Content uploaded by Roland Kirstein
Author content
All content in this area was uploaded by Roland Kirstein
Content may be subject to copyright.
257
DO ARTISTS BENEFIT FROM
RESALE ROYALTIES?
AN ECONOMIC ANALYSIS OF
A NEW EU DIRECTIVE
Roland Kirstein and Dieter Schmidtchen
ABSTRACT
According to a new European Union directive, artists, whose works are
resold, are entitled to a share of the sales price. The principal aim of this
initiative is to let the artists participate in the economic success of their
work. Our analysis shows that the new directive is most likely to place the
artists in a worse economic position. The analysis of the relation between
the artist and his dealer as an incentive compatible contract leads to further
objections against the new EU directive. However, the paper also illus-
trates under which conditions a resale royalty is, at least, an incentive
compatible contract.
1. INTRODUCTION
Economic theory may be able to provide answers to simple questions law-
makers should ask. The title of this paper is an example of such a simple
question: will the effects of a new EU directive meet the expectations of
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Law and Economics in Civil Law Countries, Volume 6, pages 257–274.
Copyright © 2001 by Elsevier Science B.V.
All rights of reproduction in any form reserved.
ISBN: 0-7623-0712-9
law-makers, or will the outcome be just the opposite? Without a behavioral
theory, it is hardly possible to predict the impact of a new law on the behavior
of the concerned actors. Economic analysis does not promise to be simple, nor
does it promise to give simple answers. However, it provides at least some
answers to lawyers’ questions.
The European Council has recently adopted1the European Commission’s
“Amended Proposal for a European Parliament and Council Directive on the
resale right for the benefit of the author of an original work of art”.2As soon
as the European Parliament agrees to the Councils position, the new EU direc-
tive would require the member states of the European Union to establish resale
rights. This “droit de suite” provides a right for artists to collect a royalty based
on the resale price of their work.3The European Commission’s initiative was
driven by three motives:
(a) to ensure “that creators of graphic and plastic works of art share in the
economic success of their original works of arts” (Recital 2).
(b) “to redress the balance between the economic situation of creators of
graphic and plastic works of art and that of other creators who benefit from
successive exploitation of their work.” (Recital 2).
(c) a better functioning of the internal market in the Union. According to the
Commission, “disparities with regard to the existence and application of
the artist’s resale right by the Member States have a direct negative impact
on the proper functioning of the internal market in works of art” (Recital
7). The application or non-application of such a right is considered to be
a factor that contributes to distortions of competition as well as displace-
ments of sales within the Community (Recital 6).
The Commission expects about 250,000 artist to benefit from the introduction
of a resale royalty in the EU.4Currently, Belgium, Denmark, France and
Germany enforce resale rights regularly. Droit de suite also exists in Greece,
Italy, Luxembourg, Portugal and Spain, but there is no enforcement on a regular
basis.5Britain, the country with the most important art market in Europe, does
not have such a legislation, as it is the case in Ireland, Austria, and the
Netherlands; these EU member states opposed until recently the proposal.6In
March, the Council has agreed upon introducing the new directive, supported
by the votes of 13 member states, whereas Austria and Belgium did abstain,
see FAZ (2000). Since then, some modifications of the Commission’s initiative
have been made in order to achieve consent. In the United States, only California
has adopted a droit de suite.7
In this paper, we take a closer look on the first of the Commission’s above
mentioned motives. Our analysis is based on the idea that a dealer has three
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
258 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
different economic functions which provide the sources of gains from trade
between the artist and the dealer.8
(a) the dealer may have better access to capital markets, i.e. pay a smaller
interest rate, than the artist and therefore serves as an intermediary for
cheap credits;
(b) the dealer may (perhaps as well as the artist) spend effort on the enhance-
ment of consumer valuation in the resale market for the work of the artist;
(c) if the artist is risk-averse, the dealer may provide insurance.
The impact of the introduction of a droit de suite on the dealer’s economic
functions is analyzed separately in the subsequent sections. Section 2 focuses
on the function of the dealer as a credit intermediary, and therefore excludes
risk-aversion and effort considerations. It is assumed that the dealer’s discount
factor is higher than the artist’s, reflecting the lower interest rate the dealer
pays to obtain credit. We show that the artist’s lifetime income is always
decreased when resale royalties are introduced.
A similar result was derived by Karp and Perloff (1993). Their model neglects
the discounting of future payoffs, whereas in our model discounting plays an
important role. Furthermore, the Karp and Perloff (1993) model places the artist
in a surprisingly strong market position: he sets the initial market price and the
resale royalty, whereas the dealer only chooses promotional effort and the resale
price. Our analysis goes beyond this in taking into account that artists may have
a weaker position in the initial market.
In Section 2, the analysis is limited to the distribution of a given cooperation
rent, an aspect that was particularly stressed in the EU initiative. This neglects
the incentive effects a droit de suite may have. Section 3 takes into account
that a droit de suite may affect the two parties’ effort to promote the value of
the artist’s work. Promotional effort may increase the cooperation rent that can
be distributed between the two parties. First, we follow Karp and Perloff (1993)
and assume that the future valuation of the artist’s work depends only on promo-
tional effort of the dealer. We show that the optimal resale royalty is zero. An
argument in favor of a resale royalty (“share-cropping”) contract can only be
derived if customers’ valuation in the resale market depends on the efforts of
both artist and dealer. However, it would require additional reasons to justify
a mandatory droit de suite, since artists and dealers are free to agree upon resale
royalties by contract.
Section 4 introduces risk-aversion on the part of the artist. In such case, the
introduction of a droit de suite would have two effects on his situation: it forces
the artist to accept a risky lottery instead of a sure income; and secondly, even
if a resale royalty had a positive net effect on the artist’s monetary lifetime
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 259
259
income due to its incentive effect, his lifetime utility may be decreased. This
effect, which we call the “paradox of risk-aversion”, can occur if the income
of the artist increases over time. Section 5 briefly summarizes the arguments,
and discusses in which direction legal harmonization in Europe should reason-
ably move.
2. RESALE RIGHTS AND LIFETIME INCOME
2.1. Initial Market and Resale Market
Droit de suite intends to let the artist participate in the future economic success
of his work. In some cases indeed, the resale price largely exceeds the price
that was initially paid by the dealer. A widely held view of the art business is
the image of a
collector who, having purchased a work of art for relatively little, resells it for a great deal
more, pocketing the entire profit and leaving the artist whose effort created the work and
whose subsequent accomplishments may have contributed to ist increase in value, with no
part of such increase. It is the image of Robert Rauschenberg and Robert Scull in tense
confrontation after the 1973 auction at which Scull resold for 85,000 Dollars, a work for
which he had originally paid Rauschenberg less than 1,000 Pounds.9
However, there is a simple economic reason for the enormous differences in
the prices at which dealers buy and sell art: from the dealers point of view, the
works of young and unknown artists are lotteries. He will frequently be unable
to resell the pieces later, and of those sold only a few will be worth large sums.
Consider a simple example: a typical dealer buys one hundred pieces of art,
ten of which will be worth 100,000 some years later. The others will yield
no considerable return. Neither the dealer, nor the artists know the future value
of a certain piece of art. Assume both parties to be risk-neutral. The interaction
between dealer and artist, concerning one piece of work, takes place in three
steps (for simplicity, assume that the piece is resold only once):
(1) the dealer buys a piece of art from the artist (initial market);
(2) nature reveals the artist’s true type (i.e. the valuation of the collectors);
(3) the dealer sells the piece to a collector (resale market).
The initial market price is governed by three factors: the dealer’s maximum
willingness to pay, the artist’s minimum willingness to accept, and the market
conditions. Let us first derive the dealer’s maximum willingness to accept. We
assume the dealer to be aware of the fact that only 10% of all purchases will
turn out to be valuable in the final market.10 Hence, when considering his
situation in the initial market, he expects an average resale price of 10,000
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
260 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
in the resale market for each of the pieces he buys. Assume furthermore that
the dealer discounts future gains by 20%. This discount factor of 0.8 reflects
the interest rate and the time until the resale is realized.11 Thus, the dealer’s
present value of this portfolio is 8000, which is the maximum amount the
dealer would be willing to pay.12
The artist will accept a price in the initial market if it exceeds the present
value of the expected return the artist earns if he stores the work until he sells
it on his own.13 Let us assume that the artist’s discount factor is 0.6, thus smaller
than the dealer’s. This reflects the fact that the dealer may have better access
to the capital market (a lower interest rate means a higher discount factor).
Additionally, young artists will only very occasionally be able to provide
collateral if they apply for a credit directly. The artist’s minimum willingness
to accept is the discounted value of the average return in the resale market,
6000.
The difference between the dealer’s 8000 and the artist’s 6000 is the
cooperation rent that is created if the dealer and the artist conclude a contract.
If the initial market is competitive, then the elasticities of demand and supply
govern the distribution of this cooperation rent by the initial market price.
In bilateral monopolies, it is the relative bargaining power of the parties that
determines the price and thereby the distribution of the cooperation rent. Thus,
the split rate could be a result of a competitive market process as well as of
bilateral monopoly negotiations and is assumed to be independent of whether
a droit de suite is introduced or not. This approach allows us to model the
effects on prices without having to explicitly model the market structure.14
A split rate close to 1 could be interpreted as the market situation of a young
artist who is already well established (in economic terms: whose work is hetero-
genized), and who has some market power in the inital market. This would be
the case if dealers outbid each other to come into business with this artist. On
the other hand, a split rate close to zero refers to the situation of a completely
unknown young artist whose work, from the dealer’s point of view, is a perfect
substitute to the work of other artists. In this case, the artists outbid each other
to get access to a dealer.
For the sake of simplicity, we assume that the cooperation rent is equally
shared among the dealer and the artist. Therefore, the initial market price,
without a droit de suite is calculated as
0.5 8000 + 0.5 6000 = 7000 (1)
The cooperation rent of 2000 is split equally between the two parties. Each
one gets 1000 of the cooperation rent. This is realized by an initial market
price of 7000.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 261
261
2.2. Introduction of a Droit de Suite
If a droit de suite is established, either by mandatory law or by contract, then
the artist holds a right to a share
<1of the resale price. A resale royalty is
a kind of excise tax on art that must be paid each time the piece is resold.15
It can also be seen as an attenuation of the buyer’s property rights. Both
points of view allow for the same prediction: the buyer’s maximum willing-
ness to pay is decreased. A rational dealer will discount the expected resale
royalty and subtract its present value from the amount he is willing to pay if
no droit de suite were in place. We assume that the dealer (and not the collector
who buys the work in the resale market) must bear this “excise tax”.16 Thus,
the maximum willingness to pay drops to (1
)8000.
The artist compares two options: to store the work yields 6000, which is
the expected return, weighted with the artist’s discount factor. If he uses this
option, then the droit de suite plays no role. On the other hand, the artist may
sell his work to the dealer. Let pdenote the initial market price paid by the
dealer if a droit de suite is in place. The artist not only receives the current
return p, but also the claim on the resale royalty. However, this additional
income is uncertain and only due in the future; it therefore must be discounted.
The present value of this claim is 6000
. Hence, an initial market price
pis acceptable to him if p+ 6000 is greater than 6000. Stated equiva-
lently, the artist accepts a price pif
p>(1
) 6000 (2)
The right hand side of inequality (2) denotes the artist’s minimum willingness
to accept if a droit de suite is introduced. Calculating the market price pthe
same way as in equality (1) leads to
p=0.5 . (1
)8000 + 0.5 (1
)6000 (3)
since the split rate is assumed to be independent of whether a droit de suite is
introduced or not. Thus, pequals (1
)7000. The introduction of a droit
de suite, therefore, leads to an initial market price which is necessarily smaller
than the one without resale royalties.17
The difference in market price with and without droit de suite is
7000.
The artist’s return at the time he sells his piece of art to the dealer is reduced
by this amount. The net effect of a droit de suite on the present value of the
artist’s lifetime income is the present value of the resale royalty minus the
decline in the initial market price:
6000 7000 = 1000 (4)
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
262 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
Under the assumptions we have made, this net effect is always negative. Note
that our result not only proves true for the example we used, but also for any
other split rate or other resale market returns.18 The net effect would be positive
if, and only if, the artist’s discount factor would exceed the dealer’s. However,
in such a case, a cooperation rent would cease to exist: gains from trade can
only occur between the artist and the dealer if the discount factor of the latter
exceeds that of the former.19 If this is the case, then the droit de suite puts the
artist definitely into a worse position. Hence, the EU directive not only fails to
achieve its most important goal, to let the artists participate in the economic
success of their work, but leads to the opposite result.
Note that the artist loss increases in the split rate.20 The better his position
in the initial market, the more the artist looses when a resale royalty is intro-
duced. Hence, if a droit de suite is established, successful artists have more to
lose. This theoretical result explains the observation of Parachini (1990, F3),
according to which 40 established artists (among them Wilem de Kooning and
Roy Lichtenstein), in 1988, opposed a U.S. Congressional initiative to introduce
a resale royalty in all U.S. states. According to the report, the artists’ reason
for their opposition was that a droit de suite “might make it even harder for
unknown artists to attract the interest of collectors”. In the light of our analysis
this is correct, yet appears to be somewhat hypocritical.21
3. INCENTIVE COMPATIBLE CONTRACTING
3.1. Dealer’s Effort
The analysis of the relationship between an artist and a dealer within a simple
market framework might be seriously misleading. Often the relationship has a
pooling character, like team production, which can best be analyzed using the
concept of relational contracts. According to this concept the artist and the dealer
are playing a repeated game in good faith. Neither is trying to gain an advantage
at the expense of the other side. Rather, both see themselves as members of a team
acting co-operatively in order to increase the economic value of the relationship.
A dealer incurs substantial costs in buying, owning, conserving and promoting
sales. He has usually tied up specific investments in the work of art, expecting
a reasonable rate of return or, in economic terms, a “quasi rent”. If resale
royalties reduces this quasi rent, the incentives of the dealers may be affected
in a way that harms the artist.
As in Karp and Perloff (1993), we first analyze the situation in which only
the promotional effort of the dealer is relevant for the valuation by the customers
in the resale market. Let us, for simplicity, distinguish two cases:
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 263
263
(a) the dealer does not spend additional effort, hence the average resale price
remains 10,000;
(b) or the dealer undertakes additional effort to promote the value of the artist’s
work.
Assume that the additional effort costs the dealer an amount of e, and increases
the average value of the artists work to, say, 15,000.22 Thus, high effort
increases the average return in the resale market by 5000. In this section, we
neglect discounting of future payoffs and focus on effort only, hence the average
resale price reflects the parties’ maximum willingness to pay and minimum
willingness to accept, respectively, in the initial market. To analyze the impact
of promotional effort on the value of the artist’s work in the resale market, we
have to distinguish two concepts: the first-best effort and the individually
rational decision.
The first-best effort maximizes the common profit of the dealer and the artist,
disregarding the possible conflict between the two parties. It would be first-best
to spend high promotional effort if the increase in valuation in the resale market
exceeds the effort costs. Hence, it is first-best to spend high effort if, and only
if,
e< 5000 (5)
The individual decision of the dealer is not necessarily based on this compar-
ison. The dealer spends high effort if his individual share of the return exceeds
his effort costs. Recall that, if a droit de suite is established, the dealer only
receives a share (1
)of the return in the resale market. Therefore, he is moti-
vated to spend high effort if, and only if,
e<(1
)5000 (6)
A comparison of the inequalities (5) and (6) leads to the following results:
(a) if e> 5000, then it is efficient not to spend high effort, and the dealer
makes the efficient choice;
(b) if e<(1
) 5000, then it is efficient to spend high effort, and the dealer
makes the efficient choice;
(c) if eis in between (i.e. (1
)5000 <e< 5000), then it would be
efficient to choose high effort, but the dealer is not motivated to do so and
chooses low effort instead.
The interval in which the dealer behaves inefficiently vanishes if
= 0. Thus,
from an efficiency point of view, the optimal mandatory resale royalty would
be
= 0. This is not only in accordance with the result in Karp and Perloff
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
264 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
(1993, 165),23 but also with the standard result of contract theory, given that
both the artist and the dealer are risk-neutral. In such a situation, a “sell the
shop” contract would be optimal: the dealer pays a fixed amount to the artist
and becomes residual claimant. In contrast to this, a mandatory droit de suite
would force the parties into a “share cropping” contract, which induces the
dealer to choose a suboptimal low effort.24
A “sell the shop” contract allocates the entire risk to the dealer. If he were
assumed to be risk-averse, whereas the artist is risk-neutral, then a reallocation
of risk would be efficient. In such a case, a resale royalty could be an efficient
contract. This, however, does not seem to be the world the EU Commission
had in mind when making its proposal.
3.2. Artist’s Effort
Let us turn to the second scenario: the expected return on the resale market
depends on the effort of the artist rather than of the dealer. An artist has several
ways to increase the value of a piece he has already sold to a dealer, such
as producing additional pieces of high quality, investing in his skills,
seeking further inspiration, and making contacts with potential buyers and
intermediaries.25
In principle, the artist’s effort may have a substitutional or a complementary
effect on the expectedreturn for his earlier work. Here, we consider the comple-
mentary effect: future work of high quality is assumed to increase the valuation
of the life’s work of the artist.26 If the artist spends high effort, bears costs
f> 0, whereas in the case of low effort his costs are 0. Let us again assume
that this increase amounts to 5000. Thus, it is first-best to spend high effort
if, and only if, f> 5000. From the artist’s point of view, it is individually
rational to spend high effort if his share of the increased return (namely
5000) exceeds the effort costs f. Just as in the case above, we have three
cases to distinguish:
(a) if f> 5000, then the artists abstains from high effort, which is efficient
to do so;
(b) if f< 5000
, then the artists chooses high effort, which is efficient to
do so;
(c) if
5000 < f< 5000, then high effort would be efficient, but from the
artist’s point of view it is better to choose the inefficient low effort.
The interval in which the artist may behave inefficiently vanishes if
= 1, which
implies that the artist not only receives a share of the resale price, but the whole
return. In such a setting, a dealer is not necessary, and the entire risk should
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 265
265
be borne by the artist. It would be optimal if the artist simply stores his pieces
and waits until they have gained their resale value. If, however, the artist were
assumed to be risk-averse, then it would be inefficient to let him bear the entire
risk.27
In this case, a contract should balance the artist’s desire for insurance against
the motivational effect of the residual claim. However, such a contract would
not lead to the first-best solution, which is not attainable due to the risk-aversion
of the artist. A resale royalty may at most be second-best: the artist sells the
main part of the residual claim in turn for an up-front payment to the dealer,
yet the prospect of receiving a share of the uncertain resale returns keeps the
artist motivated to spend at least some effort on the enhancement of these
returns.
3.3. Both parties’ efforts
Things are more complicated when the expected resale market return depends
on the efforts of both the artist and the dealer. First of all, the contract should
deal with the fact that now both parties may have an incentive to spend less
than the optimal effort (double-sided moral hazard). Secondly, the parties’
efforts may influence each other. In particular, they may have a complementary
or a substitutional effect. We consider the complementary effect here: future
work of high quality does not only tend to increase the value of the life’s work
of the artist, but also increases the marginal effect of the dealers effort (and
vice versa).28
To keep matters simple, we limit the choices of the parties to high and low
effort. Let us denote the high effort of the dealer as e=1 and of the artist as
f= 1, whereas e=f=0represents both parties’ low effort. The expected return
on the artist’s work is denoted as q = q(e+f). Hence, three levels of q must be
distinguished: q(2), if both parties contribute; q(1) if only one party spends
effort; q(0), if neither party contributes.
The marginal effect of either party’s effort on the expected return is positive,
thus q(2) > q(1) > q(0). The two party’s efforts are strategic complements, hence
the marginal effect of one actors contribution is higher if the other has made
a contribution as well: q(2)
q(1) > q(1)q(0).
The parties may bear different costs for high effort: let us denote the artist’s
cost as cand the dealer’s as k. If both spend effort, the (expected) social net
surplus of the two parties is q(2)
c
k, which we assume to be positive and
greater than q(0): it would be efficient if both were spending effort. However,
this is not necessarily the outcome if the parties maximize their own (expected)
payoff. Table 1 shows the interactive decision situation as a game in strategic
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
266 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
form. Recall that the artist’s share of the expected return is denoted as $$,
hence the dealer’s share is (1
).
The efficient outcome (e = 1, f = 1) is a Nash equilibrium29 if, and only if,
the following relations hold simultaneously:
(a)
q(2)c>
q(1); this condition guarantees that f=1is the best reply
for the artist if the dealer chooses $e = 1$, and
(b) (1
)
q(2)
k>(1
)q(1), which implies that e=1is the best reply
to f=1.
Both conditions say nothing more than the individual costs of choosing high
effort should be smaller than the individual gain, given the other side chooses
high effort as well. The two conditions can be equivalently expressed as
c<
[q(2)q(1)]<q(2)q(1)k(7)
If the parties agree upon a value of
that satisfies condition (7), then this
contract turns the efficient outcome into a Nash Equilibrium.30 Note that the
interval in (7) is non-empty if, and only if, q(2)
q(1) >c+k. If this condition
is violated, then there is no value of rthat turns (e = 1, f = 1) into a Nash-
equilibrium.
A numerical example might help to understand the meaning of condition (7):
let q(2) = 100,000, q(1) = 10,000 and q(0) = 0. This satisfies the comple-
ment property, since 100,00010,000 > 10,0000. Let furthermore be
c= 9000 and k= 45,000, which satisfies the efficiency condition
($100,00045,0009000 > 0$). Table 2 shows the decision situation in this
example (the figures are in 1000).
Using these parameters, condition (7) translates to 9000/90,000 <
< 45,000/
90,000, or 0.1 <
< 0.5: In this example, a very low value of
implements
efficient behavior as a Nash equilibrium.31 This is due to the cost structure:
the dealer’s costs were assumed to be nine times as high as the artist’s costs.
The low share for the artist is not only efficient, but also reflects fairness
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 267
267
Table 1. The Effort Game.
considerations: the party that has to bear the higher costs should get the greater
share of the output.
4. RISK-AVERSE ARTISTS
Until now it was assumed that the actors are risk-neutral. It has been shown in
the previous sections that, leaving aside incentive effects, the introduction of a
droit de suite tends to harm the artists. If, on the other hand, both sides’ efforts
to promote the resale value of the work are relevant, a resale royalty can be
the incentive compatible contract, as it was shown in Section 3.3, and may lead
to a higher monetary income for the artist.
The assumption of risk-neutrality will be maintained in this section with
respect to the dealer because he is able to spread his risk over a portfolio of
different assets. The individual artist, however, relies on only one source of
income, especially if he is completely devoted to his work. Therefore; it seems
reasonable to assume that artists are risk-averse.32
Risk aversion is modeled by a concave utility function. The marginal utility
of income is positive (more money brings more utility), but the rate at which
an additional unit of income increases the utility is diminishing. There are two
reasons why this kind of utility function makes it more likely that a droit de
suite is harmful for the artist. First of all, resale royalties oblige the artist to a
lottery that can adversely affect his utility if he is risk averse. It could be bene-
ficial to receive a higher (and secure) initial fixed income rather than a lower
one which is combined with a risky payment in the future.33 If the artist had
the right to waive the resale right (and the dealer would agree), the artist could
buy himself out of the lottery. However, a waiver is not permitted under the
EU directive.
Additionally, if the artist is unable to obtain credit against his claim, the droit
de suite shifts income from his youth to his older age. The artist is forced to
accept a lower income now, in exchange for some uncertain future gain, shifting
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
268 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
Table 2. Example Effort Game.
income to a perhaps more prosperous stage of his life.34 This transfer of a part
of the artist’s current income to the future may be harmful even if the present
value of the lifetime income stream is increased.35
A numerical example should make clear this “paradox of risk-aversion”:
Assume that a young artist has an annual income of 16.900. The introduction
of resale royalties leads to a decrease of, say, 2.500. His new annual income
is then 14.400. Making use of a concave utility function like U(x) = 0.5,
where xis the income and Uthe utility, the artist gets 130 utility units before,
and 120 utility units after the introduction of the droit de suite: thus, the droit
de suite leads to a current loss of ten utility units.
Assume now that, in his later life, the artist has an annual income of
90 000 and the resale royalty brings him 3.025 as additional income.36 This
is equivalent to 300 utility units without the droit de suite, and 305 utility units
after the introduction of it. The introduction brings an additional utility of
5 units. These figures are displayed in Table 3, where the label “no dds” stands
for the situation without a droit de suite, whereas “with dds” denotes the
situation if a resale royalty is introduced.
Adding up the income as well as the utility in the two years under consid-
eration, the artist earns 106.900 without droit de suite, and 107.425 if a
droit de suite is introduced; hence, the resale royalties increase his monetary
income by 525. However, his lifetime utility is decreased: without droit de
suite, overall utility adds up to 430 utility units, whereas with resale royalties
it is only 425 utility units. The utility loss at a younger age due to the intro-
duction of a droit de suite exceeds the utility gain from the resale royalty.
The economic reason for this is straightforward: Having an additional in
a situation where the income is low can bring much more utility than having
this additional in a situation where the income is already high. According to
our example, this even holds true when discounting of future income and utility
as well as the riskiness of the resale royalty is neglected. If future income were
subject to discounting, this effect would be even greater.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 269
269
Table 3. Lifetime Income and Utility.
5. CONCLUSION
The aim of this paper was to analyze whether the introduction of a mandatory
droit de suite will make artists better off. The answer is, to say the least, unclear.
Leaving the effort issue aside and focusing only on the distribution of a given
expected return on the resale market, the droit de suite clearly places the artists
in a worse position. Taking the incentive effect into account, three cases must
be distinguished:
(a) a resale royalty would be counterproductive if only the dealer’s effort is
required to promote the value of the artist’s work. However, the droit de
suite might be at least second-best if the dealer were assumed to be risk-
averse, whereas the artist is risk-neutral.
(b) If, on the other hand, only the artist’s effort is relevant to promoting the
value of his work, then he should be the residual claimant. The artist should
circumvent the dealer and sell his work on his own behalf.
(c) If both the artist’s and the dealer’s efforts may increase the expected return
in the resale market, then a share cropping contract can implement efficient
behavior.
However, even if a resale royalty forms part of a contract on the initial market,
an increase in the artists’ lifetime income does not necessarily mean that their
position is improved. If they are risk-averse, they are likely to lose. First of all,
a droit de suite increases the volatility of their lifetime income. Additionally,
the income shift from younger to older age induced by resale royalties can
decrease the utility derived from the artist’s lifetime income even if the monetary
value of his income is higher than without a droit de suite.
Even if a resale royalty were efficient, this would not yet justify a mandatory
droit de suite. The artist and the dealer would have an incentive to do this
voluntarily. Such a contract may yet be difficult to specify or to enforce.37
Prohibitively high transaction costs are thus a standard rationale for mandatory
legislation;38 in this case the law may sensibly provide a standard contract. If
droit de suite is inefficient, but distributes the smaller “cake” in a way that is
more favorable for the artists, this might provide another reason for mandatory
legislation.
However, there may arise practical difficulties in enforcing the proposed legis-
lation.39 Dealers may try to circumvent the droit de suite by selling works
of art in a jurisdiction where no resale royalties apply. This would prevent
the dealer’s maximum willingness to pay from decreasing, but nevertheless
Europe would lose a share of the art market.40 Furthermore, one should carefully
determine whether Europe is the optimal legal area for the issue in question.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
270 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
If not, then harmonization itself is not a reasonable goal. Even if one agrees
that Europe is the optimal legal area, then the additional question arises towards
which direction harmonization should take place.41 One should keep in mind
that the droit de suite provides the greatest losses for the most talented artists.
Taking this into account, harmonization should rather be directed towards the
abolition of the droit de suite in the countries where it exists than towards
the introduction of it in the other countries.
NOTES
We are grateful to Brandon Ahearn, Gerrit de Geest, Lewis Kornhauser, Ejan Mackaay,
and other participants of the conference on “L’économie du droit dans les pays de droit
civil” in Nancy, June 2000.
1. See FAZ (2000).
2. OJ (1998).
3. Perloff (1998, 645). Taken literally, this principle would give the artist a claim on
a share of the resale price even if it is lower than the initial price. Karp and Perloff
(1993, 167f.) discuss resale royalties based on the difference between the resale and the
initial price (“capital gains”), however without deriving a systematic effect of the two
regimes on the artist’s income. See also Bolch, Damon and Hinshaw (1988, 75).
4. EU Commission (1999).
5. See O’Hagan (1998, 86), and FAZ (1999).
6. FAZ (1999).
7. See Karp and Perloff} (1993, 163), who also mention Algeria, the former
Czechoslovakia, Chile, Poland, Sweden, Tunisia, and Uruguay to provide such guarantees
– however, in most countries the artists do not actually receive resale proceeds.
8. In Kirstein and Schmidtchen (2000), an additional function is analyzed: an expert
dealer creates value by his ability to tell good from less talented artists.
9. Weil (1983), c.f. O’Hagan (1998, 100, Fn. 18).
10. In Kirstein and Schmidtchen (2000), we take a closer look on expert dealers that
are able to distinguish between good artists and less talented ones.
11. If the annual interest rate is denoted as iand the resale takes place tyears after
the initial purchase, then the discount factor for the future returns is 1/(1+i)t. Obviously,
the discount factor is higher if the interest rate is lower.
12. If the dealer had operating costs, these would further decrease his maximum
willingness to pay, but the effect of the droit de suite would qualitatively be the same.
13. O’Hagan (1998, 90) remarks that the artist “always has the possibility of holding
on to his/her work for investment purposes, if there is reason to believe that the work
will appreciate”. Only for simplicity we assume the artist produces at zero marginal
costs.
14. Karp and Perloff (1993, 165) also analyze the case of an artist-owned gallery.
In terms of our model, this kind of vertical integration would be reflected by a split
rate of 1 (the artist internalizes the whole cooperation rent) and equal discount
factors. Obviously, the dealer-artist cannot be made better off by introducing a droit de
suite.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 271
271
15. The EU Commission rejects the view that resale royalties are “just a further tax”,
since the proceeds go to the artist, not the state, see EU Commission (1999). However,
the economic effect on the resale market is just the same as if a tax were imposed.
16. This assumption seems to be rather reasonable when the supply on the resale
market is inelastic, which is the case in our model: the amount traded on the initial
market is the constant supply on the resale market.
17. See Perloff (1998, 645), Karp and Perloff (1993), Hansman and Santilli (1996, 69).
18. See Kirstein and Schmidtchen (2000).
19. For the moment, we have left aside other economic reasons for cooperation rents,
such as insurance, screening, or promotional effort. Insurance and promotion are analyzed
in the subsequent sections.
20. This is proven in Kirstein and Schmidtchen (2000).
21. The bill has not passed Congress, see Karp and Perloff (1993, 163). Landsburg
(1989) argues in the opposite direction: a droit de suite may shift income from less
known to successful artists. However, Solow (1990) rejects this argument, see Karp and
Perloff (1993, 174f.), endnote 6.
22. In principle, this increase can have two reasons: the resale value of the artists
work is increased, or the probability that an artists turns out to be successful is increases
(or both).
23. Even though these authors do not perform a rigorous derivation of optimal
contracts, they come to the result that resale royalties reduce the dealer’s incentives to
spend promotional effort.
24. This holds in general if the dealer’s choice set is continuous; if he chooses his
effort from a discrete set, such as high or low, share cropping could be optimal as well.
However, “sell the shop” would still be optimal in this case and nothing is gained by
a mandatory resale royalty.
25. The impact of the artist’s activities on the value of a piece of art sold is not taken
into consideration in Karp and Perloff (1993).
26. The substitutional effect would be reflected by the idea of Coase (1972): the later
output of a monopolist may be seen as a substitute for his earlier works which tends to
decrease the value of each piece of work. See Karp and Perloff (1993, 169) and
Schmidtchen, Koboldt and Kirstein (1998, 789).
27. We take a closer look on risk-aversion and droit de suite in the next section.
28. The resulting interactive decision-making situation is called a “supermodular
game”, see Fudenberg and Tirole (1992). In case of a substitutional effect, the marginal
effect of one party’s effort on the expected return is diminishing in the effort chosen
by the other party.
29. A strategy combination is a Nash equilibrium if no player has an incentive to
deviate from it.
30. It is possible that (e=0, f=0) as well is a Nash equilibrium; a case which is
not of interest for our analysis here.
31. In fact, four equilibrium constellations are possible, depending on the value of
(and neglecting ties):
(1) 0 < <0.1: The strategy f=0is dominant and e=0is the best reply.
(2) 0.1 < <0.5: Both e=f=1and e=f=0are Nash equilibria.
(3) 0.5 < <0.9: f=0is dominant and e=0is the best reply.
(4) 0.9 < <1: The stategies e=1and f=0are dominant.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
272 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
Note that 0.1 < < 0.5 implements efficient behavior as one out of two Nash equilibria.
32. A decision-maker is risk-averse if he prefers not to take part in a “fair” lottery.
A lottery is called fair if the ticket price equals the expected gain. Consider a lottery
that pays 100 with probability 0.1 and nothing otherwise. Its expected value hence is
10. If the lottery is fair, then a risk-neutral decision-maker would be indifferent whether
to participate or not. A risk-averse decision maker, on the other hand, would strictly
prefer to keep the ticket prize. He would take part if he had to pay less than the expected
gain. If the risk-averse is indifferent between the lottery and a ticket price of, say, 7,
then the difference of 3 is the “risk premium” this decision-maker wants to earn before
trading in the ticket prize for the prospect on an uncertain return.
33. See Karp and Perloff (1993, 171).
34. O’Hagan (1998, 89).
35. This argument was presented in Schmidtchen, Koboldt and Kirstein (1998).
36. According to Filer (1986), young artists earn less, but show a steeper lifetime
income profile than the average workers of the same age.
37. A very simple way – which is completely free of enforcement costs – for an
artist to guarantee participation in the increasing value of his work would be to withhold
a few pieces.
38. Karp and Perloff (1993), Hansman and Santilli (1996), and Schmidtchen, Koboldt
and Kirstein (1998, 790) also discuss reasons like paternalism or asymmetric information.
However, these reasons appear to be less convincing, as long as perfect rationality is
assumed. For a general discussion of inalienability, see e.g. Rose-Ackerman (1985).
39. See O’Hagan (1998, 88f.).
40. In EU Commission (1999) it is argued that the introduction of a droit de suite
would not lead to the loss of “thousands” of jobs.
41. The EU commision complains that the works of painters like David Hockney
enjoy much weaker protection than those of the Spice Girls, see FAZ (1999) and EU
Commission (1999). Therefore, harmonization is claimed to be necessary not only among
the member states of the EU, but also between groups of artists.
REFERENCES
Bolch, B. W., Damon, W. W., & Hinshaw, C. E. (1988). Visual Artist’s Rights Act of 1987: A
Case of Misguided Legislation. In: Cato Journal, 8(1), 71–78.
Coase, R. (1972). Durability and Monopoly. In: The Journal of Law and Economics, 15, 143–149.
EU Commission (1999). Proposed Directive on Artist’s Resale Right – Clarification; document
dated Dec. 14th, 1999, downloaded on June 23rd, 2000 at http://europa.eu.int/comm/
internal\_market/en/intprop/99–68.htm
FAZ (1999). Zwist um Folgerecht für Künstler. Finnische EU-Ratspräsidentschaft drängt Briten zu
neuen Vorschlägen. In: Frankfurter Allgemeine Zeitung, 255, 2.11.1999, 18.
FAZ (2000). Zementiert. Neue Richtlinie zum Folgerecht. In: Frankfurter Allgemeine Zeitung, 66,
18.3.2000, 51.
Filer, R. K. (1986). The ‘Starving Artist’ – Myth or Reality? Earnings of Artists in the United
States. In: Journal of Political Economy, 94, 56–75.
Fudenberg, D., & Tirole, J. (1992). Game Theory. MIT Press, Cambridge, London, 2nd part.
Karp, L. S., & Perloff, J. M. (1993). Legal Requirements that Artists Receive Resale Royalties.
In: International Review of Law and Economics, 13, 163–177.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
Do Artists Benefit from Resale Royalties? 273
273
Kirstein, R., & Schmidtchen, D. (2000). Do Resale Royalties Make Artists Better Off? An Economic
Analysis of a New EU Directive; Center for the Study of Law and Economics, Discussion
Paper 2000–07, Saarbrücken.
Landsburg, S. (1989). Price Theory and Applications. The Dryden Press 1989.
O’Hagan, J. W. (1998). The State and the Arts. An Analysis of Key Economic Policy Issues in
Europe and the United States. Celtenham: Edward Elgar.
Hansmann, H., & Santilli, M. (1996). Author’s and Artist’s Moral Rights and the Droit de Suite:
A Comparative Legal and Economic Analysis; Paper presented at the 5th Symposium of
Law and Economics, Travemünde, 27–30.3.96.
OJ (1998). Official Journal of the European Union 125/8, 23. 4. 1998.
Parachini, A. (1990). Artist’s Rights Bill Awaiting Bush’s Signature. In: Los Angeles Times, Nov.
6th, F3 (cit. Karp/Perloff (1990, 174)).
Perloff, J. M. (1998). Droit de Suite. In: P. Newman (Ed.), The New Palgrave Dictionary of
Economics and the Law (pp. 645–647). London/Basingstoke/New York: MacMillan.
Rose-Ackerman, S. (1985). Inalienability and the Theory of Property Rights. In: Columbia Law
Review, 85, 931ff.
Schmidtchen, D., Koboldt, C., & Kirstein, R. (1998). Rechtsvereinheitlichung beim “droit de suite”?
Ökonomische Analyse des Richtlinienentwurfs der Europ¤ischen Kommission. In: B.
Großfeld, R. Sack, T. M. J. Möllers, J. Drexl & A. Heinemann (Eds), Festschrift für
Wolfgang Fikentscher. Mohr Siebeck, Tübingen.
Solow, J. L. (1992). An Economic Analysis of the Droit de Suite. University of Iowa Working
Paper.
1
2
3
4
5
6
7
8
9
1011
11
12
13
14
15
16
17
18
19
2011
21
22
23
24
25
26
27
28
29
3011
31
32
33
34
35
36
37
38
39
40
274 ROLAND KIRSTEIN AND DIETER SCHMIDTCHEN
