply in the generation and diffusion of technical change New York: Hatwood Academic Publishers. Thong, J. Y. (1999). An i ntegrated m odel of i nformation s ystems a doption in s mall b usinesses. Journal of Management Information Systems 15, 187 214 Tornatzky, L. G. & Fleischer, M. (1990). Th e process of technological innovation Lexington, MA: Lexington Books. U.S. Census Bureau (2007). County and city data book: 2007 Retrieved from http://www.census.gov/prod/www/abs/ccdb07.html Van Tessel, J. (2001). Digital over broadband: Harvesting bandwidth Boston, MA: Focal Press. Waterman, D. (2000), CBS Viacom and the e ffects of m edia m ergers: An e conomic p erspective Federal Communica tions Law Journal 52(3), 531 550. Wilbur, K. (2008). A two sided empirical mode of television advertising and viewing markets. Marketing Science 27(3), 356 378. Wi nslow, G. (2011). Shopping for e fficiencies. Broadcasting & Cable 141(14), 16 2 0. Wirth, M O., & Bloch, H. (1995). Industrial organization theory and media industry analysis. Journal of Media Economics 8 (2), 15 26. Wirth, M. O., & Wollert, J. A. (1984). The effects of market structure on television news pricing. Journal of Broadcasting 28 (2) 215 224 WRALTechwire (2009). WRAL content now available through iPhone application Retrieved from http://wraltechwire.com/business/tech_wire/news/story/4252222/ Yanich, D. (2010). Does o wnership m atter? Localism, c ontent, and the Federal Communications Commission Journal of Media Economics 23(2), 51 67.
147 BIOGRAPHICAL SKETCH Heejung Kim received a Ph.D. degree in mass communication from the Un iversity of Florida During her Ph.D. s tudies s he was awarded Alumni Graduate Fellowship from the University of Florida A native of Seoul, Korea, Heejung is a graduate of Yonsei University, Seoul, Korea, with a B.A. in English L iterature and an M.A. in mass communication. She also rec eived her M.A. in telecommunications from Indiana University Bloomington under an Associate Instructorship. Heejung was a research associate in broadcast and telecommunications policy at the K orea Information Society Development Institute. Her research an d teaching fields focus on economics of communication s industries, communications policy, and business strategies of communications firms.