Prevision model and empirical test of box office results for sequels

HAL Id: hal-03200091

https://u-bourgogne.hal.science/hal-03200091

Submitted on 13 Oct 2021

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-

entic research documents, whether they are pub-

lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diusion de documents

scientiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

Prevision model and empirical test of box oce results

for sequels

Bertrand Belvaux, Rémi Mencarelli

To cite this version:

Bertrand Belvaux, Rémi Mencarelli. Prevision model and empirical test of box oce results for sequels.

Journal of Business Research, 2021, 130, pp.38-48. �10.1016/j.jbusres.2021.03.008�. �hal-03200091�

Prevision model and empirical test of the box office results for sequels

Bertrand Belvaux

Rémi Mencarelli

IAE Dijon – CREGO E.A. 7317 - University of Burgundy

2 boulevard Gabriel, BP 26611, 21066 Dijon Cedex, France

Tel : +33 380 39 64 70

E-Mail : [email protected]

IAE Savoie Mont Blanc – IREGE – University of Savoie Mont-Blanc

4, chemin de Bellevue, 74944 Annecy-Le-Vieux Cedex, France

Tel: +33 450 09 24 00

Email: [email protected]

* Corresponding author

Prevision model and empirical test of box office results for sequels

Abstract

As studios release an increasing number of movie sequels, scholars have begun to examine

this strategic choice. Prior studies use standard models of box office performance to evaluate

sequels’ performance and have mainly compared the box office results of the original movie

with those of its sequel. However, sequels hold a unique position in the motion picture market

since they are strongly associated with the original movie. Using the accessibility-

diagnosticity framework, this research investigates the drivers behind the success of sequels

and examines specifically the original movie’s impact through the role of reviews. The results

– from 232 movies (116 original movies and 116 first sequels) – demonstrate the direct

impact of the original movie’s reviews on the sequel’s performance.

Keywords

Movie industry; Review; Sequel; Box office; Accessibility-diagnosticity framework

This research did not receive any specific grant from funding agencies in the public,

commercial, or not-for-profit sectors.

Prevision model and empirical test of box office results for sequels

1. Introduction

The lifespan of movies in theaters has been steadily decreasing over the last several years.

This is due to the abundant and constantly increasing number of films released, a sharp

growth in distribution (through the number of screens per movie) and alternative means to

access films (video-on-demand, streaming, screening rooms), which has led to studios

perceiving the release of new movies as highly risky. Instead, over the past few years, they

have capitalized on past successes, exploiting well-established movies with enduring

reputations and strong consumer preferences (Eliashberg et al., 2006), and thus focusing more

on motion picture sequels than on stand-alone movies (Filson & Havlicek, 2018). With this

strategic orientation, studios have aimed at minimizing risk by capitalizing "on the success of

an original movie by producing another film that reprises the same characters evolving in a

new situation" (Sood & Drèze, 2006). In 2019, the top ten box office successes worldwide

were sequels, compared to only two in 2000.

Given the enormous economic impact of sequels in the movie industry, scholars have begun

to examine this strategic choice (Filson & Havlicek, 2018). Some studies have used sequels as

a control variable in existing models of box office performance; others have evaluated and

compared the box office results of sequels with those of the original movies or

contemporaneous non-sequel movies (Ravid, 1999; Basuroy & Chatterjee, 2008; Hennig-

Thurau et al., 2009; Moon et al., 2010; Dhar et al., 2012). Most existing studies use standard

models of box office performance to evaluate sequels’ performance, even though sequels hold

a unique position in the motion picture market since they are strongly associated with an

original work (Chisholm et al., 2015). Some studies equate sequels with brand extension,

thereby indirectly associating the sequel with the original movie (Moon et al., 2010), but they

have neglected to examine the role of any pre-existing signals from the original movie,

especially the role of reviews. However, experiential information generated about the original

movie (i.e., reviews from consumers and critics) can be used by moviegoers to make

inferences and judge the sequel and, in turn, affect the box office performance of the sequel.

Hence, this research investigates what drives the success of sequels and demonstrates that

over and above the usual characteristics that affect the performance of a movie (i.e.,

characteristics of the film, studios’ actions and third-party information), the drivers of a

sequel may also include the reviews associated with the original movie. Using the

accessibility-diagnosticity framework, which has been recently applied to motion picture box

office performance (Joshi & Mao, 2012; Knapp et al., 2014; Carrillat et al., 2018), we

hypothesize and empirically show that these peripheral signals do not originate from the

sequel itself, but impact the sequel’s box office performance, specifically during the release

phase. This result explains the unique position sequels hold in the movie market compared to

stand-alone movies and why a specific performance model should be developed for sequels.

In addition to the indirect effect of the original movie on the sequel through a brand extension

effect, we highlight the direct impact of the original movie’s reviews on the sequel’s

performance with the accessibility-diagnosticity framework, demonstrating that reviews from

an original movie can be applied to a sequel and thus have an effect over time. From a

managerial perspective, our results suggest that it would be interesting for cinema executives

to reactivate positive reviews of the original movie at the launch of a sequel.

In this paper, we first review the literature dedicated to the box office performance of motion

pictures and, based on the accessibility-diagnosticity framework, we develop a specific model

for sequels. The methodology used and the results obtained are then presented. To test the

model, we use data from 232 recent movies (116 original movies and 116 first sequels).

Finally, the contributions and limitations of the study are noted, and avenues for future

research are proposed.

2. Conceptual framework

2.1. Motion picture box office performance: original movie versus sequel

Researchers have produced a substantial number of studies analyzing the box office

performance of films. Like any other product, a movie’s success is based on the interaction

between supply and demand. The supply strategy is developed by two key players in the

motion picture industry, the producer and the distributor; the major studios usually assume

these two functions internally (Hadida, 2009). The producer defines the main characteristics

of a movie: genre, country of origin, MPAA ratings, selection of the director and the actors.

The associated budget is also defined by the producer according to the expected performance

of the film and its anticipated profits (Dhar et al., 2012). The producer entrusts the distributor

with a mandate to define the marketing actions associated with the release strategy (Prieto-

Rodriguez et al., 2015). In order to maximize revenues, the distributor determines the optimal

combination of resources – such as advertising expenditures and distribution intensity through

number of screens – by estimating the movie’s potential audience.

With regard to demand, a movie’s attractiveness is based on its reputation and image among

consumers (Liu, 2006). Even if a movie is an original experience (Nelson, 1970; Chen et al.,

2012), potential moviegoers can develop preferences without consuming the film. They use

signals that resonate with their past experience. For example, the genre, the actors or the

director are signals that can influence a priori the consumer’s preference for a movie. The

audience can also use external information such as expert reviews or interpersonal

communications (Holbrook, 2005). The influence of third-party information used by

moviegoers has been demonstrated in various studies (Liu, 2006; Karniouchina, 2011; Lee et

al., 2015). Third parties are people who share their experiences and opinions of a film with

potential moviegoers. These information sources are usually perceived as credible and

trustworthy (Liu, 2006; Chakravarty et al., 2010), and moviegoers view them as an indicator

of quality (Hennig-Thurau et al., 2006). Third-party information is all the more important for

entertainment products because it sparks discussions among consumers (Liu, 2006).

Consequently, a vast majority of moviegoers base their decision processes on third-party

information (Karniouchina, 2011). Many studies have analyzed the role of third parties and

distinguished:

• the sources: expert-based reviews (professional critics’ reviews) and non-expert

reviews (consumer reviews). These third parties play the role of influencer or

predictor (Basuroy et al. 2003);

• the characteristics: volume (total number of reviews, which contributes to the

reputation and awareness of a film’s existence; i.e., the informative dimension) and

valence (nature – positive or negative – of the reviews, which contributes to the

movie’s image; i.e., the persuasive dimension) are the main characteristics

used to

describe reviews (Liu, 2006).

To sum up, the box office success of a movie is based on three main factors: the

characteristics of the film, the intensity of the marketing strategy (mainly through the

advertising budget and number of screens) and the reviews (from critics and moviegoers).

In this research context, a number of studies have analyzed sequels’ performance (Table 1).

Broadly speaking, the term "sequel" is used to refer to the production of a film based on the

Other characteristics were identified, including consistency (Karniouchina, 2011), dispersion (Godes &

Mayzlin, 2004), entropy (Dellarocas et al., 2007) and observability (Dellarocas et al., 2010).

success of a first, original movie (Sood & Drèze, 2006). These studies of sequels aimed to

analyze their box office performance from two perspectives.

[insert Table 1 here]

First, some studies compare sequels with non-sequel movies. Initial studies use the sequels as

a control variable. Some conclude that this variable does not affect the existing models

(Basuroy et al., 2003; Liu, 2006), while others show that it has a significant impact (Basuroy

et al., 2006; Hennig-Thurau et al., 2006; Boatwright et al., 2007; Karniouchina, 2011). Other

studies have evaluated the box office performance of sequels by comparing them to non-

sequel movies released during the same period. Basuroy and Chatterjee (2008) show that

sequels do better than contemporaneous non-sequels; Dhar et al. (2012) confirm this result. In

both studies, the authors provide explanations based on brand extension theory (Aaker &

Keller, 1990). They conceptualize sequels as brand extensions, as do several other studies

(Sood & Drèze, 2006; Hennig-Thurau et al., 2009). In other words, the studios use existing

brand names and look for ways to build on the similarity of characteristics between the

original movie and the sequel (e.g., genre, stars, directors, storyline) in order to enhance brand

awareness. The perceived similarity between the original movie (i.e., the brand-parent

category) and the sequel (i.e., the extension category) leads to a process of assimilation (Sood

& Drèze, 2006), shapes quality perceptions (Hennig-Thurau et al., 2009) and positively

influences the box office success of the extension (Basuroy & Chatterjee, 2008; Dhar et al.,

2012).

Second, some studies compare the sequel’s performance with that of the original movie.

Using an economic approach (Lazear, 2004), Basuroy and Chatterjee (2008) assume that the

box office earnings of the original film will always be greater than those of the sequel due to

the inevitable phenomenon of regression to the mean. They show this effect empirically on a

sample of 11 sequels. Using a larger sample and a model based on brand extension value,

Hennig-Thurau et al. (2009) provide empirical evidence that sequels generate high average

revenue and reduce the associated risk. Using an individual-level analysis, Moon et al. (2010)

emphasize that sequels have a better box office performance due to the original movie’s brand

power, the initial success of the original film and the established fan base of the original. They

confirm this assumption during the release phase but find that this effect dissipates quickly

after the first two weeks. Moon et al. (2010) also observe that sequels receive lower ratings

from moviegoers as a result of satiation (lack of novelty and surprise for an experiential

product). Dhar et al. (2012) use a sample of movies spanning 26 years and demonstrate

empirically that sequels attract more moviegoers than stand-alone movies released during the

same period, but fewer than the original movies. However, they generate more profit than do

the original movies. Filson and Havlicek (2018) extend this to global film franchises and

show that the performance of sequels tends to decline as extensions are introduced.

Several factors can explain the contradictory results obtained in these studies: the instability

of the results given the small numbers of sequels in some of the samples, the different study

periods in changing and complex environments and the adoption of different time horizons to

determine a motion picture’s box office performance. Although they arrive at different results

and conclusions, these studies generally compare the sequel’s box office results with those of

its original movie or a contemporaneous non-sequel movie. They do not examine the specific

mechanisms underlying a sequel’s box office performance, in particular the role of pre-

existing signals from the original movie. These mechanisms deserve attention and should be

considered in light of the initial release of the original movie on the market (Chisholm et al.,

2015).

2.2. Specific mechanisms of sequels’ box office performance

Based on previous research dedicated to box office performance, we assume that, like stand-

alone movies, sequels’ box office performance depends mainly on the characteristics of the

movie (genre, director, actors, MPAA rating, country of origin, producer status), the intensity

of the studios’ actions (production budget and screens) and the reviews (from critics and

moviegoers) of the sequel. We do not propose formal hypotheses about these effects as they

are largely supported by previous empirical studies.

In addition to these expected effects, we assume that the original movie affects the sequel’s

box office performance through the persistence of its reviews. To identify these specific

mechanisms, we use the accessibility-diagnosticity framework (Feldman and Lynch, 1988),

which has been successfully applied in the context of motion picture box office performance

(Joshi & Mao, 2012; Knapp et al., 2014; Carrillat et al., 2018). According to this framework,

consumers rely on the most diagnostic and accessible cues to make a judgment. More

precisely, a piece of information is more likely to be used as a cue in a judgment if it is

perceived as relevant to the judgment (diagnosticity), it is accessible for use in the judgment

and other information is less accessible (accessibility). We use arguments based on this

theoretical framework to explain the influence of the original movie’s reviews on the sequel’s

box office performance.

First, consumers will use available signals to assess the quality of the sequel during the

release phase (Akdeniz & Talay, 2013). In this context, moviegoers can draw on existing and

credible signals based on their own consumption experience of the original movie and/or on

information generated by other moviegoers who saw the original movie. Consumers will also

associate the sequel with the original movie (Basuroy et al., 2006; Hennig-Thurau et al.,

2009). In other words, information generated about the original movie (through consumption

and/or by third parties) contributes a priori to the reputation and image of the sequel. These

cues are also more instructive than factual information (e.g., genre, MPAA ratings) as they are

experiential information. Consequently, because of their accessibility and diagnosticity, the

reviews associated with the original movies become an important information source for

sequel judgments, especially during the release phase. Consequently, moviegoers can be

motivated to choose to see the sequel early, without waiting for the emergence of sequel-

specific reviews.

However, according to Joshi and Mao (2012), in the first week after release, direct

experiential cues about the sequel become more accessible. Consequently, the effect of the

information used to evaluate the movie quality earlier on, including the reviews of the original

movie, is likely to dissipate. It is thus possible to assume the reviews of the original movie

influence the short-term box office performance of the sequel but not its long-term

performance.

Previous studies suggest both a volume and a valence effect of reviews on box office

performance (Table 1). However, for expert-based reviews, the volume effect is less

persistent over time, in contrast to the volume effect of consumer reviews (Larceneux, 2007).

The volume of critics’ reviews also remains broadly constant, with each expert evaluating all

the films that are launched. Finally, several studies highlight the prominence of a valence

effect for critics’ reviews (Desai & Basuroy, 2005; Basuroy et al., 2006; Dellarocas et al.,

2007). Therefore, we assume that the volume and valence of consumer reviews and the

valence of critics’ reviews of the original movie can directly impact the short-term box office

performance of the sequel.

H1a. Consumer reviews (volume and valence) of the original movie have a positive impact on

the sequel’s short-term box office performance (first week) but do not affect the sequel’s

long-term performance (after the first week).

H1b. Critics’ reviews (valence) of the original movie have a positive impact on the sequel’s

short-term box office performance (first week) but do not affect the sequel’s long-term

performance (after the first week).

Second, the effect of the original movie’s reviews on the sequel’s box office performance will

probably be moderated by the time interval between the release of the two movies (Basuroy &

Chatterjee, 2008; Filson & Havlicek, 2018). The persistence of reviews has already been

documented in different studies (Godes & Silva, 2012; King et al., 2014). As noted by King et

al. (2014), persistence of reviews means that existing reviews significantly influence future

ones. According to the accessibility-diagnosticity framework, a short time period makes the

reviews of the original movie more accessible to potential consumers of the sequel. In

consequence, a short time period between the two movies should increase the influence of the

original movie’s reputation and image on the sequel’s short-term box office performance.

H2. The shorter the time interval between the release of the original movie and the sequel, the

stronger is the impact of consumers’ and critics’ reviews on the sequel’s short-term box

office performance (first week).

Third, the reviews generated for the original movie can influence the box office performance

of the sequel by affecting the reviews of the sequel. Like Duan et al. (2008), we consider

reviews to be an endogenous variable. In other words, consumers’ and critics’ reviews of a

sequel depend on the reviews previously generated for the original movie. First, for a sequel,

the evaluation can be general due to the emergence of brand equity (Moon et al., 2010). The

great similarity between the original movie and the sequel, which increases the perceived

diagnosticity of cues associated with the original movie, will favorably impact the

endogenous nature of the reviews (Joshi & Mao, 2012). Second, considering the consistency

and perceived similarity between the sequel and the original movie (Sood & Drèze, 2006), the

assessment criteria will probably be the same and will be reactivated in a similar way (Moon

et al., 2010). Therefore, by distinguishing the consumers’ and critics’ reviews, we posit that

the reviews associated with the original movie can influence the sequel’s box office

performance through their impact on the sequel’s reviews. Formally:

H3a. The consumer reviews (volume and valence) of the original movie have an indirect and

positive impact on the sequel’s long-term box office performance through the consumer

reviews of the sequel.

H3b. The critics’ reviews (valence) of the original movie have an indirect and positive impact

on the sequel’s long-term box office performance through the critics’ reviews of the

sequel.

Figure 1 summarizes the model of box office performance for a sequel. The model is based on

standard effects identified for a movie (intra-movie effects). It also explicitly integrates the

influence of the original movie and the associated signals that predated the release of the

sequel (inter-movies effect).

[insert Figure 1 here]

3. Sequel box office prevision model

3.1. Data

Original data were collected on 232 movies (116 original movies and 116 first sequel movies)

on the American market between 1999 and 2014 (Appendix 1). The data were collected from

boxofficemojo.com, the-numbers.com, rottentomatoes.com and imdb.com (like Chakravarty

et al., 2010). As reviews on the IMDb platform began appearing in 1999, franchises that

started before this date are excluded from the database. Indeed, certain factors are likely to

change strongly over such long periods of time (e.g., economic conditions, critics’ and

audiences’ tastes) and can therefore alter results (Elberse & Eliashberg, 2003).

Numerous studies assess box office performance dynamics by considering both the overall

performance and that of the first week following a movie’s release (Basuroy et al., 2006).

However, to test our hypotheses and avoid data nesting, we make a distinction between results

in the first week (short-term performance) and after (long-term performance). To account for

the temporal dimension, these data were adjusted for inflation over the period and for

seasonal variations (annual and monthly trends). Seasonal correction coefficients were based

on a broader film database (6,650 films shown over the same period).

We use three sets of variables to explain the commercial performance of a sequel: movie

characteristics, studios’ actions and reviews (Table 2).

[insert Table 2]

Like Hennig-Thurau et al. (2012), we choose several movie characteristics: country of origin

(USA/outside the USA), status of the studio (major/independent), genre, MPAA

classification, director and actors. For genre we use the four major categories on the IMDb

website: action-adventure, animation, drama-thriller and comedy. Like Moon et al. (2010),

two categories are used for the MPAA classification: R and non-R (PG-13, PG and G). To

analyze the role of the director and actors (the two main actors), the total box office

performance of their previous movies is used as an indicator of attractiveness (Hennig-Thurau

et al., 2006). Using past success as a predictor of future success is the simplest indicator for

measuring the effect of the director and/or "star" actors (Rein et al., 1987; Karniouchina,

2011).

In addition, we include two main variables associated with studios’ actions (Hennig-Thurau et

al., 2006): the production budget

and the number of screens at the movie’s release. These

reflect the ambition of the producer and distributor, respectively.

Finally, for the reviews, the valence of the critics’ reviews was captured by the Metascore

index from the IMDb website, summarizing and standardizing some thirty professional

sources, the scope of the critic being constant (Hennig-Thurau et al., 2012). For consumers,

the reviews variables were captured through the evaluations posted on the IMDb website

(average score and number of ratings). For the temporal dynamic (Liu, 2006), the sequel’s

reviews were recorded during the first week (intra-movie effect) and those of the original

movie from its release period until the date of release of the sequel (inter-movie effect).

3.2. Model specification

Like Zufryden (1996) and Elberse and Eliashberg (2003), we use log-log modeling. This

makes it possible to take into account the multiplier effects of variables

and analyze the

effects in the form of elasticities. The logarithmic transformation of continuous data also

reduces the strong variance between films and limits heteroscedasticity. Moreover, the

formulation of a multiplier model is consistent with the multi-level theoretical models used to

estimate the commercial performance of a movie (Duan et al., 2008). Finally, in order to

compare results, we use the same choice process as in several other research studies looking

at sequels (Elberse & Eliashberg, 2003; Basuroy et al., 2006; Hennig-Thurau et al., 2006).

The relationship between the endogenous variable and the exogenous variables is written as:





 



 







 









 





The production budget corresponds to what the North American film industry calls the "negative cost" or the

expense of producing the film excluding distribution and promotional expenses. The amount is adjusted for

inflation.

For example, there will be no income unless a copy is presented.

where X is the vector of continuous variables and Z the vector of dummy variables.

In order to evaluate the original movie’s impact on the sequel, the analysis was carried out in

two stages: first, the test of the intra-movie model (internal effects of the sequel’s

performance) and then the test of the inter-movie model (effect of the original movie on the

sequel).

3.3. The intra-movie model of sequel performance

The first step is to estimate the sequel’s performance through its internal dynamics: the

movie’s characteristics, the studios’ actions and the reviews. As mentioned in the conceptual

framework, we made no hypotheses on these effects given their validation in previous studies.

The production budget and the number of theaters can raise problems of endogeneity as they

directly explain performance and are also determined by the expected performance of the

film. A three-equation model is therefore used:

Performance = f (constant, budget, theaters, country, genre, MPAA, producer, director, actors, valence of critics’ reviews,

valence of consumer reviews, volume of consumer reviews)

Budget = f (constant, country, genre, MPAA, producer, director, actors)

Theaters = f (constant, budget, country, genre, MPAA, producer, director, actors)

Parameters of the model were estimated using the structural equation modeling (partial least

squares method). The classic model for forecasting a movie’s box office performance is

satisfactory in the sequel case (Table 3). The performance of a sequel thus depends at first

mainly on studios’ actions (production budget and number of theaters) and professional

reviews, and then, after the first week, on spectator’s reviews. The production budget

positively affects the commercial performance of the movie (β = 0.251, p < 0.01 during the

Continuous variables are transformed into logarithmic format. Some values can take a zero value; the applied

transformation is: Y = ln (X+1). Dummy variables are introduced in exponential form to reflect the absence of

the factor.

first week and β = 0.254, p < 0.01 after the first week). The number of theaters also has a

positive effect (β =1.695, p < 0.01 during the first week, β = 1.365, p < 0.01 after the first

week). The effects of reviews also play a role in the commercial performance of the sequel.

The influence of professional reviews is strongest immediately after the film is released (β =

0.584, p < 0.01), before the influence of consumer reviews kicks in (both valence β = 0.441, p

< 0.01, and volume β = 0.420, p < 0.05). On the other hand, the movie’s characteristics do not

affect the result. The "star" effect does not impact the sequel’s commercial performance.

These initial results remain broadly consistent with those identified in the literature regarding

the commercial performance of a movie, with a key role played by studios’ actions and

reviews. However, we note that the movie’s own characteristics have almost no effect on

commercial performance, which suggests the predominant influence of the sequel as a brand

(Sood & Drèze, 2006).

[insert Table 3]

3.4. The inter-movies model of sequel performance

The second step is to analyze the box office release from an inter-movie perspective. This

consists of assessing the extent to which a sequel’s performance is affected by the original

film. The analysis model is therefore based on explaining residuals of the first model. It

reduces the number of parameters to be estimated and limits the endogeneity effects between

the variables of the two movies (original movie and sequel).

We determine both the direct and the indirect effects of the original movie’s reviews on the

commercial performance of the sequel by using mediation tests on the hypotheses. Indirect

effects are analyzed via the influence of variables related to the reviews of the sequel. This

test follows the procedure described by Zhao et al. (2010) using the structural equation

method.

The objective is to use the sequel’s reviews to analyze the direct and indirect effects

of the original movie’s reviews on the sequel’s performance.

The results first highlight the direct effects of the audience reviews associated with the

original movie on the sequel’s performance (Table 4). The audience reviews of the original

movie (valence and volume) thus directly affect the performance of the sequel in the first

week (β = 0.458, p < 0.01; β = 0.129, p < 0.01). However, after the first week, they no longer

directly affect the performance of the sequel. Moreover, the professionals’ reviews of the

original movie do not directly affect the performance of the sequel in either the short (first

week) or long term (after the first week). The results obtained therefore support H1a and

reject H1b.

In the sequel box office model, the time interval between the original movie and the sequel

can moderate the effect of the original movie’s reviews on the sequel’s box office

performance (H2). Regressions were performed with an interaction effect (number of months

between the two movies) explaining the sequel’s performance through the reviews of the

original movie (Table 5). The time interval between the original movie and the sequel has a

weak and negative effect on the relationship between consumer reviews (volume) and the

sequel’s short-term box office performance (β = - 0.005, p < 0.1). In other words, the shorter

the time interval between original and sequel, the stronger is the effect of the consumer

reviews (volume) on the sequel’s performance. The effects of other signals associated with

the original movie on the performance of the sequel are not moderated by the time interval;

these include the valence of consumer reviews (β = -0.009, n.s) and the valence of critics’

reviews (β = 0.001, n.s). These results partially support H2. The nature of the reviews’

characteristics can explain the results. The quantitative aspect (volume) of the reviews

The analyses were performed by the Lavaan module of R software for analysis of covariance structures and the

partial least squares method with Smartpls software for verification.

increases their accessibility, reactivates them easily and enhances consumer awareness of the

sequel (Liu, 2006), contrary to the qualitative nature of the cues (valence of reviews from

consumers and critics), which can have inconsistent effects over time (Godes & Silva, 2012).

Secondly, the results show the indirect effects of audience reviews of the original movie

(volume and valence) on the long-term performance of the sequel (Table 4). A good audience

evaluation of the original film thus generates a favorable opinion of the sequel (β = 0.484, p <

0.01), which positively determines the sequel's box office results. Similarly, a large volume of

reviews for the original movie positively affects the volume of reviews associated with the

sequel (β = 0.837, p < 0.01), which then determines its commercial success. On the other

hand, while the valence of critiques of the original movie affects the valence of the critiques

of the sequel (β = 0.701, p < 0.01), there is no significant indirect effect on performance.

These results therefore support H3a and reject H3b.

[insert Table 4 here]

[insert Table 5 here]

While these results only fully support two hypotheses (Table 6), they do confirm the unique

situation of the sequel, which benefits from the previous existence of the original movie.

Unlike an original movie, where audience reviews only can play a role after the first week

given the time it takes for them to emerge (Liu, 2006), a sequel immediately benefits from the

public reaction to the original movie, which will directly affect the short-term performance of

the sequel. The results confirm the role played by some pre-existing experiential information

– both informative (volume) and persuasive (valence) – associated with the original movie.

Because of their accessibility and diagnosticity, reviews of the original movies accelerate the

viewers’ decision-making about the sequel during the first week. Moreover, the influence of

the reviews of the original movie on those of the sequel emphasizes the endogenous nature of

the reviews (Duan et al., 2008). Finally, consistent with the accessibility-diagnosticity

framework (Joshi & Mao, 2012), the direct impact of these quality signals becomes diluted as

other quality signals directly associated with the sequel emerge.

[insert Table 6 here]

4. Conclusion

4.1. Contributions to the literature

In developing a model to forecast a sequel’s box office performance, we make three main

contributions to the literature. First, we identify the specific mechanisms underlying a

sequel’s box office performance and thus the need to develop dedicated performance models

for sequels. Whereas the performance of a stand-alone movie is affected by cues related to a

studio's prelaunch decisions (i.e., production budget, advertising budget, distribution intensity,

star power, director) and cues related to third-party information (i.e., professional reviews

from critics and consumer reviews) (Bharadwaj et al., 2017), the performance of a sequel is

also affected by the experiential information associated with the original movie. These

peripheral signals impact the sequel’s box office performance during the opening week. This

result explains why sequels hold a unique position in the movie industry and can have a

competitive advantage compared to their contemporaneous non-sequel movies (Hadida,

2009). This strategy will specifically reduce the risk of failure, but does not necessarily ensure

success (which is based more on the quality of the sequel).

Second, our paper adds to previous studies that have specifically analyzed sequels’

performance. These studies have mainly compared the performance of the sequel with the

performance of the original movie (Basuroy & Chatterjee, 2008; Hennig-Thurau et al., 2009;

Moon et al., 2010; Dhar et al., 2012; Filson & Havlicek, 2018), using a brand extension effect

to indirectly take into account the effect of the original movie on the sequel. Based on the

accessibility-diagnosticity framework, the test in our model goes further by acknowledging

the explicit impact of the original movie on the sequel’s box office performance through the

role of the reviews. The brand and product-related characteristics (genre, stars, director,

storyline) are not the only signals affecting the performance of the sequel. The experiential

information associated with the original movie and reactivated in this brand extension context

will also directly impact the sequel’s performance.

Third, our results demonstrate the persistence of reviews, whereas previous studies mainly

emphasize the short-term effects of reviews in the context of a stand-alone movie's box office

performance (Liu, 2006). Here, the accessibility-diagnosticity framework offers a convincing

theoretical explanation for this persistence effect of reviews on sequels and confirms its

relevance in assessing the effect of these signals (Carrillat et al., 2018). Specifically, the

volume and valence of consumer reviews affect the sequel’s short-term performance given the

availability and credibility of the signals that guide moviegoers in their decision-making

process by minimizing uncertainty. The direct impact of these experiential signals then

becomes diluted as other direct experiential cues about the sequel become more accessible.

Consequently, as expected, the original movie's reviews have no direct effect on the long-term

performance of the sequel. Contrary to the idea that the influence of reviews diminishes

quickly over time (Duan, Gu & Whinston, 2008), this result suggests that reviews can be

reactivated from an original movie to apply to a sequel and thus have an effect over time.

4.2. Managerial implications

From a managerial perspective, the results of this work can help studios during the

development and launching of sequels, as the decision-making of the major studios regarding

this type of movie remains a "complex equation" (Eliashberg et al., 2006).

At a strategic level, the existing reviews of the original movie are interesting signals available

to the studios. First, analysis of these reviews can help studio executives select original

movies for potential sequels that can be successfully launched onto the movie market.

Second, during the release phase, these signals can reassure risk-averse studio executives

about a sequel’s potential. While the existence of an original movie provides an overall

revenue boost for the sequel (Moon et al., 2010), the reviews associated with the original

movie will specifically facilitate the sequel’s theatrical release. Third, the reviews of both

critics and consumers will provide an indicator of the nature of the reviews that will be

generated as part of a sequel.

At an operational level, our results encourage managers to call upon these quality signals

during the release phase of the sequel. Currently, studios only have a short, one-off window to

attract a potential audience: the average run for a movie is four weeks in a context where

more films are launched each year (708 new movies were released throughout 2019,

compared to 641 in 2018

) and where development and launch costs are increasing, especially

for sequels (Bharadwaj et al., 2017). Therefore, the main objective for cinema executives is to

use as many quality signals as possible during the release phase in order to reduce information

asymmetry and ensure rapid commercial success for the sequel. Our results provide an

incentive to studios to integrate reviews into their operating strategy more effectively,

especially in promotional operations. They can reactivate these quality signals in two ways.

First, studios can directly and explicitly refer to the reviews of the original film in the

promotional campaigns at the launch of a sequel. They can be mentioned in trailers, teasers,

posters or websites dedicated to the sequel by emphasizing the number of moviegoers who

recommended the original movie (volume effect) or recalling good memories associated with

the original movie (valence effect). Studios can also use internet ad campaigns targeting

moviegoers who recommended the original film through display campaigns, for example. The

importance of these reviews can also be mentioned in the press kits that the studios provide to

Motion Picture Association Industry Report (2019)

movie critics. The use of reviews in terms of their experiential nature can lead to a positive

attitude towards the sequel in the opening weekend and reassure moviegoers who want to

avoid purchasing a "lemon" (Akerlof, 1970; Carrillat et al., 2018). Secondly, studios can

reactivate these signals indirectly by emphasizing the perceived similarity between the

original film and the sequel. The aim here is to underline the existing proximity between the

original film and the sequel, whether in the genre, the star cast, the director or the storyline.

Studios can also facilitate this reactivation of reviews by reducing the time interval between

the release of the two movies.

4.3. Limits and perspectives

This research has some limitations. It may be interesting to analyze in more depth the time

horizon associated with commercial performance. Indeed, the model developed in this

research distinguishes between short-term (revenue during the first week of movie screening)

and long-term performance (revenue after the first week of movie screening). However, given

the current speed of the emergence of reviews associated with new products (Hennig-Thurau

et al., 2015), and using the accessibility-diagnosticity framework, analyzing the sequel’s daily

revenues would allow for a more detailed evaluation of how long the quality signals (access

and relevance) associated with the original movie affect the sequel’s success. It also seems

relevant to adopt a differentiated time horizon with the reviews by distinguishing, like Berger

and Schwartz (2011), the reviews generated in the launch phase (immediate word of mouth)

from the ongoing reviews developed throughout the product’s presence on the market

(ongoing word of mouth). However, new products are likely to generate substantial reviews in

the launch phase. Since sequels are by definition a derivative work, born out of a

standardization rather than innovation approach, this distinction would make it possible to

assess both the volume of reviews of a sequel throughout its presence in theatres, and the

weight of the effect the reviews of the original movie have on those associated with the

sequel. This perspective could also make it possible to examine the volume of reviews of

critics whose weight is especially significant when starting a film (Larceneux, 2007).

Beyond a better consideration of the temporal dimension, other variables characterizing

sequels should also be integrated into the proposed model. From this point of view, this

research simplified the format by reducing the analysis to the impact of the original movie on

the first sequel. However, studios are tending to develop longer and longer formats (Filson &

Havlicek, 2018), prompting consideration of the number of sequels to assess the persistence

of these signals according to the place of the sequel in the franchise (Basuroy & Chatterjee,

2008). Similarly, as in Sood and Drèze (2006), the homogeneous nature of movies could be

integrated using indicators such as titles (descriptive or numbered, e.g., Terminator Genisys or

Fast and Furious 8) or the continued participation of star actors from one movie to another.

The degree of similarity between the movies (i.e., diagnosticity of the franchise equity) is

likely to increase the reactivation effects of the signals associated with the original movie

(Joshi & Mao, 2012). It might also be interesting to distinguish the nature of sequels by

incorporating an order perspective. Indeed, in this research, the term sequel generically covers

all works derived from an original work with no distinction of the nature of the derivative

works. However, studios deploy different script devices (i.e., prequel, sidequel, spin-off,

reboot, remake, crossover). Integrating these script formats into box office forecasting models

could also help refine the results.

Finally, the hypotheses proposed in this paper could be extended to other markets and product

categories. First, this research drew from data from the U.S. market to test the box office

forecasting model of a sequel. It might be interesting to assess the validity of the model in

other markets, as the taste for this type of film product may vary depending on the cultural

context (Akdeniz & Talay, 2013). Secondly, this research was limited to the analysis of the

theatrical movie market. However, there are other experiential products where companies

minimize the risks of a new launch by capitalizing on the success of an original product. From

this point of view, strategies in the publishing, video game or music industries could be

similar to those implemented in the movie industry. It would therefore be interesting to

examine the role played by quality signals associated with products of the same brand in these

sectors. Similarly, it might be interesting to assess the role of these signals in the creation of a

product providing a similar experience (film, video game, comic book, book) on multiple

media platforms.

Appendix – Movie sample

• Harry Potter

• The Lord of the Rings

• Spider-Man

• Shrek

• Twilight

• Transformers

• X-Men

• The Fast and the Furious

• Pirates of the Caribbean

• The Hunger Games

• Iron Man

• The Hobbit

• Despicable Me

• Ice Age

• Madagascar

• The Hangover

• The Bourne Identity

• Meet the Parents

• The Matrix

• Alvin and the Chipmunks

• Monsters, Inc.

• Night at the Museum

• The Chronicles of Narnia

• Scary Movie

• The Mummy

• Captain America

• Cars

• Ocean's Eleven

• Saw

• American Pie

• Sherlock Holmes

• How to Train Your Dragon

• National Treasure

• Thor

• Kung Fu Panda

• Taken

• Diary of a Mad Black

Woman

• The Da Vinci Code

• Spy Kids

• Bruce Almighty

• Analyze This

• 21 Jump Street

• 300

• Grown Ups

• Fantastic Four

• Rio

• Hulk

• Final Destination

• Sex and the City

• Clash of the Titans

• Cloudy with a Chance of

Meatballs

• Resident Evil

• Scooby-Doo

• The Expendables

• Charlie's Angels

• Big Momma's House

• Underworld

• The Conjuring

• Cheaper by the Dozen

• Step Up

• The Smurfs

• Anchorman

• Journey to the Center of

the Earth

• The Ring

• Stuart Little

• The Princess Diaries

• Lara Croft

• Insidious

• Legally Blonde

• Horrible Bosses

• xXx

• Ghost Rider

• Diary of a Wimpy Kid

• Percy Jackson

• Think Like a Man

• Miss Congeniality

• The Grudge

• Red

• Barbershop

• Pitch Black

• Cats & Dogs

• Kill Bill

• The Purge

• Hellboy

• Are We There Yet?

• Shanghai Noon

• Daddy Day Care

• Dolphin Tale

• Bridget Jones's Diary

• The Best Man

• Garfield

• The Transporter

• Sin City

• Deuce Bigalow

• The Sisterhood of the

Traveling Pants

• Kick-Ass

• Nanny McPhee

• 28 Days Later...

• The Whole Nine Yards

• Jeepers Creepers

• Agent Cody Banks

• Hostel

• Silent Hill

• The Hills Have Eyes

• Hoodwinked

• A Haunted House

• Harold & Kumar

• The Last Exorcism

• Crank

• Jonah: A VeggieTales

Movie

• Johnny English

• House of 1,000 Corpses

• The Collector

• G.I. Joe

• The Tigger Movie

• Atlas Shrugged

Table 1 – Sequel’s box office performance: literature synthesis

Authors

Exogenous variables

Endogenous

variable: movie

revenues

Movie sample

Performance of sequels

Demand

Supply

Ravid (1999)

Critics’ reviews (volume

and valence)

Production budget, MPAA ratings,

Presence of stars, Presence of

participants who were nominated for or

received awards, Film’s release date

Domestic,

international and

video revenues

175 movies (1991

– 1993)

Sequels are considered as a signal. 11 films are sequels in

the database. The sequel variable positively affects the box

office performance.

Basuroy et al.

(2003)

Critics’ reviews (volume

and valence)

Production budget, Presence of stars

(actors and directors), Screens, MPAA

ratings, Film’s release date

Weekly revenues

175 movies (1991

- 1993)

Sequels are considered as a control variable. 11 films are

sequels in the database. Models are not affected by the

presence of sequels.

Basuroy et al.

(2006)

Critics’ reviews (valence

and homogeneity),

Consumer reviews

Advertising expenditures, Screens,

Major distributors, Competition,

Expected performance, Presence of

stars, Sequel

Revenues (opening

week, second week

and beyond)

175 movies (1991

- 1993)

Sequels are considered as a signal. 11 films are sequels in

the database. Sequel and ad expenditure have a positive

interaction impact on the box office revenues.

Hennig-

Thurau et al.

(2006)

Critics’ reviews, Consumer

reviews

Budget, Presence of stars, Cultural

familiarity (sequel, remake, adaptation),

Advertising expenditures, Screens,

Production budget

Revenues (opening

week, beyond the

opening week)

331 movies (1999

- 2001)

Sequels are considered as a studios’ action. Number of

sequels in the database is not specified. Sequels positively

affect the box office revenues during the opening week.

Liu (2006)

Critics’ reviews (volume

and valence), Consumer

reviews (volume and

valence, pre- and post-

release)

Production budget, MPAA ratings,

Genres, Presence of stars, Screens (first

week), Movie’s life (in weeks)

Weekly revenues

40 movies (2002)

Sequels are integrated in preliminary analysis. They have

no significant effect on the aggregate box office revenues.

Boatwright et

al. (2007)

Critics’ reviews (volume

and valence)

Production budget, Communication

budget, Presence of stars, Screens,

Movie’s appeal

Revenues (opening

week), Decay rate

of sales

466 movies (1997

– 2001)

Sequels are a characteristic of the films. 8% of a sub-sample

(317 movies) are sequels. Sequels have the greater market

potential.

Basuroy &

Chatterjee

(2008)

Critics’ reviews (volume

and valence)

Screens, Movie characteristics (MPAA

ratings, Director or actor awards,

Production budget), Seasonality

Domestic revenues

(for up to the

fifteenth week)

167 movies

(1991- 1993)

Sequels are considered as a brand extension. 11 films are

sequels in the database. The sequel's box office performance

is weaker than the original movie's performance but higher

than the contemporaneous non-sequels' performance.

Moon et al.

(2010)

Critics’ reviews (volume

and valence), Consumer

reviews (volume and

valence)

Movie characteristics (genre, sequel,

MPAA ratings, distribution, films life,

video release lagging days), Movie

costs (production budget, advertising

expenditure, screens)

Revenues (box

office, video rental,

video sales)

246 movies (2003

– 2005)

Sequels are considered as a brand extension. 36 films are

sequels in the database. Sequels have a positive impact on

revenues only in the first two weeks but receive lower

ratings than originals.

Karniouchina

(2011)

Consumer reviews (volume

and valence, pre- and post-

release, movie and star)

Movie characteristics (advertising

budget, sequel, nationality), Stars’

characteristics (bankability, recency of

success, academy awards, sex appeal)

Revenues (opening

week, second week

and beyond)

140 movies

(2005)

Sequels are considered as a control variable. Number of

sequels in the database is not specified. The volume of

reviews is higher for sequels.

Dhar et al.

(2012)

Consumer reviews

(valence)

Movie characteristics (MPAA ratings,

genre, run time), Seasonality (month,

holiday)

Revenues (opening

week, total)

1,990 movies

(1983-2008)

Sequels represent 6,68% of the database. Sequels have a

better box office performance than non-sequels but they

generate less attendance than original movies.

Hennig-

Thurau et al.

(2012)

Critics’ reviews (valence),

Consumer reviews

(valence)

Movie characteristics (star power,

sequel, MPAA ratings, genre), Target

audiences, Studios’ actions (advertising

expenditures, release strategy), Buzz

(prerelease)

Revenues (first

week, after the first

week)

1,370 movies

(1998-2006)

Sequels are a characteristic of the films. Number of sequels

in the database is not specified. The influence of reviews on

the box office performance is weaker for a sequel than for a

contemporaneous non-sequel.

Akdeniz &

Talay (2013)

Critics’ reviews (valence)

Movie characteristics (production

budget, sequel, star power, genre,

Revenues (first

week)

1,116 movies

(2007-2011)

103 films are sequels. The positive relationship between

sequels and box office performance persists internationally.

awards), Screens, Advertising

expenditures, Characteristic of the

distributor (major), Seasonality,

Competing movies (including local

movies)

However, this relationship varies by culture (e.g., the

relationship wanes in individualist cultures versus

collectivist cultures).

Bharadwaj et

al. (2017)

Critics’ reviews (volume,

valence, writing style)

Sequel, Genre, Production budget,

Advertising budget, MPAA ratings,

Director power, Star power, Level of

competition, Studio, Release

Domestic revenues

(total)

115 movies

(2009-2011)

Number of sequels in the database is not specified. Sequels

– considered as a prelaunch marketing decision – have a

significant impact on box office performance.

Filson &

Havlicek

(2018)

Critics’ reviews (valence),

Consumer reviews

(valence)

Production budget, Advertising

expenditures, Genre, MPAA ratings,

Screens, Awards, Change in the

franchise (ratings, lead actor, director)

Revenues (domestic

and foreign

markets) and ROI

(revenues / budget)

433 movies (prior

to 2014)

The box office performance of franchises deteriorates as

installments are introduced. Part of the box office revenues

attributable to foreign markets rises as further installments

are introduced.

Table 2 – Summary of the data sample

Variable

Characteristics

Mean

Standard Deviation

Original

movie

Sequel

Original

movie

Sequel

Box office performance

Revenues (M$) during the first

week (adjusted for inflation and for

seasonal movements)

35.84

37.31

24.55

32.13

Revenues (M$) after the first week

(adjusted for inflation and for

seasonal movements)

95.65

78.32

67.48

74.34

Characteristics of

the movie

Country of origin

Binary (USA versus non USA)

89% (USA)

93% (USA)

Producer status

Binary (Major versus Independent)

88% (Major)

MPAA ratings

Binary (R versus PG, PG-13 and G)

30% (R)

28% (R)

Genre

Adventure / Action

Animation

Drama / Thriller

Comedy

53%

15%

12%

20%

53%

15%

12%

20%

Director

Success: cumulated revenues before

the movie release (M$)

276.56

335.87

429.99

442.60

Stars

Success: cumulated revenues of the

two main stars of the cast before the

movie release (M$)

1496.59

2203.58

1535.68

1727.67

Studio’s actions

Production budget

M$ (adjusted for inflation)

57.47

73.11

47.57

58.40

Screens

Number of screens (release)

2945.21

3233.01

746.34

691.82

Reviews

Critics’ reviews

Valence: rating (average score of 30

professional reviews) – metascore

55.4

48.4

14.66

15.70

Consumer

reviews

Valence: average score – IMDb

First week

15 weeks

Volume: number of ratings – IMDb

First week

15 weeks

6.70

605.10

6.44

116.84

0.99

572.25

1.40

176.58

Table 3 – Model of the sequel's box office performance: intra-movie perspective

First week

After the first week

Endogenous

variables

Exogenous variables

Budget

Screens

Revenue

Budget

Screens

Revenue

Production budget

Number of screens

Country of origin

Producer status

Genre (Adventure / Action)

Genre (Animation)

Genre (Drama / Thriller)

MPAA ratings (R)

Success of the director

Success of the stars

Critics’ reviews (valence)

Consumer reviews (valence)

Consumer reviews (volume)

0.021

0.089

0.273***

0.331***

0.121

0.186***

0.169***

-0.122

0.203***

-0.079

-0.026

0.007

0.225***

0.054

0.070

0.118*

-0.034

0.251***

1.695***

0.211

-0.053

0.052

0.047

0.064

-0.028

0.030

-0.009

0.584***

0.026

0.093

0.274***

0.323***

0.096

0.192***

0.181**

-0.123

0.211***

-0.075

-0.026

0.006

0.221***

0.089

0.075

0.117*

-0.026

0.254***

1.365***

0.105

0.018

-0.015

0.220

0.025

0.248***

0.033

-0.012

0.474

0.441***

0.420**

Adjusted R

0.662

0.654

0.633

0.662

0.654

0.772

*** significant at the 1% level, ** at the 5% level, * at the 10% level

Table 4 – Direct and indirect effects of the original movie’s reviews on the sequel’s box office

performance through the sequel’s reviews

Reviews

original

movie

Box office

performance

sequel

Direct effect

Indirect effect

Total

effect

a x b + c

Reviews

original movie

→ Box office

performance

sequel

(c)

Reviews

original movie

→ Reviews

sequel

(a)

Reviews

sequel

→

Box office

performance

sequel

(b)

Indirect effect

(a x b)

Consumer

reviews

(valence)

First week

0.458*

After the first

week

0.066

0.484***

0.180*

0.087*

0.153*

Consumer

reviews

(volume)

First week

0.129***

After the first

week

-0.183

0.837***

0.409***

0.343**

0.159*

Critics’

reviews

First week

-0.018

0.701**

0.011

0.008

-0.010

After the first

week

0.092

0.701**

-0.043

-0.030

0.062

*** significant at the 1% level, ** at the 5% level, * at the 10% level

The model is just-identified. Fit indices are not significant (RMSEA = 0; CFI = 1; TLI = 1).

The total variance of Y explained by X is based on a direct effect c (X➔Y) and an indirect effect a x b (a X ➔

mediator and b mediator ➔ Y).

Table 5 – Moderating effect of the time interval on the relation between original movie’s

reviews and sequel’s performance

First week

Consumer reviews

Valence

-0.009

Volume

-0.005*

Critics’ reviews

Valence

0.001

*** significant at the 1% level, ** at the 5% level, * at the 10% level

Table 6 – Results summary

Only the interaction coefficient is reported.

Figure 1 – Conceptual model

Hypothesis

Expected result

Empirical result

H1a

Reviews (vol. consumers)

original movie

➔ Short-term box office

sequel

Effect: +

Supported

Reviews (vol. consumers)

original movie

➔ Long-term box office

sequel

No effect

Supported

Reviews (val. consumers)

original movie

➔ Short-term box office

sequel

Effect: +

Supported

Reviews (val. consumers)

original movie

➔ Long-term box office

sequel

No effect

Supported

H1b

Reviews (val. critics)

original movie

➔ Short-term box office

sequel

Effect: +

Not supported

Reviews (val. critics)

original movie

➔ Long-term box office

sequel

No effect

Supported

Time interval ➔ [Reviews (vol. consumers)

original movie

➔ Short-term box office

sequel

]

Moderation effect: +

Supported

Time interval ➔ [Reviews (val. consumers)

original movie

➔ Short-term box office

sequel

]

Moderation effect: +

Not supported

Time interval ➔ [Reviews (val. critics)

original movie

➔ Short-term box office

sequel

]

Moderation effect: +

Not supported

H3a

Reviews (vol. consumers)

original movie

➔ Reviews (vol. consumers)

sequel

➔ Long-term

box office

sequel

Effect: +

Supported

Reviews (val. consumers)

original movie

➔ Reviews (val. consumers)

sequel

➔ Long-term

box office

sequel

Effect: +

Supported

H3b

Reviews (val. critics)

original movie

➔ Reviews (val. critics)

sequel

➔ Long-term box

office

sequel

Effect: +

Not supported

References

Aaker, D. A., & Keller, K. L. (1990). Consumer evaluations of brand extensions. Journal of

Marketing, 54(1), 27-41. https://doi.org/10.1177%2F002224299005400102

Akdeniz, M. B., & Talay, M. B. (2013). Cultural variations in the use of marketing signals: A

multilevel analysis of the motion picture industry. Journal of the Academy of Marketing

Science, 41(5), 601-624. https://doi.org/10.1007/s11747-013-0338-5

Akerlof, G. A. (1970). The market for 'lemons': quality uncertainty and the market

mechanism. Quarterly Journal of Economics, 84(3), 488-500.

https://doi.org/10.2307/1879431

Basuroy, S., & Chatterjee, S. (2008). Fast and frequent: investigating box office revenues of

motion picture sequels. Journal of Business Research, 61(7), 798-803.

https://doi.org/10.1016/j.jbusres.2007.07.030

Basuroy, S., Chatterjee, S., & Ravid, S. A. (2003). How critical are critical reviews? The box

office effects of film critics, star power, and budgets. Journal of Marketing, 67(4), 103-

117. https://doi.org/10.1509%2Fjmkg.67.4.103.18692

Basuroy, S., Desai, K. K., & Takuldar, D. (2006). An empirical investigation of signaling in

the motion picture industry. Journal of Marketing Research, 43(2), 287-295.

https://doi.org/10.1509%2Fjmkr.43.2.287

Bharadwaj, N., Noble, C. H., Tower, A., Smith, L. M., & Dong, Y. (2017). Predicting

innovation success in the motion picture industry: The influence of multiple quality

signals. Journal of Product Innovation Management, 34(5), 659-680.

https://doi.org/10.1111/jpim.12404

Berger, J., & Schwartz, E. M. (2011). What drives immediate and ongoing word of mouth?.

Journal of Marketing Research, 48(5), 869-880. https://doi.org/10.1509%2Fjmkr.48.5.869

Boatwright, P., Basuroy, S., & Kamakura, W. (2007). Reviewing the reviewers: The impact

of individual film critics on box office performance. Quantitative Marketing and

Economics, 5(4), 401-425. https://doi.org/10.1007/s11129-007-9029-1

Carrillat, F. A., Legoux, R., & Hadida, A. L. (2018). Debates and assumptions about motion

picture performance: A meta-analysis. Journal of the Academy of Marketing

Science, 46(2), 273-299. https://doi.org/10.1007/s11747-017-0561-6

Chakravarty, A., Liu, Y., & Mazumdar, T. (2010). The differential effects of online word-of-

mouth and critics’ reviews on pre-release movie evaluation. Journal of Interactive

Marketing, 24(3), 185-197. https://doi.org/10.1016/j.intmar.2010.04.001

Chen, Y., Liu, Y., & Zhang, J. (2012). When do third-party product reviews affect firm value

and what can firms do? The case of media critics and professional movie reviews. Journal

of Marketing, 76(2), 116-134. https://doi.org/10.1509%2Fjm.09.0034

Chisholm, D. C., Fernández-Blanco, V., Ravid, S. A., & Walls, W. D. (2015). Economics of

motion pictures: the state of the art. Journal of Cultural Economics, 39(1), 1-13.

https://doi.org/10.1007/s10824-014-9234-1

Dellarocas, C., Gao, G., & Narayan, R. (2010). Are consumers more likely to contribute

online reviews for hit or niche products? Journal of Management Information Systems,

27(2), 127-158. https://doi.org/10.2753/MIS0742-1222270204

Dellarocas, C., Zhang, X. M., & Awad, N. F. (2007). Exploring the value of online product

reviews in forecasting sales: The case of motion pictures. Journal of Interactive Marketing,

21(4), 23-45. https://doi.org/10.1002/dir.20087

Desai, K. K., & Basuroy, S. (2005). Interactive influence of genre familiarity, star power, and

critics’ reviews in the cultural goods industry: The case of motion pictures. Psychology &

Marketing, 22(3), 203-223. https://doi.org/10.1002/mar.20055

Dhar, T., Sun, G., & Weinberg, C. B. (2012). The long-term box office performance of sequel

movies. Marketing Letters, 23(1), 13-29. https://doi.org/10.1007/s11002-011-9146-1

Duan, W., Gu, B., & Whinston, A. B. (2008). The dynamics of online word-of-mouth and

product sales – an empirical investigation of the movie industry. Journal of Retailing,

84(2), 233-242. https://doi.org/10.1016/j.jretai.2008.04.005

Elberse, A., & Eliashberg, J. (2003). Demand and supply dynamics for sequentially released

products in international markets: The case of motion pictures. Marketing Science, 22(3),

329-354. https://doi.org/10.1287/mksc.22.3.329.17740

Eliashberg, J., Elberse, A., & Leenders, M. A. (2006). The motion picture industry: Critical

issues in practice, current research, and new research directions. Marketing Science, 25(6),

638-661. https://doi.org/10.1287/mksc.1050.0177

Feldman, J. M., & Lynch, J. G. (1988). Self-generated validity and other effects of

measurement on belief, attitude, intention, and behavior. Journal of Applied

Psychology, 73(3), 421-435. https://psycnet.apa.org/doi/10.1037/0021-9010.73.3.421

Filson, D., & Havlicek, J. (2018). The performance of global film franchises: installment

effects and extension decisions. Journal of Cultural Economics, 42(3), 447-467.

https://doi.org/10.1007/s10824-018-9316-6

Godes, D. & Mayzlin, D. (2004). Using online conversations to study word-of-mouth

communication. Marketing Science, 23(4), 545-560.

https://doi.org/10.1287/mksc.1040.0071

Godes, D., & Silva, J. C. (2012). Sequential and temporal dynamics of online opinion.

Marketing Science, 31(3), 448-473. https://doi.org/10.1287/mksc.1110.0653

Hadida, A. L. (2009). Motion picture performance: A review and research agenda.

International Journal of Management Reviews, 11(3), 297-335.

https://doi.org/10.1111/j.1468-2370.2008.00240.x

Hennig-Thurau, T., Houston, M. B., & Heitjans, T. (2009). Conceptualizing and measuring

the monetary value of brand extensions: the case of motion pictures. Journal of Marketing,

73(6), 167-183. https://doi.org/10.1509%2Fjmkg.73.6.167

Hennig-Thurau, T., Houston, M. B., & Sridhar, S. (2006). Can good marketing carry a bad

product? Evidence from the motion picture industry. Marketing Letters, 17(3), 205-219.

https://doi.org/10.1007/s11002-006-7416-0

Hennig-Thurau, T., Marchand, A., & Hiller, B. (2012). The relationship between reviewer

judgments and motion picture success: re-analysis and extension. Journal of Cultural

Economics, 36(3), 249-283. https://doi.org/10.1007/s10824-012-9172-8

Hennig-Thurau, T., Wiertz, C., & Feldhaus, F. (2015). Does Twitter matter? The impact of

microblogging word of mouth on consumers’ adoption of new movies. Journal of the

Academy of Marketing Science, 43(3), 375-394. https://doi.org/10.1007/s11747-014-0388-

Holbrook, M. B. (2005). The role of ordinary evaluations in the market for popular culture:

Do consumers have “good taste”? Marketing Letters, 16(2), 75-86.

https://doi.org/10.1007/s11002-005-2774-6

Joshi, A., & Mao, H. (2012). Adapting to succeed? Leveraging the brand equity of best sellers

to succeed at the box office. Journal of the Academy of Marketing Science, 40(4), 558-571.

https://doi.org/10.1007/s11747-010-0241-2

Karniouchina, E. V. (2011). Impact of star and movie buzz on motion picture distribution and

box office revenue. International Journal of Research in Marketing, 28(1), 62-74.

https://doi.org/10.1016/j.ijresmar.2010.10.001

Knapp, A. K., Hennig-Thurau, T., & Mathys, J. (2014). The importance of reciprocal

spillover effects for the valuation of bestseller brands: introducing and testing a

contingency model. Journal of the Academy of Marketing Science, 42(2), 205-221.

https://doi.org/10.1007/s11747-013-0350-9

King, R. A., Racherla, P., & Bush, V. D. (2014). What we know and don’t know about online

word-of-mouth: a review and synthesis of the literature. Journal of Interactive Marketing,

28(3), 167-183. https://doi.org/10.1016/j.intmar.2014.02.001

Larceneux, F. (2007). Buzz and recommendations on the Internet. What impacts on box-

office success? Recherche et Applications en Marketing, 22(3), 45-64.

https://doi.org/10.1177%2F205157070702200304

Lazear, E. P. (2004). The Peter Principle: A theory of decline. Journal of Political

Economy, 112(S1), 141-163. https://doi.org/10.1086/379943

Lee, Y. J., Hosanagar, K., & Tan, Y. (2015). Do I Follow My Friends or the Crowd?

Information Cascades in Online Movie Ratings. Management Science, 61(9), 2241-2258.

https://doi.org/10.1287/mnsc.2014.2082

Liu, Y. (2006). Word of mouth for movies: its dynamics and impact on box office revenue.

Journal of Marketing, 70(3), 74-89. https://doi.org/10.1509%2Fjmkg.70.3.074

Moon, S., Bergey, P. K., & Iacobucci, D. (2010). Dynamic effects among movie ratings,

movie revenues, and viewer satisfaction. Journal of Marketing, 74(1), 108-121.

https://doi.org/10.1509%2Fjmkg.74.1.108

Nelson, P. (1970). Information and consumer behavior. Journal of Political Economy, 78(2),

311-329. https://doi.org/10.1086/259630

Prieto-Rodriguez, J., Gutierrez-Navratil, F., & Ateca-Amestoy, V. (2015). Theatre allocation

as a distributor’s strategic variable over movie runs. Journal of Cultural Economics, 39(1),

65-83. https://doi.org/10.1007/s10824-014-9220-7

Ravid, S. A. (1999). Information, blockbusters, and stars: A study of the film industry. The

Journal of Business, 72(4), 463-492. https://doi.org/10.1086/209624

Rein, I. J., Kotler, P., & Stoller, M. R. (1987). High visibility. UK, Heinemann Professional

Publishing Ltd.

Sood, S., & Drèze, X. (2006). Brand extensions of experiential goods: Movie sequel

evaluations. Journal of Consumer Research, 33(3), 352-360.

https://doi.org/10.1086/508520

Zhao, X., Lynch Jr, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and

truths about mediation analysis. Journal of Consumer Research, 37(2), 197-206.

https://doi.org/10.1086/651257

Zufryden, F. S. (1996). Linking advertising to box office performance of new film releases-a

marketing planning model. Journal of Advertising Research, 36(4), 29-42.