Do ‘African American’ films perform better or worse at the box office? An empirical analysis of motion picture revenues and profits

Abstract

This article investigates the box office performance of films defined as being ‘African American’, with respect to their cast and content material, against those which are not. Using a large sample of films released in the North American market from 1997 to 2007, the analysis shows that, in general, African American films earn higher revenues yet are typically produced on lower budgets. Regression results of revenues show that this difference is highly statistically significant. Further, the profit functions are also statistically different, leading to the conclusion that, ceteris paribus, African American films perform better at the box office.

Notes

¹See, for example, Smith and Smith (1986), Prag and Casavant (1994), De Vany and Walls (1999), Eliashberg and Shugan (1997), Albert (1998), Ravid (1999), Hand (2001), Collins et al. (2002), Basuroy et al. (2003), Elberse and Eliashberg (2003), Reinstein and Snyder (2005), Goettler and Leslie (2005), Jansen (2005), Walls (2005a, b) Ravid et al. (2006), Gemser et al. (2007), Einav (2007) and Moul (2007). A useful survey is provided by Eliashberg et al. (2006).

²The top 10 African American films ranked in terms of revenue were Bad Boys 2, Bringing Down the House, Big Momma's House, Remember the Titans, Save the Last Dance, Dreamgirls, Norbit, Are We There Yet?, Barbershop and Shaft. A full list of the titles used in this study is available from the author on request.

³See De Vany and Walls (1996, 1999, 2004), Walls (1997, 2005a, b), Hand (2001) and McKenzie (2008).

⁴Obviously the definition of profit being used here is incomplete as it does not account for income from other streams (e.g. international theatrical, DVDs, cable).

⁵Although the dependent variable is an aggregation over many weeks, the inclusion of calendar week dummy variables captures the fact that films typically earn a large portion of their revenue in their first week, which may impact strongly on cumulative revenue. Einav (2007) examines the underlying seasonality of demand in detail.

⁶Ulmer's lists have been published since 1997 and to date has six volumes. Volume 1 was used for 1997–1999 films, Volume 3 for 2000–2002 films, Volume 5 for 2003–2005 films and Volume 6 for 2006–2007 films.

⁷Fifty-three weeks are used due to leap years and the fact that there are 365 days in a year. This ensured Boxing Day, the biggest single day of the year, always fell in week 53.

⁸Walls (2005a) extends work of De Vany and Walls (2004) to estimate a conditional stable regression of motion picture profit.