Star Power: The Movie Has to Have It, Not the Actors

Stars. “Essentially worthless . . . absolutely essential.”

That’s how screenwriter William Goldman assessed the importance of stellar screen presences 15 years ago in his well-known book on the movie business, “Adventures in the Screen Trade.” Only one star matters, he wrote, and that star is the picture.

Now two economic professors applying advanced “probability mass distribution models” and “stochastic dynamic processes analysis” to more than 2,000 Hollywood movies released over a 12-year period have come to essentially the same conclusion in a 33-page scientific paper (56 pages if the attached charts and graphs are counted), which will be presented today at the American Economic Assn.’s annual meeting in New York.

Arthur De Vany, a UC Irvine economics professor and lifelong movie lover, acknowledges that the research amounts to a ponderous explanation of why Goldman’s famous dictum about the movie business--”Nobody knows anything”--is essentially correct. He insists, though, that his work with W. David Walls of the University of Hong Kong debunks some common myths about stars and has valuable lessons for film executives who want to keep their studios afloat.

After all of the economists’ complex calculations and obscure discussion of DF-betas and asymptotic Pareto rank equations, they do reach some surprising conclusions, among them that the two current stars who most affect movie box office are women (Michelle Pfeiffer and Sandra Bullock).

They also found that, despite conventional wisdom, stars are more important in giving a movie “legs” to stay on screens for the long haul than they are in helping a movie open big. Longevity is more important to a movie’s profitability than a huge opening weekend, the findings show.

The presence of a star will get a movie booked in more theaters initially, but “stars give more kick to screen counts later than at the opening of a movie’s run,” the professors write.

“By its fifth week, a movie with a star will have nearly twice as many theaters as a movie without a star. And by the 10th week, nearly three times as many.”

Even so, the researchers say it is impossible to determine whether it is the presence of stars that gives the movie a boost or if their findings merely show that top stars have access to the best stories and select material more likely to connect with audiences.

In other words, nobody knows anything.

“Forecasting the success of a movie is like predicting an earthquake,” De Vany said in an interview.

Of the top 100 stars (they include producers and directors in their calculations), only 19 have a detectable effect (5% or more) on the probability that a movie will be a hit. Based on data collected from 1984 to 1996, those 19 are, in order after Pfeiffer and Bullock:

3. Tom Cruise

4. Jim Carrey

5. Jodie Foster

6. Brad Pitt

7. Steven Spielberg

8. Oliver Stone

9. John Travolta and Kevin Costner

11. Tom Hanks

12. Cher

13. Harrison Ford

14. Michael Douglas

15. Robin Williams

16. Clint Eastwood and Mel Gibson

18. Eddie Murphy

19. Arnold Schwarzenegger

There is a surprisingly large body of academic research on the ability to forecast movie box office. De Vany said, however, that he and Wall are the first researchers to reject traditional ways of looking at it.

“They [other researchers] were all thinking in terms of bell curves with a hump in the middle and with most [of the data] clustered around it,” he said. “Well, there is a hump, but it’s like a ski slope.” A handful of box-office blockbusters are at the top, “then it goes down like Bob Hope’s nose,” he said.

A studio executive heeding the study’s findings would stop opening movies so frequently on a huge number of screens, which he called a “knife-edge strategy” dependent on high early grosses. The danger of this approach is that it dilutes earnings at each theater and can result in exhibitors abandoning a movie early.

According to his research, the single most important factor in determining whether a movie will be a hit is a long run. A wide release is the second factor, followed by the presence of a star. “Budget is the least important [factor], once these other factors are controlled for,” the report says.

De Vany also said it would be wise for studios to drop the current practice of green-lighting individual movies and switch to a strategy of choosing portfolios of movies. That way, the success of one movie could benefit investors whose funds are distributed among films in the diversified portfolio.

And he said he would advise studios to continue to use stars but not with the primary goal of having huge opening-weekend profits. Most important he said is “a wonderful story” that will engage audiences and have lasting power.

Studios also should stop trying to forecast a movie’s success by using current methods and switch to scientific probability analysis, he said.

“By making strategic choices in booking screens, budgeting and stars, a studio can position a movie to improve its chances of success,” the professors write. “But, after [a] movie opens, the audience decides its fate. . . .

“Weeks after the opening, after the effects of marketing, opening power, special effects and big budgets are gone, what remains is the most important determinant of a movie’s success of all--that the audience loves it,” the report says.

Goldman could hardly have put it better.