ChapterPDF Available

The Financial and Economic Risks of Film Production



Content may be subject to copyright.
The financial and economic risks of film production
Michael Pokorny and John Sedgwick
It would generally be accepted as self-evident that the film industry provides a very
unstable environment for the development of a coherent investment strategy. Indeed,
realistic investors in film production should expect their returns to be in the form of
reflected glamour and kudos, rather than in strictly defined balance sheet profits. Even
the hard-headed investor would allocate only a small proportion of his investment funds
to film production, ensuring that the remainder of his investments generate sufficient
profits to compensate for the film losses that are likely to be generated.
For the one brutal constant in the film industry, from its very beginnings in the late 19
Century, is that only a relatively small proportion of film outputs ever make profits.
Even in the case of Hollywood, the world’s consistently dominant film producer from the
1920s, only about a half of its films have been profitable. And in the case of these
profitable films the vast majority have only been modestly profitable, with just a handful
of films, annually, generating significant profits. Indeed, part of Hollywood’s success
derives from the perpetuation of the myth that these handfuls of ‘hit’ films represent the
norm rather than, in fact, being the very rare exceptions.
In pure investment terms the low incidence of hit films would not be a problem, if there
was a basis or methodology for predicting with any level of accuracy which of the large
numbers of films released annually will turn out to be hits. The problem is that such a
methodology would not appear to exist – in essence, it is impossible to predict, on a
consistent and regular basis, which of the hundreds of annual film releases will turn out to
be profitable, the process being to all intents and purposes an apparently random one. It
is this empirical regularity that has been pithily captured in screenwriter William
Goldman’s throwaway line about Hollywood that ‘nobody knows anything’, an
observation that has now become part of film industry ‘wisdom’.
To be sure, there are
many who argue that they have particular insights into the complex process of profitable
film production, particularly those film producers who at any point in time are seeking
investment funds for their next productions. But in the same way that there are those
who argue that they have particular insights in the manner in which stock market prices
evolve over time, the empirical support for such claims are weak if not non-existent. For
the paradox is that while film investment is inherently risky, Hollywood is a manifestly
profitable industry and has been so for almost a century, albeit within the context of
marked profitability cycles. Many of the major studios/distributors that make up
Hollywood have been there from its establishment, although these studios have
experienced regular changes in ownership and structure.
Our objective in this chapter is to present a detailed description of the risk environment of
film production, within the context of Hollywood, and to attempt to identify the various
influences on this environment. In turn, we will draw conclusions about the nature of the
strategies that might be employed to attenuate the risks of film production, and the
implications that these have for the evolution and structure of the industry. Despite its
claims to the contrary, the risk environment in which Hollywood operates is not unique,
and shares many of the characteristics of other industries in which only a small
proportion of outputs are profitable, but sufficiently so to more than compensate for the
loss generating outputs. Examples of such industries are book publishing, music
recording and pharmaceutical manufacturing. What these industries appear to share are
research, development and production processes that are characterised by a seemingly
blind and extensive search for success, and where failure is not so much an unfortunate
consequence of this process but integral to it.
The risk environment of film production
There are two main datasets that will be drawn on here. The first derives from the films
released onto the North American market between 1929 and 1942 by three of the five
Hollywood major studio/distributors at the time, Metro–Goldwyn–Mayer (MGM), Radio-
Keith-Orpheum (RKO) and Warner Bros (WB). This dataset consists of 1,796 films, for
each of which data are available on production costs, revenues received by the
studio/distributor, and in the case of MGM and RKO, the profits that each of these films
generated. In the case of WB we had to apply a methodology for estimating its film
The second dataset derives from the 4,164 films released onto the North American
market between 1988 and 1999. The dataset consists of the North American box-office
revenues generated by these films, and estimates of the production costs of just over half
of these films. This dataset does not contain any profitability data, and we had to
estimate film profits for the films for which production cost estimates were available.
The resulting dataset consists of 2,116 films, of which 1,458 were distributed by the
major studios. Note, that in order to make the profitability data from the 1990s
comparable with those of the 1930s the 1990s data relate to profits generated only from
theatrical release and therefore do not include profits generated from all ancillary
Figure 1 presents a scattergraph of film profits generated in the North American market
against film production costs for the 1930s, and Figure 2 presents the equivalent graph
for the 1990s. For contextual purposes the titles of a number of the films produced
during both periods are also included.
The outstanding and common feature of these two graphs is that the variability of film
profits increases as production budgets increase, reflecting the increasing levels of risk
that are associated with high budget film production. The obvious attraction of high
budget production is that it offers the possibility of a single film generating very high
profits. However, as is also clear from Figures 1 and 2, high budget films can also
generate substantial losses, and even when profitable, generate profits that are often
comparable to those generated by much more modestly budgeted films. However, there
are a number of more subtle differences between Figures 1 and 2. By and large, high
budget film production was relatively unsuccessful in the 1930s, whereas it was the main
source of profits during the 1990s. Thus the rate of return on high budget production
during the 1930s was less than that on lower budget production, whereas during the
1990s the profitability of high budget production exceeded that of lower budget
production. It is not simply that Hollywood became better at producing high budget
films over time (which it did) but Hollywood recognised that the successful production of
high budget films was essential to its survival in the post-World War II period, when the
advent of television and other demographic and life style changes resulted in its near
collapse. It was only by producing high budget, ‘event’ movies – what today would be
termed ‘blockbusters’ – that Hollywood could hope to differentiate its output from that of
television and attract audiences back to moviegoing. Starting with developments such as
suburban multiplex cinemas in the 1960s Hollywood developed increasingly
sophisticated strategies for maximising film revenues, and with the continual
development and diffusion of a range of movie consumption technologies the revenues
that can be generated by successful movies is now virtually limitless. Such an
environment stands in stark contrast to that of the 1930s when theatrical revenues were
the only source of film revenues and, at best, a successful movie had a life of only 15
months, with most movies being on release for a considerably shorter time than this.
An alternative means of presenting the profitability data in Figures 1 and 2 is via
frequency distributions – a histogram representation of the data that provides a more
explicit reflection of the manner in which profits and losses are spread across the
profitability range. These frequency distributions are shown in Figure 3 (for the 1930s)
and Figure 4 (for the 1990s). Again, the similarities between these distributions are clear,
and in particular, the long right hand tails that both of these distributions exhibit. That is,
it is the films that appear in this ‘long tail’ – the blockbusters – that dominate the profit
distribution, and generate profits that can compensate for the losses made by the large
number of films falling in the shorter left hand tail.
Clearly such blockbusters are
relatively rare, but are crucial for the financial success of the industry, and provide the
defining characteristic of the industry.
In order to provide a more detailed interpretation of these long tails, consider the films
which generated profits in excess of $0.5m (1929 prices) during the 1930s, and those
films which generated profits in excess of $20m (1987 prices) in the 1990s. During the
1930s there were 90 such films – 5 per cent of the films produced – and in the 1990s
there were 52 such films – 3 per cent of the films produced. The films that fall into these
long tails can be seen directly from Figures 1 and 2. In the case of the 1930s – Figure 1 –
a horizontal line can be drawn at 0.5 on the vertical axis, and all the films lying above this
line are those that fall into this long tail. Thus the film lying the furthest to the right in
Figure 3 can be seen to be Mrs Miniver, having generated US profits of $3.1m. In the
case of the 1990s – Figure 2 – a horizontal line can be drawn at 20 on the vertical axis, all
the films above this line defining the long tail, with Titanic at the extreme right hand
point of this tail, having generated estimated US profits from theatrical release of $90.2m.
We can compare and contrast these long tails in more detail. However, in order to
compare like with like, we will consider only the films distributed by the majors. Clearly
these are all the 1,796 films in the 1930s data set, and for the 1990s 1,458 of the 2,116
films for which profits could be estimated fall into this category. Thus consider the top
10 per cent most profitable films during both periods, a somewhat more extensive
definition of the long tail than used above. For the 1930s these 180 films generated 84%
of the profits generated by all films, and for the 1990s the 146 most profitable films
generated 85% of the profits generated by all the films produced by the majors. So in this
regard, the contribution of the long tails was remarkably similar between the two periods.
However, there are marked differences in the nature of these top films between the two
periods. Thus consider the incidence of high budget films amongst the top profit earners
– the extent to which strategic decisions on the part of film producers were translated into
profitable films, on the assumption that a high level of investment in a film is a direct
reflection of the producer’s confidence in the quality and appeal of the film. We will
here define a high budget film as a film costing more than twice the average cost of all
films produced in the film’s year of release. According to this definition 179 high budget
films were produced in the 1930s, 10 per cent of all films produced, absorbing 30 per
cent of production costs, and in the 1990s, 247 of these films were produced, or 17 per
cent of the total, absorbing 39 per cent of production costs, thus confirming the greater
extent of high budget production in the latter period. In the 1930s 46 of these high
budget films – 26 per cent of the high budget films produced – were located in the long
tail, with a similar proportion lying in the long tail in the 1990s – 68 high budget films, or
28 per cent of the high budget films produced. But note also that these long tail high
budget films accounted for 26 per cent of the long tail films in the 1930s, but 46 per cent
of the long tail films in the 1990s. However, the main difference between the two periods
is the contribution to overall profits made by these long tail high budget films. In the
1930s, the 46 long tail high budget films accounted for only 25 per cent of aggregate
profits whereas in the 1990s the 68 long tail films accounted for 42 per cent of aggregate
profits. In terms of all high budget films produced, the 179 produced in the 1930s
generated just 17 per cent of profits (on 30 per cent of costs, with 44 per cent of these
films making losses), whereas the 247 produced in the 1990s generated 45 per cent of
profits (on 39 per cent of costs, and 33 per cent making losses). That is, the focus of
profitability in the 1930s was on lower budget production, whereas the profitability of the
contemporary industry derives from higher budget production.
Clearly, the film producer’s objective is to locate his film in this long tail, and thereby
derive the disproportionate returns that will be generated by so doing. However, as is
also clear from Figures 3 and 4, the likelihood of doing so is minimal. Indeed, the clear
implication of Figures 1 to 4 is that that any film production strategy based on the success
of single, one-off film projects is doomed to failure. Rather, a more sensible strategy for
a rational, profit-maximising film producer is to produce a wide range of films, annually,
in the hope that at least some of these will produce profits that will compensate for the
losses that a large proportion of these films will inevitably generate. That is, we could
characterise the successful film studios/distributors as constructing diversified annual
portfolios of films, diversified according to production budget and genre, and allocations
of stars, directors and screenwriters. The issue then is not so much which of the films in
the portfolio are profitable, but simply that the portfolio itself is profitable.
However, this is not to suggest that such a strategy can now render film production
riskless – at best, such a strategy can exert some control over the risk of film production,
but certainly not eliminate it. But what it does mean is that such a profitable film
production strategy requires considerable financial resources so that a fully diversified
film portfolio can be constructed – the successful film producer must have the capacity to
absorb significant losses in addition to enjoying the fruits of the profitable films.
It is therefore not surprising that the consistently successful film studios/distributors are
large organisations, that themselves are often parts of even larger media/entertainment
conglomerates, allowing for further opportunities for risk spreading. Conversely,
relatively small independent studios/distributors will not have the ability to develop
sufficiently diversified film portfolios, nor have the financial strength to absorb the losses
that any film portfolio will inevitably generate.
To this point our objective has been to describe the nature of the financial risk
environment within which film producers operate. However, we have not made any
attempt to identify the factors that influence or determine this risk environment. That is,
why is it the case that even highly experienced film producers cannot predict with any
reliability how consumers will react to their films, and hence which of their films will
make profits and which will generate losses? Our answer to this question derives from
the recognition that consumers themselves enter a risk environment when choosing to
view a movie, in the sense that while the consumer will have certain expectations prior to
viewing a given movie, there will often be a divergence between actual experience and
expectation. Indeed, it is the possibility of the actual viewing experience markedly
exceeding the expectations of that experience that provides the essential stimulus to
continued movie consumption. We will now go on to discuss in more detail the nature of
the risk environment from the film consumer’s perspective.
The risk environment of film consumption
One of the essential characteristics of films is that they are ‘experience’ goods: audiences
can only form a full assessment of the product when the act of consumption is complete.
The standard model developed by Nelson (1970, 1974) depicts experience goods as a
category of good, the quality of which can be established only after it has been consumed
(‘experienced’) by the buyer. Thus consumers identify the ‘best’ brand of a given good
by repeatedly consuming the range of brands that are available, and on the basis of the
experience so derived, decide upon the preferred brand. In consuming a new brand the
consumer will form expectations concerning this brand derived from the past
consumption of competing brands, together with any information provided by
advertising/promotion. But it is only once the act of consumption is complete that a full
assessment of the new brand can be made, within the context of the extent to which
expectations were disappointed, met or exceeded. Even if consumers identify a clearly
preferred brand, they may still continue to sample experience competing brands, if for no
other reason than to confirm their original choice.
Thus films clearly fall into the category of an experience good – irrespective of the
information available to the consumer prior to viewing the film, in the form of word-of-
mouth and promotional/advertising messages, such information will be an imperfect
substitute for the experience derived from the act of consumption. However, in contrast
to most other experience goods, films are generally consumed only once, as the key
characteristics that filmgoers seek is novelty – they seek ‘surprises’ and innovation in the
filmgoing experience, and in particular, do not seek exact replications of previous
filmgoing experiences.
Thus in choosing to view a given film the film consumer will have formed expectations
of the pleasures that are likely to be generated by the experience. These expectations will
derive both from the consumer’s previous film consumption experiences and the
marketing/promotional activities surrounding the film. However the consumer will also
be aware that these expectations are likely to be confounded, in either a positive or
negative sense. Indeed, the definition of a hit film is a film that generates, amongst the
film consuming population, a pervasive and positive divergence between the pleasures
derived from actually viewing the film and the expectations of those pleasures prior to
consumption. This pervasiveness, typically, would be built up via a word-of-mouth
information cascade, as the consistently positive messages relating to the film’s qualities
are disseminated throughout the population.
However, as is clear from Figures 1 to 4,
such hit films are relatively rare. For the dilemma that the film producer faces is that
consumer tastes in film are ultimately unpredictable. What the consumer is seeking in
the film consumption experience is novelty – ‘surprises’ – but the consumer is incapable
of articulating what form such novelty might take, and indeed probably has no interest in
doing so. Certainly the consumer will form attachments to specific film ‘markers’ such
as stars and genre, and producers will use these markers in an attempt to perpetuate
previously successful film formulas. However such strategies generally have limited life
cycles, and in effect serve the purpose of fully exploiting previous hit formulas until the
next hit emerges.
As a simple reflection of the dilemmas facing the film producer consider the production
just of high budget films, films for which producers presumably have clear expectations.
Now, the budgets of such films allow for the incorporation of high production values, in
the form of stars, directors, screenwriters, cinematography, and as such, producers would
hope that these films would stimulate extensive consumer interest, thereby generating
high box-office revenues, if not profits. Thus we could interpret the size of production
budgets as directly reflecting the level of producer expectations concerning film success.
Similarly, the extent of consumer reaction to films – the level of satisfaction derived –
would be reflected in the box-office revenues generated. Were producers able to
accurately anticipate consumer reactions then we would expect there to be a close
correspondence between production budgets and film revenues – higher budget films
would generate higher revenues. The simplest way of measuring the extent of this
correspondence is via a rank correlation coefficient. That is, if films are ranked
according to production budget from smallest to largest, and the films are then ranked
according to box-office revenue, from smallest to largest, then if the rankings by
production budget exactly match those by box-office revenue this would produce a
maximum rank correlation coefficient of 1. This would mean that the highest budget film
would produce the highest revenue, the second highest budget film would produce the
second highest revenue, and so on, down to the lowest budget film producing the lowest
revenue. The implication would be that film producers are able to perfectly anticipate
consumer responses, and could always be sure that a high film investment would pay off
in terms at least of consumer responses (this does not of course mean that the film would
necessarily be profitable – so much may have been spent on production that even the high
revenues that such a film generates would not be enough to cover the costs). Conversely,
the minimum rank correlation coefficient of −1 would be produced if the production cost
ranks were the reverse of the box-office ranks – the highest budget film generates the
lowest revenue, the second highest budget film generates the second lowest revenue, and
so on down to the lowest budget films generating the highest revenue. In such a situation
we could interpret film producers as being hopelessly confused with regard to
anticipating consumer responses, consistently investing in the wrong film projects. A
rank correlation coefficient of 0 would imply a random correspondence between
production cost and revenue ranks – a random distribution of high budget films amongst
high and low ranking revenue generating films (and vice versa).
We can use both our 1930s and 1990s datasets to derive these rank correlation
coefficients. We will define a high budget film, as we did above, as any film costing
more than twice the cost of the average cost of all films produced in the film’s year of
release. During the 1930s 179 high budget films so defined were produced by MGM,
RKO and Warner Bros. In the 1990s the major studios produced 247 high budget films.
The (Spearman) rank correlation coefficients between film production costs and box-
office revenues were 0.461 for the 1930s and 0.254 for the 1990s, implying only a
moderate correspondence between producer expectations and consumer satisfaction. In
turn we would argue that this is as direct reflection of the difficulty producers had in
predicting the evolving nature of consumer tastes. However, even if these rank
correlation coefficients are interpreted as reflecting reasonable success in producers being
able to anticipate consumer responses, within the context of a relatively volatile market,
the rank correlation coefficients derived from production budgets and film profits
demonstrate the fundamental difficulties facing producers. For the 1930s the rank
correlation coefficient between film budgets and film profits was −0.193 and for the
1990s is was 0.059. Thus while producers might have had some modest success in
generating higher revenues from higher budget films, this success was countered by the
fact that such strategies had no impact on profits – in essence, in attempting to stimulate
consumer satisfaction through high production value films, producers were unable to
control these budgets sufficiently to ensure a commensurate impact on profits.
Another way of characterising the risk environment faced by film consumers can be
derived by drawing on a simple experiment.
This involved the generation of a self-
selecting sample of filmgoers, achieved via an invitation to complete a short on-line
questionnaire. Apart from providing some basic demographic information and data on
film viewing habits, respondents were asked to evaluate their last movie viewing
experience by indicating, first, the nature of their expectations prior to viewing the movie
(on a 5 point scale ranging from ‘very high’ to ‘very low’) and then, second, to indicate
the level of satisfaction actually derived once the movie had been viewed (on a 5 point
scale ranging from ‘very satisfied’ to ‘not satisfied at all’). Thus we could interpret the
difference between these two 5 points scales as reflecting the extent of divergence
between expectations and realisations – the extent of ‘disconfirmation’ of the expected
moviegoing experience – with a value of −4 (‘very high’ expectations ‘minus’ ‘not
satisfied at all’) indicating considerable disappointment with the film, and a value of 4
(‘very low’ expectations ‘minus’ ‘very satisfied’) indicating unexpected pleasures, with a
zero value indicating a close correspondence between expectations and realisations.
Figure 5 presents a histogram of this disconfirmation variable, together with an
interpolated smooth curve fitted to the data. Thus we could conclude that within this
sample at least there is a marked incidence of a divergence between expectations and
realisations of the movie consumption experience, and that this distribution is a broadly
symmetrical one.
Given the unscientific manner in which these sample data were selected it is clearly
inappropriate to draw any detailed conclusions concerning the population of film
consumers. However, this experiment does allow for a degree of speculation to be
undertaken. Thus note first that the data were collected on the basis of the last movie that
the consumer viewed, and hence covers a range of different movies. This experiment
could be repeated, but focusing on a given film, thus allowing conclusions to be drawn
with regard to the extent of disconfirmation associated with this film. This would imply
that in the case of hit movies, the disconfirmation distribution would be shifted to the
right, and in the case of ‘flops’, the distribution would be shifted to the left. That is, we
could interpret the word-of-mouth information cascade that builds up around any given
film, together with the marketing and promotional strategies undertaken by the film’s
distributors, as determining the manner in which this distribution evolves over time, with
the parameters of this distribution (mean, variance and skewness) reflecting consumer
responses to the film. That is, within the context of Figure 5, we would expect a hit film
to produce a disconfirmation distribution that is shifted to the right, distributed tightly
around a relatively high and positive value of disconfirmation, and perhaps with a degree
of skewness producing a long, thin left-hand tail – even for a hit film there will be some,
very small proportion of consumers for whom the film generates disappointment.
Conversely, for ‘flops’, we would expect the disconfirmation distribution to be shifted to
the left, focused on a low and negative value of disconfirmation (and possibly a long, thin
right-hand tail).
Some conclusions
The key to understanding the risks inherent in film production is to recognise that film
consumers are themselves engaged in a risk process when choosing and viewing a film.
Consumers seek novelty in the film viewing experience, but by definition are incapable
of defining the form that this novelty might take – they will ‘know it when they see it’.
For those consumers who restrict themselves to particular genres and/or stars the extent
of novelty sought in a new film may only be incremental, but the consumer would still
expect something new or novel in the film, otherwise the repeat viewing of films would
be much more pervasive than is observed to be the case.
The manner in which film producers have dealt, and continue to deal, with this
uncertainty has been to produce large and diversified annual portfolios of films, and
inducing consumers to consume extensively across these portfolios, thereby allowing the
hits of the season to emerge. Such an interpretation explains why cinema admission
prices have traditionally been relatively low and invariant across films, a pricing strategy
which can be justified in two ways. First, low admission prices stimulate consumption,
ensuring that annual film portfolios are sampled widely. And second, discriminatory
pricing policies on the part of film distributors/exhibitors – such as higher prices being
charged for what are anticipated to be ‘blockbuster’ movies – would, by implication,
infer that lower priced movies are of lower quality thereby providing a disincentive to the
consumption of such movies. Furthermore, the disappointment of consuming
unsuccessful high budget/blockbuster movies – which, as can be seen from Figures 1 and
2, occur regularly – would be intensified by a high admission price policy for such films.
In other words, any attempt at price discrimination would imply that there is a
price/quality trade-off in movie consumption, and encourage in the consumer an overly
strategic approach to the selection of films to be considered for consumption, thereby
inhibiting the emergence of the ‘unexpected’ hits.
Our discussion of the concept of disconfirmation that consumers experience in the film
consumption process and illustrated in Figure 5, also allows for an interpretation to be
made of the manner in which the information cascade evolves throughout the period of a
film’s release. It would seem reasonable to assume that consumers would experience the
greatest level of risk at the point at which a film is initially released, when information
concerning consumer reception of the film is at a minimum. Thus it is important from
the film distributor/exhibitor’s perspective that the word-of-mouth information cascade
be stimulated as efficiently as possible, an outcome that is achieved by the common
practice of blanket releasing films. Hit films will result from a high incidence of positive
disconfirmation amongst filmgoers, generating a positive word-of-mouth information
cascade, and the converse being the case for ‘flops. However, it also important to
recognise that once a film is released film producers can have little influence on the
success that the film might subsequently achieve. Indeed, our disconfirmation
interpretation of the manner in which films are received by consumers implies that
producers should be careful not to over-promote films or make exaggerated claims for
them, the potential consequence being that consumer expectations are unrealistically
heightened and hence disconfirmation reduced and positive word-of-mouth thereby
Anderson, C. The Long Tail: Why the Future Is Selling Less of More, (New York,
Hyperion Press, 2006).
Clemons, E. ‘How Information Changes Consumer Behavior and How Consumers
Behavior Determines Corporate Strategy’, Journal of Management Information Systems,
25, (2008) pp.13-40.
Elberse, A., ‘Should you invest in the long tail?’ Harvard Business Review, July-August
(2008), pp.88-96.
Goldman, W., Adventures in the screen trade (New York, 1983).
Moul, C., ‘ Measuring Word of Mouth’s Impact on Theatrical Movie Admissions’,
Journal of Economics and Management Strategy, 16, (2007) pp. 859-892
Pokorny, Michael and John Sedgwick “Profitability trends in Hollywood, 1929 to 1999:
somebody must know something,” Economic History Review (forthcoming).
Pokorny, M. and Sedgwick, J., ‘Stardom and the profitability of filmmaking: Warner
Bros. in the 1930s’, Journal of Cultural Economics, 25 (2001) pp.157-184.
Ratner, R., Kahn, B. and Kahneman, D., Choosing Less-Preferred Experiences for the
Sake of Variety, Journal of Consumer Research, 26 (1999) pp. 1-15
Sedgwick, J. and Pokorny, M. ‘The relationship between consumer risk and producer
risk: evidence from the film industry during the 1930s.’ Business History (forthcoming).
Sedgwick, J., and Pokorny, M., ‘The risk environment of film-making: Warner Bros. in
the Inter-War period’, Explorations in Economic History, 35 (1998), pp.196-220
Thompson, K., Exporting Entertainment: America in the World Film Market 1907 -
1934, (London, BFI, 1985).
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Production Costs ($m, 1929 Prices)
Estimated US Profits ($m, 1929 Prices)
Marie Antoinette
Wizard of Oz
The Good Earth
Northwest Passage
Great Waltz
The Great Ziegfeld
Sergeant York
Boom Town
Snow White
Mrs Miniver
San Francisco
Day at the Races
Mutiny on the Bounty
Abe Lincoln in Illinois
Boys Town
Honky Tonk
Gold Diggers of '33
Love Finds Andy Hardy
Hardys Ride High
At the Circus
Test Pilot
Figure 1 Scatter of US Profits against Film Costs, 1929 Prices, 1930 to 1942
(n = 1,796)
0 20 40 60 80 100 120 140
Production Cost ($m, 1987 Prices)
Estimated US Profits ($m, 1987 Prices)
Speed 2
Terminator 2: Judgement Day
Star Wars: Phantom Menace
Lethal Weapon 4
Blair Witch Project
Jurassic Park
Home Alone
Independence Day
Sixth Sense
Forrest Gump
Indiana Jones
Lion King
Father's Day
Hard Rain
Cutthroat Island
Soldier Postman
Batman Returns
Figure 2 Scatter of US Profits against Film Costs, 1987 Prices, 1988 to 1999
= 2,116)
Figure 3 Frequency Distribution of US Profits, 1929 Prices, $m, 1929 to 1941
(n = 1,796)
Figure 4 Frequency Distribution of US Profits, 1987 Prices, $m, 1988 to 1999
(n = 2,116)
Figure 5 Frequency Distribution of Disconfirmation in a Sample of 222 Film
-4 -3 -2 -1 0 1 2 3 4
Goldman (1983)
See Michael Pokorny and John Sedgwick “Profitability trends in Hollywood, 1929 to 1999: somebody
must know something,” Economic History Review (forthcoming) for a more detailed discussion of these
data sets, and in particular the methodology that was used to estimate film profits.
See Pokorny and Sedgwick ‘Profitability trends’, for a more detailed discussion of these issues.
See Anderson (2006); Clemons (2008); and Elberse (2008) for a discussion of this phenomenon applied
to internet retailing.
See Sedgwick and Pokorny (1998) and Pokorny and Sedgwick (2001) for developments of this idea.
Ratner, Kahn, and Kahneman (1999)
See Moul (2007), who tests a model estimating the significance of word-of-mouth in forming consumer
See Sedgwick and Pokorny, ‘The relationship between consumer risk and producer risk’, Business
History (forthcoming), for a fuller discussion of these issues.
This experiment was carried out by Claudia Domokos, Tobias Pospischil, Karina Sokolova, Richard
Wallace, and Michael Zieniewicz, under the supervision of John Sedgwick.
... Furthermore, the success of a movie is often tied to several costly factors such as the "big-name" stars playing in them, the "star directors" accepting to direct them, the quality of the production and the mass availability in movie theater around the world and the marketing campaign (Pokorny et al. 2012). ...
... Another outstanding feature of film production, highlighted by the paper of John Sedgwick and Michael Pokorny (2012) (Pokorny et al. 2012), is that the variability of film profit increases as production budgets increase, reflecting the increasing levels of risk associated with high budget film production. ...
This thesis is about exploring the impact of blockchain technology, and especially ’smart- contracts’, on the Belgian Tax Shelter mechanism and more generally, on Belgium film financing. Film projects financing and accounting is often referred to as both the most complex and the most creative part of film productions. In Hollywood, two types of financial instruments allows equity investment in movie projects: Ticket-linked bonds and Asset-backed securities, also called Slate financing. Film accounting is said to be creative because of some malpractices of production com- panies inflating expense accounts in order to reduce the reported profit of the movie. This practice, often referred to as “Hollywood accounting”, allow production companies to pay themselves hefty fees while reporting lower profit numbers to equity investors (Cheatham 1996). The complex part of film financing refers the not only the difficulty of raising funds for creative projects but also to the necessary layers to finance a project. As it will be exposed in the corresponding sessions, the financing of a film is close to what is done in private equity, especially in the US film industry. In Belgium, like the rest of Europe, the cinema industry is mostly publicly funded. There are many reasons for this: Historical reasons, promoting country culture, the desire to support a creative and intellectual industry, the multiplier effect on the economy, the creation of jobs, ... (Olsberg 2012) In 2004, the Belgian government introduced the Tax Shelter act that allowed companies to finance Belgian film projects in exchange for a tax relief on corporate tax and a compensation based on box office revenues. Financial intermediaries, with the purpose of matching investors and film projects, have appeared in the Belgian cinema industry and have become, with the time, an essential keystone in the Belgian movie value chain. Since then, the law has had to be modified because of abusive behaviors from some Tax Shelter intermediaries and production companies. Most of the time, the abusive practice consisted in promising a very high return on investment to tax shelter investors by proposing to buy film shares back at a high price through a put option, even if the film box office didn’t reach its profit target. Another major scandal triggered the reform: Corsan, a tax shelter intermediary, paid previous investors with funds for newcomers, building in fact a Ponzi scheme that col- lapsed when the Tax Shelter department of the Ministry of Finance imposed stricter controls on investments and film budgets. These problems of trust between different stakeholders could be resolved by a new kind of technology currently disrupting many industries and threatening incumbent companies: the distributed ledger technology(DLT), also named blockchain. Some think that it will cause the end of banks, notaries and all other kind of inter- mediaries... even so-called circular economy companies like Uber and AirBNB could be replaced by blockchain-powered smart contracts. Others think that blockchain is a threat and a fraud, only useful to scam naive users with cryptocurrencies and to buy illegal goods and services on the dark net. Whatever the convictions, in 2017, many developers raised more than $4 billion through ICOs in the United States, and more than $11billion has been raised during the first semester of 2018 (Gurrea-Mart ́ınez et al. 2018). The blockchain trend continues to grow and there are already several projects trying to disrupt the cinema industry. The objective of this thesis was to analyze the potential impact of the blockchain technol- ogy on the tax shelter funding mechanism in Belgium and on film financing in Belgium more generally. In order to assess such a new technology, which have only few years of development and only proof of concepts of implementations, it has been decided to perform a thorough review of all the necessary background in film financing, blockchain technology and briefly in financial regulation. Then, a review of the existing literature on film financing mechanisms and blockchain regulation helped understand the current possibilities and implications of an implemen- tation in Belgium for film financing. To more precisely apprehend the possibilities of blockchain implementation, interviews of experts and attendances to conferences helped refine the initial idea. Finally, a financial intermediary, SCOPE Invest, was analyzed and a blockchain im- plementation was suggested. From this implementation, conclusions were drawn which allow to iterate and to propose a new potential future use case of the blockchain, accord- ing to what already exists and based on article conclusions and European regulations. Furthermore, a third implementation was proposed, outside of the scope of tax shelter, in order to show the possible implementation of a decentralized film financing fund and its operating principle. To conclude, this dissertation found that blockchain will have a profound impact at the transaction level of the tax shelter mechanism. It is not unlikely that governments bodies, producers and intermediaries collaborate on a blockchain implementation in order to build more trust between parties and to reduce auditing and reporting costs. Moreover, more disruptive innovation will probably come along the way of the ”tok- enization” of the economy, but will require more time, according to the interviewees, to convince all the stakeholders and especially, regulatory bodies. Future research on the subject should focus on testing the implementation at various level: • Many frameworks could be used but the limitations should be assessed in order to pick the one that fits best the problem. • In the case of a DAO, the problem of regulation of the organization and of the tokens should be assessed more precisely • The behavior of the DAO investors could be analyzed to understand what differs from a traditional company/cooperative • The perception of potential investors towards the DAO and the film tokens could also be assessed
... Most major productions require multiple sources of funding, and so the very financing structures of film and high-end television production promote and reward cross-border co-production arrangements. Often, they depend on initial sources of public funding from national funding agencies which are then used to lever additional finance through complex co-production agreements and advances from distributors and sales agents (Doyle, 2013;McElroy and Noonan, 2019;Mitric, 2018;Murschetz et al., 2018;Pokorny and Sedgwisk, 2012;Sørensen and Redvall, 2021). Tax incentives and inwards investment schemes play a crucial role in this, and these are predicated on production and post-production taking place across coproducing countries in order to leverage national and regional funds. ...
Full-text available
Climate change is a concern for many nations, industries and citizens. However, for some it will also be a moment of opportunity. As witnessed in relation to the pandemic, dynamics of power have a particular purchase at moments of crisis. This article proposes a greater concern with questions of policy and power in relation to green strategies within the screen sector and highlights the role that media scholars might play in developing this critical lens. There has been a recent rise in efforts to mitigate particular environmental harms and the article outlines some of the initiatives that are emerging from both commercial and public bodies. Here are no shared systems for auditing and reporting, and few formal policies are widely recognised or adopted. As different ideas and approaches garner greater traction, travelling from one national and industrial setting to others, the article scrutinises the industrial, structural and policy obstacles which hinder a meaningful shift for film and television production to be environmentally sustainable. The article reflects on the agendas and forces at play in this space and attempts to stimulate debate about how those researching media production might productively engage in critiquing these policies and dynamics of power.
The Toronto International Film Festival (TIFF) attracts filmmakers and film audiences from around the world each year. In this article, I take an ethnographic approach to understand the way film festival organizers bring groups together and act as intermediaries in negotiating the value of particular films and cinema. I argue that film festival organizers engage in glamorous work—a strategy that employs visual cues and artful mediation to cultivate a sense of the extraordinary while obscuring the risky nature of film festival organization. This article considers the risks involved in film festival organization and the way glamorous work attempts to alleviate these risks.
Full-text available
This paper examines the risk environment of film consumption in the United States during the 1930s when moviegoing dwarfed all other paid-for leisure activities. We argue that the wide variability in the financial performance of films, reflecting the considerable risks that were involved in film production, can be interpreted as being mirrored in the risks incurred by consumers in the film consumption process. We further argue that production risk needs to be understood within the context of consumer risk. Using a dataset derived from the trade journal Variety, we examine the weekly fortunes of movies in first-run cinemas as consumers rapidly substitute movies that are currently on release for the promised pleasures of yet unseen movies. That expected utility was not always realised was commonplace, as was the pleasurable surprise that came with being thrilled by certain films. These are important results since, perhaps for the first time in modern society, they led to the emergence of the long right tail of consumer preferences for mass distributed goods.
Full-text available
Data from several experiments show that, contrary to traditional models of variety seeking, individuals choose to switch to less-preferred options even though they enjoy those items less than they would have enjoyed repeating a more-preferred option. Two explanations for this finding are tested. Results indicate no evidence of a benefit to more-preferred options due to the contrast to less-preferred alternatives. However, the results of three studies suggest that retrospective global evaluations favor varied sequences that also include less-preferred items as opposed to sequences that only include more-preferred items, even though these more varied sequences result in diminished enjoyment during consumption. Copyright 1999 by the University of Chicago.
Full-text available
This paper examines the film production performance of Warner Bros. during the 1930s, placing particular emphasis on the manner in which Warners invested in stars. Warners are shown to have acted rationally in the sense of having consistently invested in previously successful actors. An assessment is then made of how successful such a strategy proved to be. Drawing a distinction between high and medium/low budget production, the deployment of established stars in high budget productions did not appear to have constituted a successful strategy. The production of medium/low budget films, by contrast, provided a more stable environment, in which there were clear returns to the deployment of previously successful actors. Copyright Kluwer Academic Publishers 2001
Full-text available
"Information transmission among consumers (i.e., word of mouth) has received little empirical examination. I offer a technique that can identify and measure the impact of word of mouth, and apply it to data from U.S. theatrical movie admissions. While variables and movie fixed effects comprise the bulk of observed variation, the variance attributable to word of mouth is statistically significant. Results indicate approximately 10% of the variation in consumer expectations of movies can be directly or indirectly attributed to information transmission. Information appears to affect consumer behavior quickly, with the length of a movie's run mattering more than the number of prior admissions." Copyright 2007, The Author(s) Journal Compilation (c) 2007 Blackwell Publishing.
Full-text available
This paper examines the financial strategies employed in the process of film production with particular emphasis on the performance of Warner Bros during the period from 1921 to 1940. The setting of film budgets is interpreted within a context of the financial risks involved, and in particular, a portfolio theory approach is found to provide a useful framework for describing and analyzing risk. The rapid growth of the film industry in the interwar years, together with the economic volatility of the period, produced a variety of risktaking strategies, and this paper attempts to assess the relative success of these strategies.
The blockbuster strategy is a time-honored approach, particularly in media and entertainment. When space is limited on store shelves and in traditional distribution channels, producers tend to focus on a few likely best sellers, hoping that one or two big hits will carry the rest of their lists. But online retailing and the digitization of information goods have changed the commercial landscape: Virtual shelf space is infinite, consumers can search through innumerable options, and the marginal cost of reproducing and distributing products is low. What does that mean for the blockbuster strategy? In his 2006 book, The Long Tail: Why the Future of Business Is Selling Less of More, Chris Anderson, editor of Wired magazine, argues that the sudden availability of niche offerings more closely tailored to their tastes will lure consumers away from homogenized hits. The "tail" of the sales distribution curve, he says, will become longer, fatter, and more profitable. Elberse, a professor at Harvard Business School, set out to investigate whether Anderson's long-tail theory is actually playing out in today's markets. She focused on the music and home-video industries - two markets that Anderson and others frequently hold up as examples of the long tail in action - reviewing sales data from Nielsen SoundScan, Nielsen VideoScan, the online music service Rhapsody, and the Australian DVD-by-mail service Quickflix. What she found may surprise you: Blockbusters are capturing even more of the market than they used to, and consumers in the tail don't really like niche products much. Elberse outlines the implications of her research for producers and retailers, and offers strategic advice to both groups.
This article presents an overview of the development of the US film industry from 1929 to 1999. Notwithstanding a volatile film production environment, in terms of rate of return and market share variability, the industry has remained relatively stable and profitable. Film production by the film studios is interpreted as analogous to the construction of an investment portfolio, whereby producers diversified risk across budgetary categories. In the 1930s, high-budget film production was relatively unprofitable, but the industry adjusted to the steep decline in film-going in the postwar period by refining high-budget production as the focus for profitability.
Information availability has increased consumers' informedness, the degree to which they know what is available in the marketplace, with precisely which attributes and at precisely what price. This informedness has altered the demand side of market behavior: customers now discount more heavily when comparable products are available from competitors and when products do not meet their wants, needs, cravings, and longings, but they no longer discount as heavily when purchasing unfamiliar products. Changes in the demand side are producing comparable changes in the supply side: firms earn less than their expectations when competing in traditional mass-market fat spots, while earning far more than previously when entering newly created resonance marketing sweet spots. We trace the impact of hyperdifferentiation and resonance marketing on strategy, with a clear progression from a limited number of fat spots, through reliance on line extensions, and ultimately to fully differentiated market sweet spots.