SELECTED IMPLICATIONS OF ALGORITHMIC SELECTION FOR TRADITIONAL MEDIA MARKETS
The economic implications of algorithmic selection services are as wide as their fields of application in various sectors of the economy. This section focuses on media markets only, in particular on media incumbents’ profitability.
For decades traditional news companies have dominated the construction of public spheres. They were unchallenged and made high profits in advertising markets. Now both core businesses of news companies - the audience and advertising markets - are increasingly coming under pressure from activities of IT companies like Microsoft or dot-coms like Google and Yahoo!. As market makers, they squeeze themselves between traditional news companies and their two customer segments, the audience and the advertisers (Aguila-Obra et al., 2007). Their competitive advantages result from the generation of huge amounts of data and the automated algorithmic selection and placement of news, on the one hand, and from the automated selection and placement of advertisements on the other. News aggregators (e.g., Google News or Bing News) and online advertising networks (e.g., Google AdSense) are examples of such intermediaries.Thus far, the majority of research on the impact of algorithmic selection on media industries predominantly focuses on news aggregators and online advertising, revealing that they increase both the reading consumptions and the quality of news (e.g., Athey and Mobius, 2012; Chiou and Tucker, 2013; Dellarocas et al., 2013) and have impacts on price strategies and targeting methods in online and offline advertising markets (Edelmann et al., 2005; Evans 2009; Bergmann and Bonatti, 2011). Many other questions remain unanswered and call for further research, especially those regarding the combined economic impact of various algorithmic selection applications that affect both the audience and the advertising market.
Moreover, the impact of algorithmic selection on other media industries such as music or film has not yet been examined.A basis for such analysis is Porter’s (1979, 2008) concept of five forces that shape industry competition, which has been applied, for example, by Maass et al. (2009) to assess the robustness of concentration in search markets. Coupled with comparative analyses of business models and market structures, Porter’s approach also makes it possible to assess the impact of algorithmic selection applications such as news aggregators, algorithmic content production, computational advertising, music streaming or subscription video-on-demand services on the news, music, film and TV industries. Changes to the five competitive forces - the threat of new entrants and of substitute products and services, the bargaining power of both suppliers and buyers, and the rivalry among existing competitors - affect the average profitability of media incumbents.
Theoretical considerations suggest that algorithmic selection services predominantly come into effect as intermediaries or suppliers in media industries, and tend to change their profitability. In his early assessment of the Internet in general, Porter (2001) argues that the Internet tends to decrease profitability. Theoretical analyses of algorithmic selection applications indicate that the impact on incumbents’ profitability seems to vary from media industry to media industry (Latzer et al., 2014). For example, in the news industry, algorithmic selection tends to decrease average profitability overall. Although incumbents benefit from added traffic streams (Chiou and Tucker, 2013; Dellarocas et al., 2013) and from integrating algorithmic selection (e.g., news created by algorithms) as an ancillary function, intermediaries such as news aggregators or advertising networks change the competitive forces of the industry (Porter, 1979) to the disadvantage of incumbents. Increasingly high concentration in these markets is shifting the bargaining power to these intermediary platforms and allowing them to amplify their market power.
This is especially the case if they are able to establish themselves as bottleneck monopolists that control the access to products of others (Shelanski, 2013), as in the case of news publishers’ content or as evident in the struggles between book publishers or sellers, and lately Disney and Amazon. Some have turned to opt-out options, for example by blocking their sites from search engines (e.g., News Corp. blocked Google services by using robot.txt files). Opting out of search services has not been a feasible solution for publishers, however. Search engines, for example, are responsible for high visitor rates to news websites, with widely differing figures up to 35 percent (SimilarWeb.com, 2013). Further, aggregation of news in single access points also results in lower transaction and switching costs for news customers and tends to increase their bargaining power. As new entrants, online advertising networks in particular are straining incumbents’ profitability.For the music industry, in contrast, it can be argued that the adoption of algorithmic selection as key part of their services tends to increase incumbents’ profitability overall, as music-streaming services (e.g., Last.FM, Spotify) strongly stimulate (legal) music consumption and have been revenue drivers in recent years. Although they also established themselves as intermediaries they are faced with a highly concentrated music industry with great bargaining power.
The differences in impact on various media industries can be explained by different business models of algorithmic selection services (market makers, layer players) and by the different stages of market development (market phases) of the relevant algorithmic selection services as well as business models and market structures of traditional industries (Latzer et al., 2014).
These first rough estimates and theoretical considerations of the possible impact of algorithmic selection on media industries are still in need of further research, and in particular need to be combined and weighed with current market data in order to receive an accurate picture of the full economic implications for media industries.
19.7