American Economic Association
Contracts between Art and Commerce
Author(s): Richard E. Caves
Reviewed work(s):
Source: The Journal of Economic Perspectives, Vol. 17, No. 2 (Spring, 2003), pp. 73-84
Published by: American Economic Association
Stable URL: http://www.jstor.org/stable/3216857 .
Accessed: 20/02/2013 10:10
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
American Economic Association is collaborating with JSTOR to digitize, preserve and extend access to The
Journal of Economic Perspectives.
http://www.jstor.org
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
Journal of Economic Perspectives?Volume 17, Number 2?Spring 2003?Pages 73-83
Contracts Between Art and Commerce
Richard E. Caves
Economists
interested in the creative industries like arts and entertainment
have had trouble mobilizing a set of economic tools suited to understand?
ing such issues as why these industries are organized as they are and what
consequences public policies hold for them. However, some useful tools have
recently emerged from the theory of contracts. The basic structural characteristics
of creative industries?their technologies of production and consumption?
fiercely resist governance by anything approaching a complete contract. Yet they
have evolved distinctive and serviceable contract forms that seem to differ from
deal-making patterns prevalent in other sectors.
Great works of art may speak for themselves, as connoisseurs declare, but they
do not lead self-sufncient lives. The inspirations of talented artists reach consumers'
hands (eyes, ears) only with the aid of other inputs?humdrum inputs?that respond
to ordinary economic incentives. The visual artist needs a gallery to display and
promote works to potential purchasers. The author requires a publisher, the pop
musician a record label. A symphony orchestra or dance company wants a hall and
someone to market the tickets. The Hollywood movie, the Broadway play, the TV
sitcom?each demands a diverse group of creative talents as well as a set of
humdrum inputs. The question of how contracts work between art and commerce
thus is nested within the larger question of why artists and humdrum inputs choose
to structure their relationships as they do.
It turns out that the organization of the arts and entertainment industries
depends heavily on the contracts that link creative and humdrum agents. These
contracts vary in form, but all depend strongly on a common set of basic structural
? Richard E. Caves is Nathaniel Ropes Professor of Political Economy, Harvard University,
Cambridge, Massachusetts. His e-mail address is {rcaves@harvard.edu).
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
74 Journal of Economic Perspectives
properties.1 All fall some distance short of the theoretical ideal of a complete
contract that prescribes the signatories' actions in all possible states of nature. But
all do seem to share a "best-feasible" property that leads the deals struck in diffuse
sectors to exhibit common features.
The Basics: Bedrock Structural Properties and Contract Theory
Several bedrock structural properties seem to underpin the organization of the
creative industries. They also distinguish arts and entertainment from other eco?
nomic sectors and, in some cases, distinguish between groups of these activities.
Each property gets an identifying catch phrase.
Nobody knows refers to the fundamental uncertainty that faces the producer of
a creative good. All inputs must be incorporated and the good presented to its
intended customers before the producer learns their reservation prices. Producers
make many decisions that affect the expected quality and appeal of the product, yet
their ability to predict its audience's perception of quality is minimal. Producers
know a lot about what has succeeded in the past, and they constantly try to
extrapolate this knowledge to the project at hand, but these efforts achieve minimal
predictive value (Goldman, 1984).
That nobody knows would matter little if the inputs to a flopped creative effort
could be salvaged and reused. However, the ubiquitousness of sunk costs denies the
producer that protection. A creative good's suppliers must snag enough rents from
each hit to cover the losses on several flops.
Another fundamental property lies in the attitudes of artists toward their work.
Art for art's sake invokes the utility that the artist gains from doing creative work.
Economists normally assume that work occasions disutility. However, artists may
accept wages for creative work that fall short of their opportunity cost in humdrum
employment, which means that artists can be viewed as a source of cheap labor. Art
for art's sake also embraces artists' tastes as to how the creative work is to be
performed, the technique or style to be employed. The artist's preferences over
how to execute creative work complicate the contracting process, particularly the
definition and allocation of decision rights.
Other basic properties involve horizontal and vertical differentiation, temporal
coordination, durability and the hazards of coordination when several artists must
collaborate. These properties will be introduced as needed.
To understand how these properties shape the deals that govern creative
1
Indeed, some economists would roll professional sports into the package. This scope is customary in
the legal field of entertainment and sports law (Weiler, 1997).
2This approach is developed in more detail in Caves (2000, especially the "Introduction").
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
Richard E. Caves 75
industries' operation, we draw heavily on contract theory.3 Here we flag some
points of connection between contract theory and the conditions of creative
industries. First, contract theory pays much attention to asymmetrical information,
which usually involves a situation in which the seller knows key characteristics of the
product not known to the buyer. However, in creative industries nobody knows, and
the core problem is one of symmetrical ignorance. Second, while principal-agent
relationships are important in creative industries, many deals in creative goods
involve joint ventures between symmetrically placed suppliers of diverse inputs.
Third, production in creative industries commonly involves inputs attached in
technologically determined sequences: a movie director cannot plan the shooting
until the screenplay has been finalized. Critical problems then arise of assigning
and transferring decision rights as the project passes from one input supplier to the
next.
Bilateral Deals: Artist and Facilitator
We start with deals of the joint-venture variety, involving one artist and one
humdrum party able to bring the artist's inspiration to potential consumers.
Examples of such a deal include visual artist and art dealer, author and publisher,
musician and record label. The contracts observed in each of these settings illus?
trate how the bedrock structures of creative industries shape the structures of deals.
Visual Artist and Art Gallery
The relationship between artist and art gallery may seem the simplest of
economic transactions, yet it is rendered complex and problematic by the bedrock
properties. Art for art's 5<z&?carries a career twist evident from young artists' training
and apprenticeship. The nineteenth-century romantic ideal sets the artist's task: to
seek new problems of creative visualization and devise compelling solutions. Art
progresses through a dialogue of problems, solutions that point to new problems
and so forth. The artist, engaged in this dialogue, takes satisfaction from the work
itself and not the acclaim (if any) that it elicits. Art is not craft or mere decoration.
The training of visual artists instills these attitudes and evaluates the student
according to creativity or originality, and not skill or proficiency (Getzels and
Csikszentmihalyi, 1976).
The romantic ideal explains why artist and dealer agree on the long-term joint
venture that we commonly observe. The artist enters into a protracted partnership
with a dealer. The dealer will display the artist's work and articulate its problemsolving
context and evolving meaning to collectors and others. This kind of
contract dominates the distribution of "serious" contemporary art. Within it, the
3
Milgrom and Roberts (1992, chapters 3-7) provide a useful introduction to contract theory. The
contrast between contracting practices in creative industries and the preoccupations of contract theory
need not imply a parallel contrast to the contracting practices in other industries.
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
76 Journal of Economic Perspectives
artist can plausibly be true to the romantic ideal. No form of vertical integration of
the production and distribution of art can preserve this ideal. If the dealer hires the
artist to paint what the dealer thinks will sell, creative autonomy is clearly gone. If
the artist undertakes the dealer function, haggling over prices with the customers,
time making art is lost, and the artist must struggle with tasks that are both
uncongenial and unfamiliar.4
The joint venture between dealer and artist involves a contract in which the
artist prepares work for periodic gallery shows. The dealer undertakes to promote
the artist's work with collectors, critics and curators, and the two parties divide the
gross revenue (perhaps after certain costs are allocated to the artist) with the dealer
claiming up to 50 percent. The contract (often done on a handshake) has no
explicit duration; if the artist's work evolves in directions unappealing to the dealer,
or clashing with the general style represented by the dealer's gallery, the two parties
may go their separate ways.
The division of gross revenue, rather than profit, is a noteworthy feature of this
contract. It likely reflects the nonobjectivity of measuring economic profit and the
opportunism of the party serving as bookkeeper (a theme that will recur in the
subsequent discussion of motion pictures). A joint-venture contract with profits
split (say) 50-50 has certain attractive, if suboptimal, incentive properties. Each
party captures a substantial if not a 100 percent share of any extra profit due to
efforts that are not otherwise compensated directly. The contract offers the flexi?
bility that reimbursed pecuniary costs can be assigned to either party.
On the other side, sharing revenue rather than profit depresses each party's
incentive to incur mutually beneficial costs. It induces each party to try to shift costs
onto the partner. Also, the contract encounters problems of moral hazard. With the
dealer taking a large share of revenue, the artist and the collector acquire an
interest in dealing with each other directly and appropriating the dealer's share;
prohibition of direct sales is standard in artist-dealer contracts and its violation a
common reason for termination. Moral hazard can also infect the dealer's bookkeeping;
sales not immediately reported to the artist become interest-free working
capital for the gallery, and lightly capitalized art galleries have disappeared overnight
along with their artists' consigned works (for more detail, see Caves, 2000,
pp. 21-30, 37-47).
Author and Publisher
The relationship between author and publisher is conceptually simpler than
the artist-dealer alliance. The art for art's sake tastes of both artist and author orient
them toward creating bodies of work over a lifetime career. However, the typical
author finishes a book infrequently, which lowers the stakes for optimizing long-run
future terms in the author-publisher deal. In trade publishing, the literary agent
plays a distinctive role. A readily confirmed implication of art for art's sake tastes is
4These choices for the emerging artist broaden considerably for the established one, for whom
exclusive representation likely loses its value.
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
Contracts Between Art and Commerce 77
that far more people invest in preparation for careers in creative work than can
expect to earn normal pecuniary returns on their investments. Before any jointventure
contract unites art and commerce, some gatekeeper must judge which
aspiring artist displays enough talent that humdrum resources invested in furthering
it can expect a normal return. In trade-book publishing, the agent serves as
marketer of the author's services to the publisher, and the author-agent deal
precedes the author-publisher deal. The logic of the process is as follows.
The agent serves as a matchmaker, knowing what sorts of manuscripts will
interest which publishers and able to discern and bring out merit in an author's
work sufficiently that a publisher will recognize it. As the author's employee, the
agent can elicit information relevant to the publisher: the author's flexibility,
punctuality and other traits pertinent to successful collaboration. The agent tends
to interact repeatedly with the publisher, which aids the credibility of the agent's
recommendation. The agent, who is compensated by a share (traditionally 10 per?
cent) of the author's royalties (Hepburn, 1968), will suffer from representing weak
or uncooperative authors whose published work brings in little revenue. The agent
with a poor track record for picking authors cannot earn a competitive income.
Hence a publisher, expecting that the agent knows more about the author, may set
aside suspicions of hidden negative information, because the established agent
could not prosper on 10 percent of the royalty streams of weak authors.
The publisher's own contract is a revenue-sharing deal that awards the author
a royalty expressed as some fraction of the book's wholesale or retail list price. If
retail margins and the publisher's variable costs are taken as given, standard royalty
rates split the publisher's gross profit (that is, before deduction of the publisher's
fixed costs) around 58-42 percent between publisher and author (Auletta, 1997).
As with other revenue-sharing joint-venture contracts, each party has only an
attenuated incentive to exert joint-profit-maximizing effort. An important device
for mending this defect is the author's advance. An advance is usually regarded only
as working capital provided to the impecunious (liquidity-constrained) author.
However, the publisher enjoys latitude on how much to spend promoting the
author's book (a sum on which author and publisher seem unable to contract).
Until the author's advance is recouped, the publisher pockets the whole extra
dollar of profit due to any promotional effort, causing the profit-maximizing choice
to be made. When the advance has been earned back, the publisher's incentive to
promote again becomes attenuated.5 This role of the advance illustrates an impor?
tant general point about revenue-sharing contracts: a party's incentive can be
intensified without upsetting the distribution of benefit between the parties by
precommitting some outlay that the transferor can then earn back.
In literary circles, a debate proceeds over the publisher's attitude toward the
agent's role. Trade publishers traditionally resented the agent's intrusion into the
personal relationship between author (artist) and publisher (handmaiden), giving
'For other implications of the advance against royalties, see Hansmann and Kraakman (1992).
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
78 Journal of Economic Perspectives
offense to the publisher's own sense of artforart's sake tastes. Furthermore, an agent
likely strikes a better bargain for his authorial principal than could the author
personally. Indeed, over the years, agents have wrested the control of subsidiary
rights (such as paperback editions and movie screenplay rights) from trade-book
publisher to agent and author. The publisher once staged subsidiary-rights auctions
and pocketed half of the proceeds, but agent and author cannot be prevented from
picking up any rents available in this way; after all, the author's talent is the one
distinctive rent-yielding asset in play.
Against these invasions of the publisher's potential cash flow are stacked the
advantages of the agent's services as gatekeeper. Without agents, publishers would
incur the cost of gatekeeping?filtering the "slush pile" of manuscripts that arrive
unheralded.
Musician and Record Label
The last of these bilateral contracts embodies a structure common throughout
the creative industries?one turning on the problem of allocating decision rights
efficiently. It retains the joint-venture form, but it adds the crucial element that the
creative good's production proceeds in steps: one party supplies its input, sinking
its cost in the process, then hands the incomplete good to the next party for
attaching its input specialty. The pop music group completes the tape for an album,
which then passes to the record label for manufacture, distribution and promotion.
Since musician and label both seek an extended career for the musician, their
obligations extend to the musician's future albums as well. A cinema film emerges
from numerous creative inputs applied in sequence. Nobody knows the consumers'
valuation of the end product, but completing another step in the sequence tends
to reduce the uncertainty of the prediction. Each participant presumably knows
more than anybody else about actions and consequences at the stage where its own
input is applied.
Consider how these parties can manage the sequence of production steps to
maximize the expected value of the end product. The first step has been completed
by agent a (the cost of a's input is sunk). The next input supplier J3is qualified to
decide whether to continue by sinking the cost of the second input: J3can observe
the outcome of a's effort and holds the relevant information about the effect of
adding its own input; nobody has better access to information about what state of
nature will prevail upon the product's completion. Most important, J3's input has
not yet been sunk, giving it a strong incentive to be well informed. The expected
value of the final product is raised if J3 obtains decision rights about whether and
how to proceed. J3, however, presumably maximizes its own expected returns from
the project, which need not call for the same actions that maximize its total
expected value. In particular, J3 has an incentive to act so that its second-stage
decisions transfer benefits from a to /3. A suitably foresighted a will of course
anticipate this and demand a side payment from J3 to compensate for this moral
hazard (Grossman and Hart, 1986).
The standard contract between record label and pop musician takes exactly
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
Richard E. Caves 79
this form. The label advances the musician a sum to cover the cost of recording
the first album plus a negotiated amount of expected royalties. Upon delivery
of the tape, the label holds the option to issue the recording. If the label
exercises this option, the clock starts for the musician's delivery of a second
album, which will bring a larger advance and higher royalty rate. Upon each
delivery, the label decides whether to proceed. If one record's royalties fail to cover
its advance, the shortfall becomes a charge against records issued subsequently
("cross-collateralization"). The options are one-sided, in that the musician may not
quit the label and record for another until the contract expires?perhaps after a
decade (Caves, 2000, pp. 61-64; Passman, 1994).
This contract has the efficiency properties previously noted. The musician or
musical group spends its own advance money to record the album, so it has an
incentive to use studio time efficiently (versus indulging in perfectionism). The
label wields decision rights over distribution and promotion, which are its specialty.
Furthermore, the contract's long duration warrants the label running losses on a
promising artist's early albums. Despite this logic, public sympathy flows to the
young musician who seems so deprived of decision rights and locked into a
one-sided relationship. For example, musicians in California have sought to obtain
legislation limiting the duration of record contracts (Ordonez, 2002); the backers
appear to be successful musicians who would benefit from being able to renegotiate
long-term contracts.
Why does not a kinder, gentler record label offer a short-term contract that lets
the performer revel in the rents that come from a big, early success? The answer lies
in the high "stiff ratio," the 80 to 90 percent of recordings that lose money. For the
label to break even in the long run, it must mine enough profits from the successes
to cover the stiffs' losses. Why is the artist without decision rights? Because, with art
for art's sake tastes, any decision right retained by the artist threatens moral hazard
and appropriation of the project's pecuniary value. In principle, the artist could
bargain to retain some decision rights in exchange for reduced pecuniary com?
pensation. However, contracting on creative decision rights is very problematic,
and few examples come to light (one is the cinema film director's "first cut" in the
assembly and editing of the raw film).
Complex Creative Goods with Motley Crews
Complex creative goods require several artistic talents along with humdrum
inputs: cinema films, television program series, symphony orchestras, stage plays
and repertory companies, dance troupes. Each is a complex institution with its own
distinctive features. We focus on cinema and TV films and on how deal-making
practice rests on underlying theoretical concepts.
Cinema Films
The long-standing Hollywood studios distribute and promote films, but each
movie rests on a separate contract linking its producer to the actors, director and
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
80 Journal of Economic Perspectives
other key talents. The studio is intimately involved, though, when it finances the
production cost of a film as well as distributing it (standard practice for big-budget
films). Each studio carries out an intensive gatekeeping (filtering) process that
keeps numerous projects "in development," a purgatory in which interested parties
rework and rewrite their projects to overcome studio skeptics and obtain the "green
light."
The producer as coordinator identifies and recruits the principal creative
participants. Vertical differentiation plays an important role in this and other
complex activities.6 A creative good's quality in the eyes of consumers can be
increased by enlarging the fixed cost expended on it. These extra fixed costs might
buy more elaborate special effects, crowds of extras and the like, but especially they
buy more skilled (costly) creative participants. Those participants come ranked
consensually within their creative communities?what is known as the A-list/B-list
property. Films seem to aim for a consistent level of quality in their inputs. This
practice might be attributed simply to complementarity among high-quality inputs:
a better leading actress adduces a better performance from the leading actor.
However, there seems to be something more involved than pair-wise complemen?
tarity. Every input needs to perform at least up to some level of dedication and
proficiency to beget a work of unified quality. This requirement corresponds
theoretically to the "O-rings property" of multiplicative production processes,
named after a key component on the space shuttle Challenger whose failure
contributed to the shuttle's explosion: an output's quality depends on all inputs
performing up to some standard (Kremer, 1993). As predicted, A-list talents work
with one another on film projects more commonly than would result from random
assignment (Baker and Faulkner, 1991). That participants be available at the right
time is another constraint on the assembly of projects, and it also feeds back to the
process of putting the deal together. Committed participants incur sunk costs of
negotiation and foreclosure of other opportunities, and so the last party to sign
gains some strategic leverage to appropriate benefits expected by committed
predecessors. Deal participants can keep this threat from their own doorways by
demanding a "play or pay" contract, which guarantees one's pay even if dismissed
from the project or the film does not get made. Hold-ups of this and other types
encounter a defense in the form of reputation in the film-making community,
where word of uncooperative behavior spreads freely (as do consensual evaluations
of A-list/B-list standings).
Pecuniary compensation in film contracts trades against other sources of utility
to creative participants, which complicates the negotiation process. Compensation
of stars includes a rent that (in principle) reflects the excess of expected gross
revenue over what the film would earn with an ordinary talent in the role. Given the
performer's pecuniary rent (if any), the performer may also may trade off cash
6Vertical differentiation is product differentiation with the available brands uniquely ranked by quality.
That is, everybody agrees that A is more desirable than B, B than Cand so on, but they need not agree
on their willingness to pay A's higher price.
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
Contracts Between Art and Commerce 81
compensation for credit in a film that appears likely to garner critical esteem.
Finally, up-front fixed compensation can be traded for deferred and contingent
compensation, normally a share of either the film's gross revenues or its net profits.
Two questions arise about Hollywood deal practices. First, if screen talents are
risk averse, their compensation must increase when it takes a contingent form.
Giving the star a risk premium must generate offsetting value elsewhere. Two
competing mechanisms that might do the trick include the incentive value of
contingent compensation for getting (say) Sylvester Stallone to do his Rocky act
once again and the reduction in risk premia demanded by other participants when
risk is loaded onto the star. Scholars have addressed but not resolved these
competing explanations (Chisholm, 1997; Weinstein, 1998; Goldberg, 1997).
The second issue lies in Hollywood accounting, which proves notoriously
adroit at causing profits (that the studio must share with the talent) to recede into
the indefinite future (Bibicoff, 1991). The types of costs invoked to devour any
tendril of net profit generally have real economic bases. The trouble is that the
studio distributing the film is both bookkeeper and residual profit claimant, subject
to severe moral hazard (fraud, in the eyes of the victims).7 A telling feature: net
profit recipients may negotiate the definition of net profit to be employed. Con?
tingent compensation hence tends to be keyed to gross revenue rather than net
profits, with generally less satisfactory effects on incentives. This issue arises most
notoriously in Hollywood, but the same problems of strategic accounting and moral
hazard arise in all the creative industries.8
TV Program Series
The contracts that govern TV program series differ considerably from the
contracts that govern cinema films. In viewers' eyes, the leading actors in a TV series
become one with the characters that they play, and successful replacements are
rare. If the program succeeds, the key actors gain strategic leverage to appropriate
the program's profit by threatening to quit. Producers ward off this hold-up by
insisting on contracts that bind key actors for long time periods. From the moment
an actor first reads for a role in a new sitcom, the actor is committed to an option
contract that commonly runs for five years. If the program is selected by a network
and episodes are ordered, the producer and the network have the option at each
season's end to continue the program and the actor's role. The renewal triggers a
previously negotiated increase in pay, but far short of the rent imputed to the actor
7
Why is the record keeping left with an intensely interested party?The only practical alternative would
be the producer, who also takes a profit share and thus also faces moral hazard. Furthermore, the studio
makes the key decisions about promotion and distribution. These it controls whether it keeps the books
or passes the receipts along to another party.
8 One reason why they arise is the sheer scale of the rents flowing to top talents in the creative
industries?a pattern explained by the superstar effect (Rosen, 1981). Superstars' contracts differ from
standard practices (as a rough generalization) more in the sums involved and the intensity of haggling
than in the forms of contract structures. However, systematic differences do exist: superstars can forego
cash for decision rights in pursuit of art of art'ssake.
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
82 Journal of Economic Perspectives
if the program is a big success. Consequently, when a successful program such as
Seinfeld or Friends outlasts the standard contract duration, the star performer's
compensation is up for grabs. The irreplaceable performers as a group then can
claim the full rent expected to flow from subsequent episodes of the program (for
example, Carter, 1998) and divide it among themselves according to bargaining
power.
The option contract is accepted practice in TV broadcasting, but it clearly
exacts its costs. Programs sometimes get cancelled because stalemates arise in
renegotiating contracts. This outcome should not happen if all parties share the
same expectations about the program's future cash flows, but they need not. The
lock-in of actors to new programs distorts the planning process. The networks
commission many more programs and pilot episodes than they will actually use
(nobody knows), and many actors are committed to options that will not be exercised.
A program selected by a network can sometimes "trade up" when a preferred actor
previously committed to another program becomes available, but an optimal
assignment of actors to roles is out of reach.
Future Research
Aided by simple contract theory, the patterns of deal making in the creative
industries tell a rich story about how parties?both artist and humdrum?structure
agreements to address complex incentive problems. One does not think of the arts
and entertainment industries as sectors organized around sophisticated commer?
cial contracts. Yet the very structural features that challenge economic calculation
serve to explain why deals are structured as they are. Revenue-sharing joint ventures
with up-front payments and real option contracts with successive transfers of
decision rights make repeated appearances in the creative industries. A great deal
of qualitative evidence supports these descriptive conclusions. However, we lack a
deeper knowledge about the trade-offs and terms of these deals that could come
from access to samples of contracts and their specific terms. Questions such as these
might receive answers: To what degree are contracts formally binding in creative
industries versus providing a basis for renegotiation when certain future states of
nature emerge? How much of a role does risk spreading play in contracts in these
highly uncertain markets, especially given artists' many forms of risk-loving behav?
ior? In creative industries where some humdrum enterprises possess elements of
market power, in what forms are these deployed in contracts with talent?
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
Richard E. Caves 83
References
Auletta, Ken. 1997. "The Impossible Busi?
ness." TheNew Yorker.October 6, pp. 50-63.
Baker, Wayne E. and Robert R. Faulkner.
1991. "Role as Resource in the Hollywood Film
Industry."AmericanJournal of Sociology.97:2, pp.
279-309.
Bibicoff, Hilary. 1991. "Net Profit Participations
in the Motion Picture Industry."LoyolaEn?
tertainmentLaw Review.11:1, pp. 23-53.
Carter, Bill. 1998. "Outbid on Pro Football,
NBC Retains 'E.R.' in Record Pact." New York
Times.January 15, pp. Al, A10.
Caves, Richard E. 2000. CreativeIndustries:
ContractsbetweenArt and Commerce.Cambridge,
Mass.:Harvard University Press.
Chisholm, Darlene C. 1997. "Profit-Sharing
versus Fixed-Payment Contracts: Evidence from
the Motion-Pictures Industry."Journal of Law,
Economics,and Organization.13:1, pp. 169-201.
Getzels, Jacob W. and Mihaly Csikszentmihalyi.
1976. The CreativeVision:A Longitudinal
Studyof ProblemFinding in Art. New York:Wiley
Interscience.
Goldberg, Victor P. 1997. "The Net Profits
Puzzle." ColumbiaLaw Review.97:2, pp. 524-50.
Goldman, William. 1984. Adventures in the
ScreenTrade.New York:Warner Books.
Grossman, Sanford J. and Oliver Hart. 1986.
"TheCosts and Benefits of Ownership:A Theory
of Vertical and Lateral Integration." Journal of
PoliticalEconomy.94:4, pp. 691-719.
Hansmann, Henry and Reiner Kraakman.
1992. "Hands-TyingContracts:Book Publishing,
Venture Capital Financing, and Secured Debt."
Journal of Law, Economics,and Organization.8:3,
pp. 628-55.
Hepburn, James. 1968. The Author's Empty
Purse and the Rise of the LiteracyAgent. Oxford:
Oxford University Press.
Kremer, Michael. 1993. "The O-Ring Theory
of Economic Development." QuarterlyJournal of
Economics.108:3, pp. 551-75.
Milgrom, Paul and John Roberts. 1992. Eco?
nomics,OrganizationandManagement.Englewood
Cliffs, N.J.: Prentice-Hall.
Ordonez, Jennifer. 2002. "ArtistsSeek to Ease
Recording-Pact Law." WallStreetJournal. Febru?
ary 25, p. B2.
Passman, Donald S. 1994. All YouNeedtoKnow
About the Music Business. New York: Simon &
Schuster.
Rosen, Sherwin. 1981. "The Economics of Su?
perstars." AmericanEconomicReview. 71:5, pp.
845-58.
Weiler, Paul. 1997. Entertainment,Media, and
theLaw:Text,Cases,Problems.St. Paul, Minn:West.
Weinstein, Mark. 1998. "Profit-Sharing Con?
tracts in Hollywood: Evolution and Analysis."
Journal ofLegalStudies.27:1, pp. 67-112.
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions
84 Journal of Economic Perspectives
This content downloaded on Wed, 20 Feb 2013 10:10:58 AM
All use subject to JSTOR Terms and Conditions