Characteristics of buyers and renters of cultural goods: the case of movies

Abstract

This article identifies the characteristics of consumers of movies, watched on videotapes, in their homes. Models of the frequency of movie video buying and renting by individuals are estimated using data from the Spanish Cultural Consumption Habits Survey (Encuesta sobre Hábitos de Consumo Cultural). This survey featured information about videotapes rented and bought in Spain throughout 1998. Applying a multinomial probit modelling framework, this study analyses whether there are two different video markets (renting and buying) and any discernible socio-economic characteristics for their respective consumers. The influence of film genres on the renting of movie videotapes is also considered.

Acknowledgements

This research has been financed by the Fundación Autor and Sociedad General de Autores y Editores (SGAE). The authors are grateful for the comments of Samuel Cameron and Chris Hand. The usual caveat applies.

Notes

¹ However, there are some studies modelling cinema consumers’ behaviour. See, for instance, Cameron (1999), Walls (2005a).

² Since DVDs were not yet popular by 1998 the data deals virtually exclusively with videotapes.

³ Fernández-Blanco and Baños-Pino (1997) analyses the effect of TV specifically in the context of the Spanish market. Macmillan and Smith (2001) do the same with respect to the British market, which is examined more generally by Hand (2002). Dewenter and Westerman (2005) study the German case.

⁴ Meanwhile, in the USA during the 1980s, TV's contribution to total Hollywood majors revenues decreased from 19% in 1980 to 15% in 1989, but the development of the pay-per-view system allowed the majors to be more optimistic. Augros (1996) forecast that their revenues in this window would steadily increase. That work suggested an increase from 12 million dollars in 1987 to 807 million dollars in 2001.

⁵ Studies on movies box office and profits are more common. See, for instance, Smith and Smith (1986), De Vany and Walls (1999), Hand (2001) or Walls (2005b). De Vany (2004) and Moul (2005) offer good surveys on movies revenues, profits and success.

⁶ The EHCC was also conducted quarterly during 1997 but during this year, the survey was in a developmental phase and the interview questionnaire content was significantly different, such that it would not support the type of work we report in this article. Accordingly, it was decided to use only the 1998 data.

⁷ This fact can be easily understood when we consider that, in 1998, 73.4% of Spanish homes had a video set. On the other hand, theatre ticket prices were increasing, while renting and buying video prices were both decreasing, probably because of the presence of strong new competitors such as piracy and pay-per-view TV systems.

⁸ It is important to note that the EHCC survey does not ask if the interviewee rented videotape, only if he or she watched a rented videotape. Furthermore, when the survey asks about video buying frequency it does not specify video type: movie, documentary, music etc – although movies represent 90% of the whole market.

⁹ Reported marginal effects account for the variation in the probability of the dependent variable due to an infinitesimal change in each independent continuous variable or the discrete change for dummy variables shifting from zero to one.

¹⁰ If we consider the positive and statistically significant effect of the variable INTVTHEATER, that measures the interviewee's interest on theatre programmes on TV, we can conclude that film buyers are also people with some significant measure of cultural consumption-based motivation.

¹¹ Walls (1998), Bagella and Bechetti (1999), De Vany and Walls (1999) and Fernández-Blanco and Prieto-Rodríguez (2003) have analysed the effect of genres on movie theatre attendance.

¹² This question was only available in the interview questionnaire used in the first quarter of 1998. Hence, our sample size is reduced to 3300 individuals.

¹³ The genres (and variables) included are the following: action (VACTION), mystery (VMYSTERY), adventure (VADVENTURE), comedy (VCOMEDY), drama (VDRAMA), romantic (VROMANCE), message (VMESSAGE), children's (VCHILD). Detailed definitions of all these variables are in the appendix.

¹⁴ The log likelihood ratios endorse the goodness of fit of these model estimates.