Effects From Privatizing A Television Market, the Influence of Mobile Advertising on Movie Box Office, and Causal Relationships Between Word of Mouth and Movie Ticket Sales

The first article in this issue finds privatization of the Croatian television market led to decreasing concentration with a foreign-owned channel replacing government television as the dominant broadcaster in the market. The second article uses data on mobile Location-Based Advertising in China to show these ads can substantially increase movie ticket sales for up to nine days after a consumer receives the ad. The third article uses causal analysis of the relationship between Word of Mouth and movie box office in Taiwan, with findings that show when marketers should encourage positive online comments and discourage negative online comments.

The first article is “Media Control: A Case for Privatization in Transitional Economies” by Fran Galetić, Marina Dabić and Timothy Kiessling. This study examines what happened when the government controlled television market in Croatia was liberalized to allow privately owned broadcasting. The market in Croatia is similar to markets in other recent members of the European Union, therefore this study offers insights about privatization for similar EU countries and candidates to join the EU.

Concentrated media ownership can threaten diversity of expression and create autocratic control of mass communication. Television strongly influences audiences, so it’s important that television programs reflect diverse opinions and ideas. Therefore, liberalization of government controlled television markets “is a very important step in each country that is moving to a democratically controlled government” (p. 113). The Croatian market was monopolized by three government controlled television channels until 2000. Privatization began in 2000, resulting in 11 national channels, both public and private, by 2014.

Concentration in the Croatian television market was measured as the share of all viewers that each national channel had, with the total for all channels equaling 100 percent. This method did not include viewers watching local or specialized channels. New channels entered the market throughout the study. The Herfindahl-Hirschman index in the Croatian television market diminished throughout the study, indicating decreases in concentration. By 2014, the HHI was below 1800, meaning the market was moderately concentrated.

The study also used regression to conduct a trend analysis that predicted the market share for each national television channel for three years after the data ended. The regressions tested four different functional forms for each television channel, and used the models with the highest R-squared. The projections found that the television channel with the largest market share is expected to enjoy continued growth at the expense of its three closest rivals. As a result, the HHI is also expected to stop decreasing and begin increasing instead. “Generally speaking, Croatian public TV channels will continue losing their audience, while the majority of other private TV channels will gain new audience” (p. 119).

The authors conclude the Croatian television market has transitioned from monopoly to oligopoly because market concentration remains high even though the number of channels is not small. However, government-controlled channels are expected to continue losing market share as private channels gain larger audiences. Some channels, including the leading channel, are now owed by foreign interests. The authors argue that foreign “firms offer a global perspective and are not influenced by the government and offer programming that will deliver an eclectic view” (p. 122). This case-study of the Croatian market suggests that deregulated markets in other nations will transition from monopoly to oligopoly, “with the former [government] monopoly suffering the greatest loss of market share” (p. 126).

The second article is “Boost Movie Ticket Sales by Location-Based Advertising: A Bayesian VAR Approach” by Zheng Fang, Yang Yang, Yanyan Xu and Wei Li. This study examines the effectiveness of movie advertising delivered to potential ticket buyers on their mobile phones. Location-based Advertising (LBA) targets potential customers based on their geographic location, maximizing mobile technology’s potential to target personalized ads in real time. However, there is not much empirical evidence about how this form of advertising influences ticket sales. This study provides such evidence.

This study examined the short and long term effects of LBA on movie ticket sales in China. The study used data from one of the largest mobile companies in the world. The company’s LBA technology allows consumers to use their mobile phones to inquire about movies and prices, and to book tickets and select seats. Consumers who are a given distance from a theater, usually less than 200 meters, receive LBA ads as text messages on their phones.

The data in the study was for more than 3 million consumers who received LBA from August 1, 2009, to July 31, 2010. Consumers who belonged to a mobile movie fan club were designated high involvement, others were designated low involvement. Consumers who had purchased movie tickets in the previous three months in response to LBA were randomly selected by the mobile company to receive behavior-based mobile ads. Another group was consumers who installed a client on their personal computers that the company used to deliver pop-up ads when they went online. The study also included data on calls from consumers to get information about movies, and controlled for weekends, high and low volume seasons such as summer vacation, and movies with blockbuster ticket sales.

The data was analyzed using Vector Auto-Regressive time-series models. Results found that an increase of 100,000 LBAs generated about 316 additional ticket sales in the short-term. The increase in LBA also generated 753 additional ticket sales in the long-term, defined as nine days after the LBA was received. The LBA had much larger effects on the high-involvement segment of consumers. The decay time for an LBA persisted for nine days among low involvement consumers, but lasted only one day for high involvement consumers. This difference is important because high-involvement consumers are less than 1% of all consumers, but account for about 20% of ticket sales.

Results also found that mobile advertisements delivered to consumers who previously purchased tickets had the largest influence on sales, followed by LBA, and then the pop-up ads delivered to personal computers.

The authors conclude that the average influence of an LBA persists for about nine days with a cumulative influence on ticket sales that is more than twice as large as the immediate, short-term influence. These cumulative effects are primarily among low-involvement consumers. Therefore, marketers should track long term effects from LBAs and not focus only on the short-term influence of these ads. The study also provides a quantifiable measure for determining if LBA is effective. Marketers should model and track the influence of LBAs among different segments of the movie audience. “The vast majority of audiences, especially low-involvement audiences, are not engaged in impulsive purchases. Audience segmentation allows marketers to more objectively evaluate the ticket sales impact of LBA” (p. 134) the authors conclude.

The third article is “Bidirectional Causality for Word of Mouth and the Movie Box Office: An Empirical Investigation of Panel Data,” by Yuan-Lin Hsu and Wen-Jhan Jane. This study examines how Word of Mouth (WOM), defined as online person-to-person communications, influence movie ticket sales. Movies are an experience good, and WOM provides a signal to consumers about the quality of a movie that they have not watched. However, most previous studies of the relationship between WOM and movie ticket sales did not examine the endogenous relationship between WOM and sales. This study argues that positive “WOM leads to more box office sales, which in turn generates more WOM and then more sales” (p. 140). The study examines the dynamic relationship between WOM and movie ticket sales using a panel Granger causality test.

The study uses data from Taiwan. The study examined ticket sales for 769 movies during a period of 29 months. The WOM data is from more than 159,000 messages posted online and gathered by a computer program. Messages were indexed by the date of posting to identify message posted before a movie was released. A subset of postings was manually coded for positive and negative terms, then a computer program used the subset to automatically classify other postings. The panel Granger causality test used weekly box office for each movie and a score based on the number of positive or negative words in a comment, the number of comments, and the ratio of positive to negative comments about a movie.

Results found a “one-way causality running from the box office to reviews is confirmed” (p??). Robustness checks using WOM in weeks 7 to 12 of a movie’s release showed ticket sales predict the volume of WOM. However, causality is bi-directional in the long-run, or after 10 weeks.

The authors conclude that “causality runs only from the box office to the number of reviews in the short run, and the causality between WOM volume and box office is bi-directional in the long run”(p. 145). They also conclude that causality “runs from the negative critics to the box office in the long run when the positive critics is (sic) controlled, and the causality runs from the positive critics to the box office in the short run when the negative critics is (sic) controlled” (p. 145). The findings support previous research that WOM influences consumer purchases. The finding also show that causality for positive and negative WOM and ticket sales is asymmetric. To stimulate ticket sales, marketers should encourage positive WOM in the short run and reduce the effect of negative WOM in the long-run. This will help create “a positive feedback loop” where increasing WOM leads to increasing ticket sales, which then increases WOM again.

We hope you enjoy these articles as much as we have.