Muldoon 1 An Examination of Creativity in Motion Picture Molly Muldoon Advisor: Professor Robert Baumann Department of Economics College of the Holy Cross Fall 2012 Abstract: Over the past 25 years, the movie industry has expanded and the number of movies released each year has roughly tripled. At the same time, average movie revenues have remained fairly constant, and during several years, revenues even declined. How can one explain the steady increase in the number of movies released while revenues were constant and even decreasing? I hypothesize that there must be more movies being released based on previously released material (less originality), or perhaps the quality of the films being released has been decreasing. The purpose of this study is to provide some potential explanations for the drastic expansion of the movie industry.
An Examination of Creativity in Motion Picture
Mar 15, 2023
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Microsoft Word - Molly Muldoon Thesis.docxMolly Muldoon Advisor: Professor Robert Baumann
Department of Economics College of the Holy Cross
Fall 2012
Abstract: Over the past 25 years, the movie industry has expanded and the number of movies released each year has roughly tripled. At the same time, average movie revenues have remained fairly constant, and during several years, revenues even declined. How can one explain the steady increase in the number of movies released while revenues were constant and even decreasing? I hypothesize that there must be more movies being released based on previously released material (less originality), or perhaps the quality of the films being released has been decreasing. The purpose of this study is to provide some potential explanations for the drastic expansion of the movie industry.
Muldoon 2
I. Introduction
Movie critics despair over the lack of creativity in Hollywood. Last year there
were twenty-seven movie sequels released which is higher than in any previous year to
date (Ebert 2011). The top seven box office films of 2011 in the U.S. were all movie
sequels, which is unprecedented in box office history (Cieply 2011). In addition, several
other movies released were based on major books. The past year in film is part of a
continuing trend. In the past fifteen years, the strong performance and box office
presence of movie sequels and other films based on previous material have been on the
rise.
The goal of this study is examine the creativity of the movie industry. Creativity
is the “divergent thinking to conceive new ideas” (Heunks 1998), and in this study,
creativity will be measured by the originality of film. A film will be considered a
creative work if it is not based on any previously released material such as a book,
television show, comic book, or remake. A film will not be considered creative in this
paper if it is based on any type of previously released material.
Preliminary work has shown that over the past 25 years, the number of movies
released per year has roughly tripled, and the average revenues per movie have remained
fairly constant1. Between the years 2001 and 2006, the number of movies released
increased by about 41% while average revenues fell by about 33% (See Figures 1 and 2).
How does one reconcile this rapid expansion of the movie industry while revenues were
simultaneously declining? After looking at the number of employed screenwriters
(defined as members of the Writers Guild of America) over these 25 years, one can see
1 Preliminary work data are taken from the Box Office Mojo and Writer’s Guild of America West websites.
Muldoon 3
that the number of screenwriters remains fairly constant and increases by about 6%
between 2001 and 2006 (See Figure 3). This small increase in the number of employed
screenwriters does not begin to explain the drastic expansion of the movie industry.
Therefore, there must be other factors driving this expansion despite falling movie
revenues.
Data in this study confirm that over the past 15 years, there has been a drastic
decline in Hollywood’s creativity as illustrated in Table 1. In 1996, roughly 58.45% of
films released were original, compared to 53.94% in 2006, and 51.36% in 2011. While it
is clear that over the past two decades there has been a sizeable decrease in the creativity
of films released, it is less apparent why the decrease occurred. The goal of this study is
to provide some potential explanations for the decline in creativity.
There are many factors that may influence the creativity of films. Perhaps movie
producers rely more heavily on previous success in times of financial hardship. During a
recession, movie producers may prefer the lower risk of previously established material.
Previous work finds that sequels on average earn more box office revenue than other
films (Terry Butler and De’Armond, 2005 and Walls, 2008). Another factor may be
competition between movie studios. If one studio produces a movie that captures high
box office revenues and that causes a sensation, other studios may attempt to replicate the
success of that studio. Other factors that may influence the originality of a film are its
budget, the influence of critic reviews, and characteristics of the film such as the rating
and the genre. Also, the size of the studio, or more specifically the number of
screenwriters employed may influence the level of creativity. In years where many
screenwriters are producing screenplays, there may be an increase in innovative ideas,
Muldoon 4
and during years with fewer screenwriters one might expect that creativity of films would
decrease. On the other hand, perhaps the movie industry is truly unpredictable and there
is simply no recipe for innovation. Regarding the movie business, screenwriter William
Goldman notes, “With all due respect, nobody knows anything” (De Vany and Walls
1999).
While it is clear that less original movies are being released than fifteen years ago,
it is less obvious whether the decline in originality has impacted the quality of films. In
this study, critic and audience reviews are used as proxies for the quality of a movie. I
assume a high proportion of favorable reviews suggests a higher film quality, while a low
proportion of favorable reviews suggests a lower film quality. I find that the perceived
quality of a film is only slightly influenced by the originality of the film. Interestingly,
both critic reviews and audience reviews were slightly higher for movies that were based
on previous material. Favorable critic reviews averaged 48.95% and 49.17% for original
movies and unoriginal movies respectively. Favorable audience reviews averaged
59.79% and 60.20% for original movies and unoriginal movies respectively.
I find that unoriginal movies earn approximately 17.4% more at the box office
than original movies, holding other observable characteristics constant. More
specifically, the effect of a sequel and remake on gross revenue is about $6.97 million
and $8.89 million, respectively, for a movie with average revenues. I also find that there
is not a decline in movie quality between 1996 and 2011 as measured by critic and
audience reviews. Instead, movie quality actually increases as movies in 2011 received
statistically significant higher reviews than movies released in 2006 and 2001. The
Muldoon 5
results of this study suggest that the decline in movie originality must be driven by a
profit incentive and not by decreasing quality in film.
II. Literature Review
The existing literature does not examine the creative component of the movie
industry. Instead, the movie industry literature examines the variables that influence a
movie’s revenue and profits, and the risk and the lack of predictability that is
characteristic of the movie industry. In addition, the literature finds that movie sequels,
along with several other factors, are a vital element to the movie industry. Terry, Butler
and De’Armond (2005) examines several determinants of box office performance. These
variables include critic reviews, award nominations, sequels, budget, release exposure
and Motion Picture Association of America (MPAA) rating. They define the sequel
variable as “a movie derived from previous material” (145). This study finds that the
number of theaters showing a movie, the critic reviews, the movie budget, and the film
being a movie sequel all have positive and statistically significant effects on box office
revenue. They find that sequels earn about $36 million more at the box office than
movies that are not sequels. They also hypothesize that the main reason sequels are
produced is because there is already an assumed audience for a sequel to a successful
film.
In a similar study, Walls (2008) looks at movie profits as a function of the film
budget, the number of screens the film is released on, whether or not the film is a sequel,
whether or not the film has a major star, the genre, the rating, and the year produced.
While being a sequel does not guarantee a movie success, Walls finds that sequels have
Muldoon 6
higher profits than non-sequels by approximately $880,000. Through his non-parametric
analysis, he finds that sequels have positive expected profitability and mean profitability.
Assuming that movie producers strive to maximize box office revenues in order to
maximize profits, the articles by Terry, Butler and De’Armond (2005) and Walls (2008)
suggest that producing more movies that are based on previous material (non-creative
movies) is a safer way to ensure steady revenues.
In a similar approach, Ravid (2003) examines the risk and the lack of
predictability of the movie business and how movie producers attempt to guarantee
revenues. He examines several “puzzles” of the movie industry, one of these being
“deviation from profit or value maximization.” He suggests that this deviation can be
explained as an attempt to reduce risk by risk-averse studio executives. He also notes that
higher ratings correspond with higher revenues. As a result, producers strive to reproduce
the formula used for successful films, which explains why movie sequels and imitative
films tend to receive higher film ratings. According to Ravid, producing sequels may be
viewed as “corporate hedging behavior” (40).
Economic theory suggests that consumers enjoy and prefer variety. However, this
is not clear when looking at the increasing numbers of unoriginal movies and the revenue
earned by each. Previous literature suggests that movies based on previous material have
higher profits than original films. The literature also suggests that other factors are
important to a film’s success such as the naming strategy, advertising, the number of
theaters showing a film, and the movie budget. Because the movie business is risky,
evidence indicates that producers try to minimize risk by reproducing films similar to
those which have been successful in the past.
Muldoon 7
III. Data Sources
I use a fifteen-year time span between 1996 and 2011. Because a large amount of
the data was collected by hand, I use data from 1996, 2001, 2006, and 2011 rather than
each year. This produces a total of 878 movies.
The existing literature highlights several variables that influence movie revenue,
including the rating and genre of the film, the release date of the film, the budget, the
critic reviews, the number of theaters and a dummy variable for whether or not the film is
an original work. In addition to the variables suggested by the literature, this study
includes ratings provided by the audience and critics, dummy variables for the year the
film is released, and the studio that releases the film. Most importantly, this study
includes variables to capture the originality of the film. The first variable that captures
the originality of the film is a dummy variable “original.” Original movies are films that
are not based on any previous material. The second method that captures the originality
(or unoriginality) of the film specifies dummy variables for each type of unoriginal
movie. Dummy variables are created for sequels, remakes, movies based on books,
movies based on plays, movies based on television shows, and movies based comic
books. These variables determine whether the effect of an unoriginal film depends on the
type of source material. In other words, a sequel and a movie based on a book may have
a different or more significant effect on revenues or reviews. Information regarding the
originality of the film is obtained from the Rotten Tomatoes, IMDB and Wikipedia
websites.
Muldoon 8
Similar to Terry, Butler and De’Armond (2005), I use data from Box Office Mojo
(boxofficemojo.com), the Rotten Tomatoes website (rottentomatoes.com), the-
numbers.com, and Movies.com. Data on the opening and eventual total number of
theaters a movie was released on, the opening and total revenue of a movie, the date of
the movie’s release, and the movie’s production studio are from Box Office Mojo. Data
on MPAA ratings, genre, critic reviews, and audience reviews are from Rotten Tomatoes.
The critic and audience reviews appear as percentages of favorable reviews and serve as
proxies in this study for the quality of a film. Since data on movie budgets is not
available for each movie, budget data are collected from Box Office Mojo and the-
numbers.com. If budget data are available for a given movie from both sources, the two
budgets are averaged
I create several other dummy variables. First, I generate dummy variables for the
quarter and year of the movie’s release. Dummy variables are also generated for the six
most common movie genres, the rating of the movie, and the top five producing studios.
IV. Results
There are six different regression models in this study. Because one of my
hypotheses about the decline in movie originality rests on the assumption that movie
studios are profit driven and that unoriginal movies earn higher box office revenues, the
first two regression models examine the determinants of box office performance. These
two models are modified versions of the model used by Terry, Butler and De’Armond
(2005) which utilizes a log linear model of revenue as a function of a movie’s critic
reviews, release date, budget, number of award nominations, and whether or not the
Muldoon 9
movie is rated R, rated G, a sequel, an action film, and whether or not the movie was
released within seven days of a major holiday.
The first two models in this study are log-linear ordinary least squares regressions
(see Table 3 for regression results). Both models have the natural log of revenue as the
independent variable, and the dependent variables capture the impact of critic ratings,
audience ratings, MPAA rating, genre, producing studio, originality, budget, and the
quarter and year in which the movie was released. These two regression models are the
same except that the first model includes a single variable to capture whether or not the
movie is original, while the second model includes more specific variables to capture the
type of source material. These variables allow for varying revenue effects of the source
material. For example, a sequel may boost a movie’s revenue more than being based on a
play, etc.
The results of these two models suggest that unoriginal movies earn
approximately 17.4% more at the box office than original movies, which is a statistically
significant estimate at 5%. When broken down into more specific categories, the results
suggest that a movie based on a book, television show, or play increases revenue,
although these figures are not statistically significant. Movies based on comic books
earned lower box office revenues than original movies, although this finding is not
statistically significant. Remakes and sequels both earn higher box office revenues than
original films. For a movie that is a sequel, the average gross revenue is about $64.6
million. Given a coefficient of 0.291, $18.8 million of the revenue is attributable to the
movie being a sequel. For a movie that is a remake, the average gross revenue is
Muldoon 10
approximately $33.1 million. Given a coefficient of 0.371, $12.3 million of the revenue
is attributable to the movie being a sequel.
The results of these two models also suggest that while critic ratings do not
significantly impact a movie’s revenue, audience ratings are an important factor; a one
percentage point increase in the favorable audience reviews lead to a 2.6% increase in
box office revenue. The results also indicate that rated movies earn statistically
significant higher box office revenues than the omitted category of unrated movies,
perhaps because rated movies have wider releases than unrated movies in general. In this
study, the top six most common genres (drama, comedy, action, foreign, horror and
family) are included in the regression2. The results suggest that dramas with average
revenues earn about $11.2 million less than movies of omitted genres, comedies with
average revenues earn about $8.29 more than movies of omitted genres, and horror films
with average revenues earn about $19 million more than movies of omitted genres.
Finally, movies produced by four of the top five studios earn statistically significantly
higher revenues than other studios.
Because my other hypothesis regarding the decline in movie originality over the
last 15 years rests on a change in the perceived quality of movies, the third and fourth
models in this study examine the impact of originality on audience reviews as measured
by the percentage of favorable reviews (see Table 4 for regression results). Both models
are linear ordinary least squares regressions. While the results suggest that originality in
general does not impact audience reviews, movie sequels in particular receive about 4.9
2 Many genre categories are omitted because there were too few movies in these categories to be included in the regression. Some of the omitted genre categories include adventure, animation, concert, crime, documentary, fantasy, musical, romance, science fiction, thriller, and western.
Muldoon 11
percentage point higher favorable audience reviews than unoriginal movies, which is
statistically significant at 5%. The results also indicate that a movie’s budget and ratings
do not prove to be statistically significant predictors of the audience reviews. Dramas
and foreign films earn about 5.78 and 20 percentage point higher favorable audience
reviews, respectively, than movies of omitted genres. While the producing studio does
not seem to significantly impact audience reviews, movies produced by Sony receive less
favorable audience reviews than movies produced by omitted studios by about 4.6
percentage points. Movies released in the first three quarters of the year receive
statistically lower percentages of favorable audience reviews than movies released in in
the last quarter.
The last two regressions estimated in this study examine the factors that influence
critic reviews (see Table 5 for regression results). Both models are linear ordinary least
squares regressions with the percentage of favorable critic reviews as the dependent
variable. The results indicate that critic reviews are not influenced by a movie’s
originality. Neither the variable to capture originality in general, nor the more specific
variables to capture the material upon which unoriginal movies are based prove to be
statistically significant. While critics seem indifferent to comedies, action films, and
family movies, the percentage of favorable critic reviews are about 7.3 percentage points
higher for dramas, 29.8 percentage points higher for foreign films, and about 11.5
percentage points lower for horror films than for films of omitted genres. As seen in
audience ratings estimations, movies produced by Fox and Sony receive less favorable
critic ratings than movies produced by omitted studios, and movies released in the first
three quarters of the year received lower percentages of favorable reviews than movies
Muldoon 12
released in the fourth quarter. Movies with higher budgets, earn a statistically significant
increase in favorable critic reviews, although at a trivial amount.
While the revenue estimations returned r-squared values of about 51%, the
audience ratings estimations had r-squared values of about 16% and the critic ratings
estimations had r-squared values of about 17%. The interpretation of these results,
suggests that about 51% of the variance in total revenues earned by a movie can be
explained by the variables in this study, while only 16% and 17% of the variance…
Department of Economics College of the Holy Cross
Fall 2012
Abstract: Over the past 25 years, the movie industry has expanded and the number of movies released each year has roughly tripled. At the same time, average movie revenues have remained fairly constant, and during several years, revenues even declined. How can one explain the steady increase in the number of movies released while revenues were constant and even decreasing? I hypothesize that there must be more movies being released based on previously released material (less originality), or perhaps the quality of the films being released has been decreasing. The purpose of this study is to provide some potential explanations for the drastic expansion of the movie industry.
Muldoon 2
I. Introduction
Movie critics despair over the lack of creativity in Hollywood. Last year there
were twenty-seven movie sequels released which is higher than in any previous year to
date (Ebert 2011). The top seven box office films of 2011 in the U.S. were all movie
sequels, which is unprecedented in box office history (Cieply 2011). In addition, several
other movies released were based on major books. The past year in film is part of a
continuing trend. In the past fifteen years, the strong performance and box office
presence of movie sequels and other films based on previous material have been on the
rise.
The goal of this study is examine the creativity of the movie industry. Creativity
is the “divergent thinking to conceive new ideas” (Heunks 1998), and in this study,
creativity will be measured by the originality of film. A film will be considered a
creative work if it is not based on any previously released material such as a book,
television show, comic book, or remake. A film will not be considered creative in this
paper if it is based on any type of previously released material.
Preliminary work has shown that over the past 25 years, the number of movies
released per year has roughly tripled, and the average revenues per movie have remained
fairly constant1. Between the years 2001 and 2006, the number of movies released
increased by about 41% while average revenues fell by about 33% (See Figures 1 and 2).
How does one reconcile this rapid expansion of the movie industry while revenues were
simultaneously declining? After looking at the number of employed screenwriters
(defined as members of the Writers Guild of America) over these 25 years, one can see
1 Preliminary work data are taken from the Box Office Mojo and Writer’s Guild of America West websites.
Muldoon 3
that the number of screenwriters remains fairly constant and increases by about 6%
between 2001 and 2006 (See Figure 3). This small increase in the number of employed
screenwriters does not begin to explain the drastic expansion of the movie industry.
Therefore, there must be other factors driving this expansion despite falling movie
revenues.
Data in this study confirm that over the past 15 years, there has been a drastic
decline in Hollywood’s creativity as illustrated in Table 1. In 1996, roughly 58.45% of
films released were original, compared to 53.94% in 2006, and 51.36% in 2011. While it
is clear that over the past two decades there has been a sizeable decrease in the creativity
of films released, it is less apparent why the decrease occurred. The goal of this study is
to provide some potential explanations for the decline in creativity.
There are many factors that may influence the creativity of films. Perhaps movie
producers rely more heavily on previous success in times of financial hardship. During a
recession, movie producers may prefer the lower risk of previously established material.
Previous work finds that sequels on average earn more box office revenue than other
films (Terry Butler and De’Armond, 2005 and Walls, 2008). Another factor may be
competition between movie studios. If one studio produces a movie that captures high
box office revenues and that causes a sensation, other studios may attempt to replicate the
success of that studio. Other factors that may influence the originality of a film are its
budget, the influence of critic reviews, and characteristics of the film such as the rating
and the genre. Also, the size of the studio, or more specifically the number of
screenwriters employed may influence the level of creativity. In years where many
screenwriters are producing screenplays, there may be an increase in innovative ideas,
Muldoon 4
and during years with fewer screenwriters one might expect that creativity of films would
decrease. On the other hand, perhaps the movie industry is truly unpredictable and there
is simply no recipe for innovation. Regarding the movie business, screenwriter William
Goldman notes, “With all due respect, nobody knows anything” (De Vany and Walls
1999).
While it is clear that less original movies are being released than fifteen years ago,
it is less obvious whether the decline in originality has impacted the quality of films. In
this study, critic and audience reviews are used as proxies for the quality of a movie. I
assume a high proportion of favorable reviews suggests a higher film quality, while a low
proportion of favorable reviews suggests a lower film quality. I find that the perceived
quality of a film is only slightly influenced by the originality of the film. Interestingly,
both critic reviews and audience reviews were slightly higher for movies that were based
on previous material. Favorable critic reviews averaged 48.95% and 49.17% for original
movies and unoriginal movies respectively. Favorable audience reviews averaged
59.79% and 60.20% for original movies and unoriginal movies respectively.
I find that unoriginal movies earn approximately 17.4% more at the box office
than original movies, holding other observable characteristics constant. More
specifically, the effect of a sequel and remake on gross revenue is about $6.97 million
and $8.89 million, respectively, for a movie with average revenues. I also find that there
is not a decline in movie quality between 1996 and 2011 as measured by critic and
audience reviews. Instead, movie quality actually increases as movies in 2011 received
statistically significant higher reviews than movies released in 2006 and 2001. The
Muldoon 5
results of this study suggest that the decline in movie originality must be driven by a
profit incentive and not by decreasing quality in film.
II. Literature Review
The existing literature does not examine the creative component of the movie
industry. Instead, the movie industry literature examines the variables that influence a
movie’s revenue and profits, and the risk and the lack of predictability that is
characteristic of the movie industry. In addition, the literature finds that movie sequels,
along with several other factors, are a vital element to the movie industry. Terry, Butler
and De’Armond (2005) examines several determinants of box office performance. These
variables include critic reviews, award nominations, sequels, budget, release exposure
and Motion Picture Association of America (MPAA) rating. They define the sequel
variable as “a movie derived from previous material” (145). This study finds that the
number of theaters showing a movie, the critic reviews, the movie budget, and the film
being a movie sequel all have positive and statistically significant effects on box office
revenue. They find that sequels earn about $36 million more at the box office than
movies that are not sequels. They also hypothesize that the main reason sequels are
produced is because there is already an assumed audience for a sequel to a successful
film.
In a similar study, Walls (2008) looks at movie profits as a function of the film
budget, the number of screens the film is released on, whether or not the film is a sequel,
whether or not the film has a major star, the genre, the rating, and the year produced.
While being a sequel does not guarantee a movie success, Walls finds that sequels have
Muldoon 6
higher profits than non-sequels by approximately $880,000. Through his non-parametric
analysis, he finds that sequels have positive expected profitability and mean profitability.
Assuming that movie producers strive to maximize box office revenues in order to
maximize profits, the articles by Terry, Butler and De’Armond (2005) and Walls (2008)
suggest that producing more movies that are based on previous material (non-creative
movies) is a safer way to ensure steady revenues.
In a similar approach, Ravid (2003) examines the risk and the lack of
predictability of the movie business and how movie producers attempt to guarantee
revenues. He examines several “puzzles” of the movie industry, one of these being
“deviation from profit or value maximization.” He suggests that this deviation can be
explained as an attempt to reduce risk by risk-averse studio executives. He also notes that
higher ratings correspond with higher revenues. As a result, producers strive to reproduce
the formula used for successful films, which explains why movie sequels and imitative
films tend to receive higher film ratings. According to Ravid, producing sequels may be
viewed as “corporate hedging behavior” (40).
Economic theory suggests that consumers enjoy and prefer variety. However, this
is not clear when looking at the increasing numbers of unoriginal movies and the revenue
earned by each. Previous literature suggests that movies based on previous material have
higher profits than original films. The literature also suggests that other factors are
important to a film’s success such as the naming strategy, advertising, the number of
theaters showing a film, and the movie budget. Because the movie business is risky,
evidence indicates that producers try to minimize risk by reproducing films similar to
those which have been successful in the past.
Muldoon 7
III. Data Sources
I use a fifteen-year time span between 1996 and 2011. Because a large amount of
the data was collected by hand, I use data from 1996, 2001, 2006, and 2011 rather than
each year. This produces a total of 878 movies.
The existing literature highlights several variables that influence movie revenue,
including the rating and genre of the film, the release date of the film, the budget, the
critic reviews, the number of theaters and a dummy variable for whether or not the film is
an original work. In addition to the variables suggested by the literature, this study
includes ratings provided by the audience and critics, dummy variables for the year the
film is released, and the studio that releases the film. Most importantly, this study
includes variables to capture the originality of the film. The first variable that captures
the originality of the film is a dummy variable “original.” Original movies are films that
are not based on any previous material. The second method that captures the originality
(or unoriginality) of the film specifies dummy variables for each type of unoriginal
movie. Dummy variables are created for sequels, remakes, movies based on books,
movies based on plays, movies based on television shows, and movies based comic
books. These variables determine whether the effect of an unoriginal film depends on the
type of source material. In other words, a sequel and a movie based on a book may have
a different or more significant effect on revenues or reviews. Information regarding the
originality of the film is obtained from the Rotten Tomatoes, IMDB and Wikipedia
websites.
Muldoon 8
Similar to Terry, Butler and De’Armond (2005), I use data from Box Office Mojo
(boxofficemojo.com), the Rotten Tomatoes website (rottentomatoes.com), the-
numbers.com, and Movies.com. Data on the opening and eventual total number of
theaters a movie was released on, the opening and total revenue of a movie, the date of
the movie’s release, and the movie’s production studio are from Box Office Mojo. Data
on MPAA ratings, genre, critic reviews, and audience reviews are from Rotten Tomatoes.
The critic and audience reviews appear as percentages of favorable reviews and serve as
proxies in this study for the quality of a film. Since data on movie budgets is not
available for each movie, budget data are collected from Box Office Mojo and the-
numbers.com. If budget data are available for a given movie from both sources, the two
budgets are averaged
I create several other dummy variables. First, I generate dummy variables for the
quarter and year of the movie’s release. Dummy variables are also generated for the six
most common movie genres, the rating of the movie, and the top five producing studios.
IV. Results
There are six different regression models in this study. Because one of my
hypotheses about the decline in movie originality rests on the assumption that movie
studios are profit driven and that unoriginal movies earn higher box office revenues, the
first two regression models examine the determinants of box office performance. These
two models are modified versions of the model used by Terry, Butler and De’Armond
(2005) which utilizes a log linear model of revenue as a function of a movie’s critic
reviews, release date, budget, number of award nominations, and whether or not the
Muldoon 9
movie is rated R, rated G, a sequel, an action film, and whether or not the movie was
released within seven days of a major holiday.
The first two models in this study are log-linear ordinary least squares regressions
(see Table 3 for regression results). Both models have the natural log of revenue as the
independent variable, and the dependent variables capture the impact of critic ratings,
audience ratings, MPAA rating, genre, producing studio, originality, budget, and the
quarter and year in which the movie was released. These two regression models are the
same except that the first model includes a single variable to capture whether or not the
movie is original, while the second model includes more specific variables to capture the
type of source material. These variables allow for varying revenue effects of the source
material. For example, a sequel may boost a movie’s revenue more than being based on a
play, etc.
The results of these two models suggest that unoriginal movies earn
approximately 17.4% more at the box office than original movies, which is a statistically
significant estimate at 5%. When broken down into more specific categories, the results
suggest that a movie based on a book, television show, or play increases revenue,
although these figures are not statistically significant. Movies based on comic books
earned lower box office revenues than original movies, although this finding is not
statistically significant. Remakes and sequels both earn higher box office revenues than
original films. For a movie that is a sequel, the average gross revenue is about $64.6
million. Given a coefficient of 0.291, $18.8 million of the revenue is attributable to the
movie being a sequel. For a movie that is a remake, the average gross revenue is
Muldoon 10
approximately $33.1 million. Given a coefficient of 0.371, $12.3 million of the revenue
is attributable to the movie being a sequel.
The results of these two models also suggest that while critic ratings do not
significantly impact a movie’s revenue, audience ratings are an important factor; a one
percentage point increase in the favorable audience reviews lead to a 2.6% increase in
box office revenue. The results also indicate that rated movies earn statistically
significant higher box office revenues than the omitted category of unrated movies,
perhaps because rated movies have wider releases than unrated movies in general. In this
study, the top six most common genres (drama, comedy, action, foreign, horror and
family) are included in the regression2. The results suggest that dramas with average
revenues earn about $11.2 million less than movies of omitted genres, comedies with
average revenues earn about $8.29 more than movies of omitted genres, and horror films
with average revenues earn about $19 million more than movies of omitted genres.
Finally, movies produced by four of the top five studios earn statistically significantly
higher revenues than other studios.
Because my other hypothesis regarding the decline in movie originality over the
last 15 years rests on a change in the perceived quality of movies, the third and fourth
models in this study examine the impact of originality on audience reviews as measured
by the percentage of favorable reviews (see Table 4 for regression results). Both models
are linear ordinary least squares regressions. While the results suggest that originality in
general does not impact audience reviews, movie sequels in particular receive about 4.9
2 Many genre categories are omitted because there were too few movies in these categories to be included in the regression. Some of the omitted genre categories include adventure, animation, concert, crime, documentary, fantasy, musical, romance, science fiction, thriller, and western.
Muldoon 11
percentage point higher favorable audience reviews than unoriginal movies, which is
statistically significant at 5%. The results also indicate that a movie’s budget and ratings
do not prove to be statistically significant predictors of the audience reviews. Dramas
and foreign films earn about 5.78 and 20 percentage point higher favorable audience
reviews, respectively, than movies of omitted genres. While the producing studio does
not seem to significantly impact audience reviews, movies produced by Sony receive less
favorable audience reviews than movies produced by omitted studios by about 4.6
percentage points. Movies released in the first three quarters of the year receive
statistically lower percentages of favorable audience reviews than movies released in in
the last quarter.
The last two regressions estimated in this study examine the factors that influence
critic reviews (see Table 5 for regression results). Both models are linear ordinary least
squares regressions with the percentage of favorable critic reviews as the dependent
variable. The results indicate that critic reviews are not influenced by a movie’s
originality. Neither the variable to capture originality in general, nor the more specific
variables to capture the material upon which unoriginal movies are based prove to be
statistically significant. While critics seem indifferent to comedies, action films, and
family movies, the percentage of favorable critic reviews are about 7.3 percentage points
higher for dramas, 29.8 percentage points higher for foreign films, and about 11.5
percentage points lower for horror films than for films of omitted genres. As seen in
audience ratings estimations, movies produced by Fox and Sony receive less favorable
critic ratings than movies produced by omitted studios, and movies released in the first
three quarters of the year received lower percentages of favorable reviews than movies
Muldoon 12
released in the fourth quarter. Movies with higher budgets, earn a statistically significant
increase in favorable critic reviews, although at a trivial amount.
While the revenue estimations returned r-squared values of about 51%, the
audience ratings estimations had r-squared values of about 16% and the critic ratings
estimations had r-squared values of about 17%. The interpretation of these results,
suggests that about 51% of the variance in total revenues earned by a movie can be
explained by the variables in this study, while only 16% and 17% of the variance…
Related Documents
![NETWORKS & CREATIVITY Session Two: Conceptualizing Creativity creativity. from itskay [kala]itskay [kala]](https://static.cupdf.com/doc/110x72/5513bb755503464b298b4703/networks-creativity-session-two-conceptualizing-creativity-creativity-from-itskay-kalaitskay-kala.jpg)
