Market failure in the market for art : characterisation of the art market, a literature review

Market Failure in the Market for Art

Characterisation of the Art Market, a Literature Review

Michiel Nivard

5941776

Economics

michiel.nivard@student.uva.nl

Supervisor

prof. dr. M.P. Schinkel

ABSTRACT

The specific characteristics of the art market give rise to some typical market

failures. This has led to an inefficient market for art causing anomalies to

appear.

1

Table of Content

1. Introduction ... 2

2. Market Analyses ... 4

3. Stardom ... 7

4. Role of the Artist ... 12

Monopoly aspect ... 12

Death Effect ... 13

5. Art return vs. Stock Return... 15

6. Conclusion ... 19

7. Further Research ... 21

2

1.

Introduction

‘Going once, going twice. Sold!’. Hearing these expressions, one immediately recognises that an auction of some kind is being held. In many economic sectors auctions are the common market place as buyers are concentrated in one area to bid directly on a displayed item that is being offered. In such cases the entire marketplace (buyers, sellers, products, infrastructure, etc.) is compressed in one single room. Examples range from the tangible flower business and livestock trade to the abstract stock markets. These markets appear to have a very clear mechanism of demand and supply and should thus be of great significance to economists for investigatory purposes. Nevertheless, even in these seemingly transparent markets, market failures exist and markets might not always be as transparent as they appear.

One of the most interesting and pure forms of auction takes place in the market for art. In this market an artist or art owner come together with potential buyers with the purpose to reach each other in a potential sales price of the offered art piece. This usually happens through the mediation of galleries, auction houses and other middlemen who play a part in the alignment of seller and buyer. As transparent and open these auctions seem to be, in fact they are complex and murky, which might result in some market failures that disturb the market for art. The investigated hypothesis will be ‘the art market is an inefficient market subjected to market failure’. It is important to assess whether an economic market is subjected to market failure as this will lead to loss of consumer and producer surplus. The overall utility cannot be maximized in these markets and that makes them inefficient. This thesis will be a literature review regarding several possible market failures subjected to the art market and discuss its problems with respect to the valuation of art in both the economic and aesthetic aspect.

The art market has always been a ‘big-money game’ where enormous sums of money are spend on individual art pieces. Perhaps because art is a secondary good and education plays a big role in buying and appreciating art, the average buyer has a wealthy background. This is reflected by the average price for so-called masterpieces that have always been very expensive compared to other goods. Recent trends have shown an increase in the average art prices as even though the world is suffering from a financial crisis art auctions are setting new revenue records with the latest record held by Christie’s is $495 million in one auction (Vogel, 2013).

This record is composed by many individually sold paintings of which the prices reflect a combination of the buyer’s preference and available income. But the prices also reflect the value of the painting; when Pollock’s Nr. 19 is sold for $58.3 million it is said to have this value. This $58.3 million figure refers to the economic value of the painting but how does it relate to its aesthetic value? Do these two entities comprise the same field or are they non-related? To what extend does the price reflect the aesthetic pleasure one can achieve from a painting, is the market perhaps the perfect mechanism to determine both economic and aesthetic value? Is the aesthetic value even objectively determinable in the first place and if so, by whom? These questions become more important as the prices paid for art become higher and higher which have led to speculation of an art market bubble (Bamberger). This thesis has the objective to define certain (possible) market failures and discuss its treats to market stability and its effect on the distinction between price and aesthetics.

3

First an analysis of the art market in general will be provided. How does this market function and what is the role of the galleries in this process? The role of information and its (asymmetrical) distribution might be a market failure. This will be discussed in section 2. In the following sections the supply side will be investigated. In section 3 the transformation of an artist to star is investigated. How does a regular, starting artist become a star in her field? What role does the public and the talent play? Might this give rise to discrepancies between aesthetics and price? As we consider the artist as the sole supplier of art because all art is unique1_{, this separates the supply side in the art} market from any other regular supply side. We will examine the role of this monopoly position of the artist in section 4. In section 5 some investigations will be shown that have been done on art regarded as investment. How are art prices moving in comparison to other investment opportunities? What effect does the speculation have on art prices and how does the combination of intentions (investment and consumption) affect the market? These four possible market failures will be analysed and after these inquiries a conclusion will be provided and some suggestions for further research will be given in section 6 and 7. Section 8 will provide an extensive bibliography.

1_{Obviously there do exist art pieces that are produced in duplicates. Consider for example the}

screen-printings by Warhol or the photography art by Andreas Gursky of which more than one are produced. Also Rodin’s famous statues have several casts, which make them non-unique. However, for simplicity, we will discuss only unique one-of-a-kind art pieces.

4

2.

Market Analyses

In economic theory, the establishment of a price (or value) mainly occurs on a market under the influence of the demand and supply mechanisms. The notion of a market stems from the place where traders meet each other to exchange goods. According to Backhouse (2002) markets first emerged during the ancient Greek civilisation and since the old days they have transformed into many different forms to meet the specialised requirements and to incorporate new technologies. Abstract markets have since then come into existence; examples are the stock market or the market for derivatives and securities. Still, the traditional market forms remain apparent; every city has its fruits/vegetables market and/or flea markets.

An art auction is another form of a traditional market in which the highest bidder receives the offered product. No restrictions are in place and an economist can immediately observe the market mechanisms. This perfect example of the demand/supply mechanism, as she who has the largest sum of money and the biggest preference for that particular art piece ‘wins’, shows empirical evidence for economic theory. The two components (preference and income) influence the price fluctuations in the market for art as shifting preferences and varying (total) income determine the prices for specific art works. This is nothing new, all economic markets function according to this principle. Preference and income determine demand and together with supply the price is established. However, in the art market there are some complications to this traditional view. Therefore a more detailed description of the market must be given. Are there not any inefficiencies or market failures?

Ashenfelter and Graddy (2003) identify the art auctions as the most important place where the price of an art piece is established. This price establishment is a good indicator of the public valuation of art and is an essential aspect when investigating the value of art. In their words; ‘art auctions provide key information for the evaluation of artistic work’ (Ashenfelter, 2003, p.764). Ashenfelter realises that the art auction is the only publicly available place where the preferences of the public is directly translated to the price of art. These prices, however, must be handled with care because of the unique nature of art. Because of this heterogeneity, the quality of art offered may differ over time. This will cause different average art prices that are not due to differences in market mechanisms but rather reflect the differences in quality of the offered objects. Still the prices at the art auctions will offer the best and most objective information of the current trends in art since other art sales are not made public. In this investigation one has to account for the powerful role played by the auction houses. It is also observed by Ashenfelter that the auctioneers decide the estimation bids that precede the auctions. These estimates play a vital role in the further workings of the auction as it is a reference point for the (usually) less informed buyers. If one also considers that the market for art auctions is highly dominated by two firms (Christie’s and Sotheby’s), it is obvious that some imperfections must occur.

Interestingly Gerard-Varet (1995) defines the auction as a secondary market for art. The preceding first stage market is of vital importance when considering the emergence of artists and the recognition of the talent. In this primary market there are many candidate artists seeking entrance in the secondary market. This is done via galleries or small, low profile art exhibitions. Only when their products are bought the candidate can be considered a serious artist as their work is now available at secondary market. This first stage purchase by a gallery or museum ‘signals their abilities

5

to the secondary markets’ (Gerard-Varet, 1995, p. 511). This first stage is crucial as it establishes the artist and creates a market with highly paid stars that capture a relative large market share and power. This first stage ‘breeding pool’ is of great importance and grants great power to the galleries that influence the emergence of new artists. This market power creates an imbalance in the art market while simultaneously strengthening to the asymmetrical distribution of information because it provides the galleries with the opportunity to gather information.

Frey (1995) also suggests that the art market is an inefficient market in which there exist some anomalies. This means that the market is not behaving rationally or efficiently. Actors are not necessarily pursuing profits, they disregard opportunity costs and sunk costs and often have an emotional attachment to the art object. This contradicts the usual assumption of rational agents that act in the market and make well-informed decisions. A lack of rationality and surplus of emotions will lead to inefficient transactions and an inefficient market. Another major restriction of the art market described by Frey is the fact that arbitrage is nearly impossible due to the limited accessibility of the market and the time delay of purchases. This results in an inelastic short run supply due to the efforts and time it takes to offer an art piece on the market. An art piece has to be handled with care and is subjected to timely transportation and security efforts. Problems that the lack of arbitrage creates are a lack of price equilibrium and price stability that are present in regular markets. There is no natural force that pulls or pushes price to a natural equilibrium. Price in the market is higher than it should be resulting in efficiency loss that could be prevented by arbitrageurs, rational actors and/or perfect information distribution.

Here some possible market failures have been identified namely the information asymmetry, the market power of certain auction houses and the entrance barriers for newcomers, emotionality and lack of arbitrage. As Ahsenfelter correctly notices, the asymmetry of information is a problem for the proper valuation of art, both in economic and aesthetic sense. Not only the auctioneers but also the valuers, galleries, and experts play a vital role in this unequal distribution of information. Ill-informed consumers will either pay unrealistic prices or be left to the mercy of ‘experts’ (galleries, auctioneers, etc.). Their dependence has indeed led to certain galleries and auction houses to gain enormous market power (Sotheby’s and Christie’s have a near duopoly). To maintain this position the insight of Gerard-Varet is interesting in the sense that the entrance barriers for new artists are imposed and kept up by the intermediaries (galleries and auctioneers). These three market failures may thus lead to inefficient markets and refrain from maximizing both producer and consumer surplus. The prices in the market for art might not maximize producer and consumer utility.

One good example of a gallery have enormous power is the art gallery of Larry Gagosian. This American art dealer has gathered many influential contemporary artists to join his gallery exclusively. This has attracted many wealthy buyers to trust the Gagosian gallery for their new purchases. As they attract each other, Gagosian has created a self-perpetuating spiral that draws both new artists and new buyers, and thus reinforces Gagosian’s market power. The weight of this growing number of influential artists (that are attracted by the number of clients) has given Gagosian a trusted name. Since the art market is a volatile market, clients appreciate any certainty one can offer. This has put Gagosian in a very powerful position; ‘In the contemporary art market, there is no standard formula for determining what an artist is worth. It's famously difficult to determine which artist will have lasting cultural significance over decades or centuries, and which will be a flash in the pan. This gives top dealers like Mr. Gagosian enormous power to influence and even set the markets

6 for the artists they represent. Anyone who wants his art must pay his prices.’(Crow, 2011). Gagosian can assure price stability, as his wealthy and trusting clients will trust his artistic repertoire, which he has gathered through the aid of his client circle.

Don Thompson has observed this self-fulfilling process and has acknowledged the troubles that might resort from it. He recognizes that the ‘reassurance given by the dealer’s brand is reinforced by the behavior of the crowd’ (Thompson, p.41). His conclusions about the power that certain galleries or dealers have (and thus establish a brand) seem to be supported by evidence as the article about Gagosian showed. But this feature of the market posses the problem of asymmetrical information, market power and restricted access to the art market for newcomers. Potential buyers are dependent on the information that the galleries have which thus gives them an advantage, an advantage they have created themselves through the power they have acquired. These are problems that might influence the pricing of art and can be considered market failures.

The possible discrepancy between aesthetic value and economic price has been a starting point for the analyses of the art market in general. As mentioned, the market consists of the buyer and the seller that are surrounded by galleries, auction houses, valuers and of course the artist herself. These different agents together with the special characteristic of the product art, makes the market unique in comparison to other markets. This character, however, also gives rise to problems concerning the efficiency of the market. Due to information asymmetries, market power and entrance barriers the art market is highly vulnerable for market failures. These market failures may lead to incorrect prices that negatively affect the valuation of the art piece. The art market suffers from inefficiency and fails to maximize consumer/producer surplus. If there is a discrepancy between aesthetics and economics, this might be one of the origins of the problem. In the next section a more specific analyses will be given on the emergence of a superstar and the entrance barriers new artists encounter. Why do superstar exist and how do they come about?

7

3.

Stardom

In classical economic theory the assumption is made that products are homogeneous. Under this constraint the total supply and demand will effectively find each other in the equilibrium price and quantity. From this it follows that all producers will get the same price for their product and can only increase profits by lowering their marginal costs or increase their supply. When several producers are in the market, this means that the only way to increase profits is to produce (and thus sell) more or by lowering the production cost. However, if product differentiation is allowed for, then producers can obtain a larger share of the market (and profit) by attracting consumers with differentiated products and can effectively raise prices since they offer a different product. In such a situation the distribution of earnings is not linear with respect to market share but relative to the demand of the product that in turn depends on its (perceived) quality.

Elasticity of Demand

A good example of such an absolute differentiated market is the art market. Here no product is alike as all artworks are unique. In relation to the previous theory it is also observed that indeed artists with qualitatively high products earn substantially more than their less endowed colleagues. Sherwin Rosen (1981) describes this phenomenon as he acknowledges the fact that due to quality differences the distribution of earnings is skewed. Phrased differently; ‘In certain kinds of economic activity there is concentration of output among a few individuals, marked skewness in the associated distributions of income and very large rewards at the top’ (Rosen, 1981, p.845). The reason for this skewness are the differentiated products or, in the case of the art world, the differences in talent. Artist with the highest talent produce art of the highest quality and earn a substantially higher income than their less talented counterparts. They can ask a much higher price as their market power depends on their talent rather than on their total output. In order to model this situation Rosen intends to incorporate talent into the model. The model he designed includes the dependent variables price (P) and market share (M). These are both increasing in talent (q), which is the independent variable and different among all artists. Figure 1a and 1b shows the relation.

Fig. 1a & 1b

When combining both figures, the revenue function with respect to market share can be obtained. As revenue is the price times the market share the revenue is increasing in the market

8 share. Since both price and market share are increasing in talent the revenue function is also increasing in talent. This results in a convex curve that illustrates the large differences in revenue are due to small differences in talent. It is especially clear near the end of the curve where the small increase in talent results in a large increase in earnings (derivative of revenue with respect to market share is positive; dR(q)/dq > 0).

Fig. 2

But why does this convexity occur? The reason is the heterogeneity of the products and the resulting imperfect substitution between the products. Because products are not of equal quality, these will not be good substitutes. Consumers are better of going to a highly talented artist costing 50, then going to 5 mediocre artists each costing 10. Or as Rosen puts it ‘lesser talent is often poor substitute for greater talent’ (Rosen, 1981, p.846). This can also be explained through the elasticity of demand. As the price will rise for a certain artist’s work, the question is to what extend this will affect its demand. Now assuming this is a highly talented artist and given the imperfect substitution, we can argue that the relative drop in demand will be lower than the relative rise in prices. The reason is that consumers have no good alternative and do not wish to settle for lower quality. They rather pay a little extra in order to preserve the quality of their purchase. This translates into an inelastic demand function; ε < 1.

Learning Process

The idea of imperfect substitution and perfect heterogeneity among products seems very applicable to the art world. Still a piece of art is no ordinary product and one other feature might be useful to incorporate in Rosen’s model. The difference with artistic products and regular products is the usage and the resulting utility. One can most certainly enjoy an acquired work of art but it could be agreed that one can increase its utility by learning more about the artist and her meaning or idea meant to communicate through the product. Suppose someone has interest in the ‘Wheatfield with Crows’ by van Gogh because she really liked the colours and composition of the painting. Her utility would then be higher once she actually acquired the piece and thus would be able to admire it all day long. Now what if she had come to learn about the background of that painting? Suppose she learns about van Gogh and the troubled period he had whilst painting ’Wheatfield with Crows’. The crows in the painting, the dark and sinister sky, together with the path leading through the baleful waving wheat

9

would most certainly become of total different meaning. Her interpretation of the painting would have become much deeper and this would contribute to her utility.

This learning process is identified by Moshe Adler (1985). The question that he asks himself is how the superstars of Rosen are identified by the public; is the excess in talent unanimously and unconditionally recognisable for the consumers? His answer would be negative and not because it is hard to recognise difference in talent or quality but because there hardly exists any difference in talent or quality. But why do we than still observe the highly unequal income distribution discussed by Rosen? This has to do with the utility increasing capabilities of learning and experience or as Adler puts it; ‘Appreciation increases with knowledge’ (Adler, 1985, p.208).

Remember the woman interested in the van Gogh painting. She is a new consumer in the art market. As she wants to maximize her utility subjected to her budget constraint, she prefers to buy something that she can discuss with other people to increase her experience and subsequently her utility. The ability to discuss the purchase of her choice depends on the market share of that artist. If the market share is very large (the artist is widely known with the public) then she would have no problem finding experienced people from whom she can learn about her purchase and increase her utility. The larger this potential group is, the lower her search cost and the higher her utility. So she has an incentive not to buy an obscure and unknown artist, but to choose one that is widely recognised and to minimize her search cost to improve her learning process and maximizes her utility.

This process can be modelled into a game theory matrix. The problem any art buyer faces resembles the coordination problem. In this standardized problem, both actors have two options but only receive a positive payoff if they both coordinate on the same option. Traditionally this involves a trip; they can either choose the cinema or the theatre, and they both have different preferences. Despite these differences they gain the highest payoff when coordinating on both going to the cinema or both going to the theatre. It is possible to remodel this game and make it applicable to the art consumers, the intuition remains.

Fig. 3

Let’s consider an individual who is interested in buying a painting. In the gallery she sees both a painting by Damien Hirst and one by Tracy Emin. She is not knowledgeable but has an intuitive preference for Emin. What is the most rational thing to do? According to Adler and

10 considering the individual has little to no knowledge of the art world she needs to coordinate on the artist who will minimize her search cost to increase her knowledge that will maximize her utility from that painting. This game is modelled in figure 3. Here is one individual who can choose between a Hirst and an Emin, and the majority who has already done the same but cannot be observed by the individual as she lacks the information. It should be evident that the majority’s payoff is independent from the choice the individual makes. As utility depends on the availability of potential discussion partners, the majority always minimizes its search cost. This advantage is not (yet) available for the individual so she has to find out what the majority chose in order to coordinate her action on that same painting. Since the payoff of the majority is independent, their payoff will always be 100 regardless of what the individual will choose. The individual, however, will only receive a positive payoff whenever she chooses the same artist as the majority. This will be the Nash-equilibriums (Hirst-Hirst and Emin-Emin). Her payoffs will be higher when choosing Emin because she favours the Emin painting to start with. If she does not pick the same artist as the majority she will not be able to increase her knowledge at a reasonable search cost and her utility will not be maximized. She will receive a negative payoff as she has paid money for the painting and is not able to maximize her utility, as it is virtually impossible to find discussion partners. With this (simplified) model it is shown that even when talent is equal among artists, still a hierarchy in earnings can exist. The main reason is that for maximum utility, knowledge is required. So an individual is better off copying the preferences of the majority than to explore their own intuition, even if there is no obvious difference in quality or talent.

Emerging Artists

Even though talent may not always be fully observed and it might not always be to ultimate factor in determining the price, talent is still different among the population. What is the effect of the distribution of talent among the population? And more importantly; how are those first superstars identified, the stars that are later followed by the majority? As talent is rare and sometimes not even recognised there has to some sort of information transfer from artists to audience. This transfer is described by MacDonald (1988) who starts from the assumption that talent is not equally distributed among the people as is suggested by the Rosen model and thus these structures need to be examined for greater insight in information transfer.

MacDonald identifies three different groups in the artistic sphere; performers, audience and critics. Performers need to convince the critics of their talent in order to persuade the audience of the ‘certain’ quality of their future work. According to MacDonald this can be modelled in the following way. He suggests that for all individuals willing to performing (any artistic act possible) there is an equal chance of success since no one has seen them before and is thus uncertain of any talent. After their performance the critics will publish their verdict of the work of the artist. Because the opportunity costs for performing are negative (starting artist receives low income opposed to other jobs), they will quit the business when receiving a bad review and no future stardom can be acquired. The reason is that the price will strongly depend on their previous performance. This results in the following two factors that influence future performance, both the opportunity cost of the artist and the review they receive from the critics. Obviously the audience can also chose to go to a performer with a ‘bad’ track record. Her work is cheaper as it received negative reviews and there still remains a possibility that the artist has improved. The problem MacDonald possess is the uncertainty of this event. Given the bad previous show(s), chances of disappointment are high. To

11

avoid risk and maximize utility the audience is willing to pay more for the talented artist. This high paying public is ‘not often dissatisfied and pays a high price for this assurance’ (MacDonald, 1988, p.166).

Two conclusions that MacDonald makes from this inquiry are supported by the research of Don Thompson in his book ‘The $12 Million Stuffed Shark’. The idea is that only young performers will enter the artistic market because their opportunity cost are still relatively low and they are attracted by the highly paid stars in the industry. This potential ‘superstar second period income’ is higher than the young entrant’s opportunity cost and is the reason why they will enter in the first place. Thompson observes this in his study, as many entrants are young and recently graduated art students. If they have not succeeded in their struggle for the top they will most likely drop out, making room for new (young) entrants. The other conclusion is the obvious power the critics have and the importance and scarcity of information as discussed in the previous section. Again Thompson agrees. He identifies the power galleries and critics have in the art community and he labels this as an unwanted market failure. This leads to unwanted hypes, exclusion of artists and buyers and the asymmetric distribution of information.

These inquiries have shown that in a market such as the art market the distribution of income will be skewed right. The result is that some artists have higher income even though this might not reflect the quality of their work. The relevance for this thesis is that this is due to several market failures and shows an inefficient market. If the incomes are skewed right, independent of the quality, this means that quality and prices (or value) is not objectively established. Rather the prices (and thus quality and value) depend on a much wider range of variables than merely the intrinsic value of the artwork. As galleries do have market power to interfere in the market, buyers tend to coordinate on other buyers, entrance barriers exist and elasticity of demand is smaller than one, the art market will not function efficient.

Don Thompson gives a good summary about the earnings of an artist. He says; ‘When an artist is hot, the economist’s concept of demand an supply does not apply.’ It becomes a different market than discussed in general economic theory. Demand increases exponentially in ‘hotness’ of the artist of that moment. The more attention a particular artist receives (galleries, visitors, shows, critical appraisal, etc.), the more potential buyers are jumping on ‘the bandwagon’. This bandwagon effect is typical for the art market and is further characterised as the crowd moving in great schools like fish. Safely hiding in the group, because ‘there is safety in numbers’ (Thompson, p.41). One could almost regard the art market as a plurality of individual monopolies held by artists or galleries that have power over the potential buyers. In the next section the focus will be on the role of the individual artist.

12

4.

Role of the Artist

Monopoly aspect

When the art market is characterised, it is apparent that no ordinary perfect competition model can be assumed. Under the assumption that every art piece is unique, there is absolute heterogeneity within the product art. As no product has a perfect substitute and all are perfectly diversified, the artist must be considered a monopolist. There is only one ‘H.M.’ by Luc Tuymans, which cannot be compared with other paintings, so the artist has the monopoly with this one good. Even if an artist creates photographs, lithographs or any other reproducible art form, she must still be considered the sole and unique producer and be characterised as a monopolist. Now the difference with a normal monopolist is that an artist (usually) only produces one good. This is opposed to a ‘normal’ monopolist who supplies the entire market of e.g. apples. Where an artist has only one product and can sell it for the highest bid, the ‘regular’ monopolist can sell more identical products and choose her optimal output (MC=MR). One difference between both monopolies was recognised in 1972 by R.H. Coase. He made the crucial distinction between durable and non-durable goods, of which art can be categorised in the former. Durable goods are goods that last for more than one period, they are durable. Art does not perish over time, it cannot be used up. Paintings need some refurbishment but can last for ages; the ‘Mona Lisa’ by da Vinci was painted between 1503 and 1506 and is still on display today.

Fig. 4

Coase used the example of land. Consider a monopolist in land, a typical durable good. In aiming at optimal profit she will charge the monopoly price (MR=MC) but the quantity sold will be

13

less than total available quantity. Some of the land will remain unsold. Since she is a monopolist and does not wish to use the land herself, she will try and sell the remaining ground. The problem is that since it is a durable good, the first period monopoly price buyers are out of the market and the remaining potential buyers are not willing to pay the monopoly price. The first period buyers are out of the equation and the demand curve will shift downwards resulting in a lower MR and monopoly price. This trend will continue until the competitive price is reached. However, previous monopoly-price buyers will not like this since their expensively bought land is now decreasing in value as equal land is sold at a lower price. Knowing this trend they will anticipate on the decreasing price until the price level of the land will altogether move downwards to the competitive price (MC=0). In other words the competitive price is the maximum price for which the monopolist can sell its durable goods since the elasticity of demand is infinite at the competitive level; any quantity will be sold for the competitive price.

A solution posed by Coase is that the supplier (the monopolist) can issue guarantees. She could assure that the price will not drop and otherwise she is legally obliged to buy the land back at a given price. Only then can the monopolist charge the monopoly price without losing consumers. When the artist is considered a parallel can be drawn to the theory of Coase although some problems arise as well. The artist can effectively be seen as a durable monopolist since the artwork usually exists over several periods. Nevertheless, we have established that a piece of art is unique and no homogeneity among the several products of artists exists. This does, however, not mean that works by one artist are not related. If one work of an artist rises in value, this will trickle down to her entire oeuvre. The value of paintings by one artist is correlated with each other. Consider the artist as a brand; if the brand is hyped then the entire value of all branded articles will rise. Obviously this works both ways; also negative publicity will affect total value of an oeuvre. As some works will have been sold, negative actions will affect first period buyers who lose value of their product. This discourages future buyers and the value will drop even more. Comparisons can be made with the stock market, if investors are not confident of future value their demand for that stock will drop and its price will too. Equally, artists will need to reassure the public of their future aesthetic quality.

Death Effect

There is one typical negative future price effect that only the artist has to cope with. This effect is described by Ekelund (2000) and lugubriously called ‘the Death Effect’. He considers an artist as a durable goods monopolist. The value of her previously sold work partly depends on the value of her present and future work. Here lies the key of Ekelund’s investigation; he investigates what the production effect is of the artist. Instead of a finite and known quantity of goods (like the landowner), the artist has an unknown and growing quantity of durable goods. The production of the artist has the potential of lowering the price of her existing (and sold) art pieces due to overproduction/supply. Like the landowner, the artist can also give some sort of guarantee (destroying original silkscreens, only sell top quality, etc.) in order to maintain monopoly prices but only her death can ultimately ensure buyers that production has stopped. Ekelund and his team have investigated this ‘Death-Effect’ on art prices.

They found that indeed for some artist there is a so-called death-effect, in which auction prices rise immediately after the death of the artist. Their explanation is that buyers recognise that supply has stopped and that there is no treat of depreciation due to overproduction of the value of its art pieces. So while many factors influence the market value of an art piece the effect of an

14 artist’s death is substantial, according to Ekelund et al. A guarantee, as suggested by Coase, to regulate future supply will only be fully reliable when the sole producer dies. How cruel this may sound, death is a definite stop of production. Still this is only one factor influencing the total demand function of art but it may explain the sudden rise in auction prices of recently deceased artists. (However, a reason for this ‘death-effect’ may also include the extra attention a deceased artist may receive; this created hype may boast demand).

Apart from the expected absolute productivity of an artist during her life, it is also interesting to regard the periods of productivity during her life. Weinberg and Galenson have concluded that due to changes in preference the most productive period in an artists’ life has changed over time. They have investigated the average auction prices of different artists and the corresponding ages of the artists when they made the paintings. They assumed that the higher average price corresponds to a higher quality. Over time they found out that the age at which the most successful art was produced declined. Their explanation was that in contrast to the main idea, experience is of less importance to artist in later art trends. As contemporary and experimental art becomes more important, experience becomes subordinate to creativity. This corresponds to the neurological idea that a person is more creatively capable in her earlier ages than later in her life.

The market for art does anticipate both the monopoly position of the artist and the possible death of an artist. Together with the previous description of the emergence of stars and the role of galleries in art markets, these aspects are potential causes for market failures. This section has tried to give a deeper impression of the functioning of the art market and the role that artist plays in the expectations of the potential buyers. It is clear that every art piece is unique which suggests a monopolistic approach should be taken. However, we have seen that the Coase theorem is applicable here and that a monopolistic approach might be problematic. It shows therefore that apart from all the described market failures there are some price mechanisms at stake in the art market that are due to consumers expectations. This might prove some market efficiency to be present in the art market but it is difficult to determine empirically.

15

5.

Art return vs. Stock Return

The previous sections have identified some possible market failures. But how do art prices behave? Is there a predictable trend in the market for art, a trend which might suggest an efficiently working market? In this section some more detailed analyses of art prices will be discussed. The idea behind many of the upcoming literature is based on the relation between the return on art investments and the return on other investment opportunities (mainly stocks). Can art be considered a serious investment opportunity? In observing the stock market and analysing the relation to the art market, a more detailed description of this art market could be given. In essence, the value of a stock (its price) is based on certain features, namely the accumulated future value of cash flows discounted for in present value. Do such specific features occur in the price mechanism of art? In other words; the stock market is used as a control variable to investigate the characteristics of the art market.

One of the first scholars to investigate this relation and to recognize the importance of it is the American economist William J. Baumol. Recovering data of sale prices of paintings over several centuries he calculated the average returns over these art pieces and compared them to the stock market return in the same period. His intentions were based on the intuition that prices of art do not follow the traditional economics of markets, there does not exist any market mechanism that drives the price within certain limits to its equilibrium. Art markets are unpredictable and cannot be treated as investments.

Baumol (1986) described the difference between regular markets and the art market. Regular market price mechanisms (price and quantity move to equilibrium due to several market factors) are absent in art markets. This may be contributed to the fact that elasticity of supply is 0, indeed as art supply is unique and thus only one product can be supplied, the elasticity is effectively 0. This suggests that all art pieces are unique and even though the price may rise tremendously, there is no opportunity of increasing the supply, supply is fixed. The second observation he makes is that there do exist monopolies in the art market. His argument is that since every piece is unique, either the current owner or the creator is the monopolist of that art piece. Thirdly the resale of an art piece may take years or even a century and are usually done in secret, so prices are often not made public. This makes it hard to gather useful data and to investigate the returns of these art pieces compared to the returns of equivalent investment opportunities. Lastly, Baumol observes that prices of art are very hard to determine, there is not an ‘objective’ way to determine the price as via a price equilibrium mechanism as observed in regular markets. As he says; ‘Who would dare to claim to know the true equilibrium [art]price?’ (Baumol, 1986, p. 11).

16 Fig. 5

As an experiment Baumol calculated, in the best possible way, the returns of resold paintings over a couple of centuries. He compared this to the rate of return of some other investment opportunities (mainly stocks). His results were clear. The average return on art was substantially lower than the returns on comparable assists. Also the risk incurred on art was much higher than in other investment opportunities. Art values and its returns are observed (and tested) as normal distributed. Over time the average return approaches zero. One of the reasons, as suggested by Baumol, for this difference in return between art and stocks, is that art is considered not only to provide monetary returns but also aesthetic returns. An acquired painting can be shown and admired. This will result in aesthetic, non-monetary returns solely obtained via art. The financial assets lack this ability and thus need higher monetary returns as no one will gain any aesthetic utility from gazing at a stock ownership contract. Also there might be some positive opportunity costs related to art as one can organize a profitable exhibition.

Nevertheless the rate of return could be higher if the buyer is more knowledgeable or has spent their entire live on studying the arts. This gives rise to the information asymmetry that is present in the market and creates an inefficient market. But still only those critics that are substantially ‘important’ enough so they are able to influence the trends, might be able to outperform the market. As these individuals are very rare, Baumol suggests that no potential art buyer should have the intention of outperforming the market en gain financial profit from the acquisition of art. On rational grounds the only reason for buying art is for the aesthetic value, buying art is only ‘a very rational choice for those who derive a high rate of return in the form of aesthetic pleasure’ (Baumol, 1986, p. 14).

In his article Baumol identifies some shortcomings of his investigation himself. Especially the calculations of the return could be interpreted as problematic. These calculations are based on the auction prices that were available in the data set. Nevertheless, these auction prices are constituted of several components. Components that are not equal over time or among different auctions. Components include risk premiums and sales commissions. The risk premium must be deduced from the rate of return and consist of the risk of damage or theft. These risks do both not occur among stock investments. Another factor influencing the rate of return are the sales commissions paid to

17

the auctioneer. These fees must be subtracted from the selling price in order to calculate the correct rate of return. But as some sale prices are secret, commission levels are even more classified and thus forms a constrain on the equation.

On top of these self-acknowledged weaknesses of the investigation, Buelens and Ginsbergh (1992) identify some other downsides of the model. They make an attempt to improve the model and find different conclusions than Baumol. In their view the data sample Baumol used has some flaws that affect the results. Some artistic periods in history are over represented and the auction period between the world wars should be considered biased data. What Buelens en Ginsbergh did was to break down the sample in different peer groups. These different artistic representatives show different rate of returns that indicate that they should be considered separately as they would otherwise interfere with each other. With their grouping of data, Buelens and Ginsburgh find that for some artistic periods there does exist a higher rate of return in the art world when compared to stock returns. They come to the conclusion that this is reasonable, as paintings require higher returns because of higher costs related to art. Art works are much less liquid and thus require a liquidity premium. They demand large transportation, storing and maintenance costs. And the insurance cost is much higher when compared to ordinary investment options. All these extra costs and risks need to be reflected in the returns, this is a reason for the higher returns found by Buelens and Ginsburgh.

Even more shortcoming on the model have been characterised by Frey and Eichenberger (1995). Apart from some of the above-mentioned shortcomings they also distinguish the problem with data, which is not abundantly available. Only auctions are analysed whereas these are not the only transaction vehicles for selling art. Secondly they are also critical about the transaction costs that are not incurred in the models, mainly because these are hard to come by and differ per time, dealer and country. Nevertheless they have a major influence on the actual price. Thirdly the taxation costs are of importance; these are often left out of the models whereas they have a big influence. In many countries art is a way of lower the overall tax burden so it should be included when calculating the returns of art. Lastly they argue that the comparable financial assets, by which the returns are compared with, could be arguable. Many different things can be substituted for in this case; apart from stocks also bonds, treasury bills, assets, etc.

Ginsburgh and Jeanfils (1995) have elaborated on the grouping of artist and the effect of prices of their works more in his article ‘Long-term Comovements in International Market for Paintings’. Here they group specific artists in groups of their period and style; examples are ‘Contemporary European Masters’ and ‘Contemporary US Painters’. The price movements of these groups are analysed in the three most influential art markets (Paris, London and New York). As a result they found that all art markets and artistic groups move closely together indicating that no price leader is apparent but all follow a roughly similar price trend. A second result is that they found no correlation with the stock markets over the long term. But there is some influence in the short run between art markets and the major stock markets (New York, London, Tokyo and Paris). Especially the Tokyo exchange shows some impact on the price of ‘Contemporary European Masters’ in the short run but this only moves one direction. Art markets are presumably too small to influence the stock markets. A possible reason for the correlation is that art can be considered a secondary good only to be bought if economy is strong. A good indicator for the economic growth is the stock market; therefore it might show a positive short-term correlation with the art market.

18 If these investigations show us one thing it is that art should be considered as a highly volatile asset. As has been described in the previous sections, a major influence on the prices of art is the amount of ‘buzz’ an artist generates and how large her ‘bandwagon effect’ is. These are hard to determine empirically and thus difficult to assess by a potential investor. The only security an artwork will provide is the aesthetic pleasure that one can derive from it (although it depends on the majority’s pick, see Adler). The best conclusion we can draw from these investigations regarding art as an investment is that art is a consumption product in the first place and investment opportunities are best considered for secondary purposes only for risk averse investors. To use the concluding remarks of Goetzmann who has done some elaborate studies in this field as well and found tremendous growth in art prices; ‘Despite this growth, however, there is little evidence that art is an attractive investment for a risk averse investor. Art, absent its aesthetic dividend flow, is only potentially attractive to an agent who would otherwise choose a relatively volatile portfolio.’(Goetzmann, 1993, p.1375). So even though art seems potentially very lucrative, the risks are high and the (monetary) returns are low. Only those who seek more than financial gains from an artwork should invest in art, especially because the real return on art is very difficult to calculate. Apart from the insecurity of determining factors also the extra costs (storage, transport, tax, insurance, etc.) need to be included and these are often hard to determine. In conclusion, the market for art is highly volatile which might be due to the many market failures identified in the previous sections. There is thus no substantial pricing mechanism to be recognised that makes a price prediction possible. This might be a sign of imperfect markets and support the inefficient art market hypothesis.

19

6.

Conclusion

Art is often characterised as an attempt to capture the elusive. It is a materialisation of a transcendental experience or the infinite forced into a finite form. When taking the existing literature on the economics of art into consideration, one could argue that the economist investigating the arts has a similar task. Within the economic theory art is a very peculiar product, for one art has a unique character and secondly the actual value of art is very difficult to determine. This atypical character apparently did not lead to a low level of attention for art; in fact art has always interested people. Especially the wealthy classes have found their way to the art market despite the uncertainty of its value.

In this thesis the aim was to investigate the market for art. The reason for this investigation was the incredible price paid for individual pieces of art and the uncertain relation to its aesthetic value. Perhaps there is a proper relation between the aesthetics and economics in the arts but there might also be some market failures causing the high spread in pricing and producer and consumer surplus to be lower than its maximized potential. Therefore the investigated hypothesis was ‘the art market is an inefficient market subjected to market failure’.

First a characterization of the art market was given and the role of several middlemen explained. The discrepancy between the price and artistic value has proved problematic especially because of information asymmetry. A figure or body that could mediate in this situation could spark sales and boost the creative arts sector and was taken on by the art gallery or dealer. With the introduction of this arbitrator, however, the buyers and sellers have created a middleman with enormous potential power. This Frankenstein has become stronger than its masters and now controls a large part of the art market through selecting the new entrants, steering the trends and hypes and determining value. For new entrants it is nearly impossible to succeed without the blessing of such galleries and buyers are forced to follow the gallery trends if they don’t want to lose potential profit or pleasure. The information asymmetry and the dependence on their recommendations have increased even more and prove to be a problematic market failure.

In the third section it has been the aim to determine the role of an artist in her market, being a monopolist and unravel the path to stardom. This has shown that the income (stardom) of individual artist does not depend on the quality of their work (talent). As this relation does not exist it means that other factors influence the market to interfere with its efficiency. Factors of such market failure might include; market power of middlemen, coordination of consumers, elasticity of demand and entrance barriers for new artists.

Apart from the identified market failures, in the fourth section some expectations of the consumers and its effect on the pricing mechanism has been investigated. Both Coase’s theorem en the Death effect has shown to be applicable to the art market and give rise to the idea of some market efficiency. This idea is further investigated in the fifth section where the relation of an art object with an ordinary investment opportunity is described. This relation has proved problematic and shown that a piece of art is a very volatile investment opportunity. A reason for this high volatility might be the previously identified market failures that complicate rational pricing mechanism and creates an inefficient market that give rise to volatile art prices. It seems that the art market finds itself in a spiral of uncertainty that reinforces itself.

20 This self-reinforcing spiral might be the characterisation of the art market that is most important. Man has always had the faculty of the mind that has given them the opportunity to imagine and create pictures of the world around them. This capacity separates man from animal and show the human capacity to reflect, sense of time and the capability of abstract thinking. Cave drawings dating from the Stone Age show an ever-existing drive to materialise mental pictures. And thus the art is a means of fulfilling this urge and an expression that distinguishes us within the animal kingdom. For these reasons art will always be an important aspect of the human society and the enormous market that has been created around it could be considered a natural human reaction. This is naturally accompanied with the enormous growth in the size of the market in both sales volume and participating agents. This continuously growing appeal to the arts together with the unique character of the art piece has indeed made the market less efficient. The many market failures are both a result and a cause of this and have created the spiral that has led to the excessive market behaviour we see today. So indeed the hypothesis can be adopted as the art market is indeed subjected to several market failures that make it an inefficient market.

21

7. Further Research

This thesis has provided an extensive literature review on the characterisation of the art market and its treats to efficiency. In this section some proposals for further research will be given.

The aim of the thesis was to find some possible market failures and establish whether or not the market for art was efficient or not. In order to make the market less inefficient the mentioned market failures should be countered or minimized. This could be an interesting topic for further research. The problem with this investigation would be the different incentives of several parties present in the market for art as their position is strong and profitable. Galleries and experts for example, will not be agreeable to lift their market power to maximize consumer surplus.

Also some aspects of the market failures itself may be up for further consideration. It is clear that some actors in the art market have more information than the majority. This may result from elaborate study or experience and gives rise to extensive information asymmetry, the nature of the emergence of this asymmetry would seem worth investigating in more detail. Bonus and Ronte (1997) have investigated this aspect of the cultural world and identify how some experts have gained the trust of the market. These experts have gained credibility in the market and can thus influence the market preferences. Is the power they have measurable? How large is their power?

Another interesting market failure worth investigating in more detail is the market power certain institutions have. This market power is partly a result from the information asymmetry, and benefits mainly the auction houses and galleries that have large market shares and signal a natural authority. How does this affect the market, is it quantifiable? How does this influence prices and/or market supply? An interesting case is described by Ashenfelter (2003) about the cartel case of Sotheby’s and Christie’s. These two major auction houses were accused of collusion on prices and abusing their market power. Galleries have dominant positions in the art market as well, they decide who exposes and sells. This requires proper regulation and supervision to prevent power abuse to exist. Further question could be about the incentives of the galleries. Are these based on rational grounds, do the most talented artist get exposed? Or do galleries misuse their dominant position and only seek the easiest revenue?

As we have seen that the returns of art are hard to determine and that they vary widely over time and market, Frey and Eichenberger also identify the importance of non-monetary returns. They recognize that a big part of the return of art is in the aesthetic value, or as they call it, the psychic returns. What are its determinants and how do they affect the art market? How does this relate to the emotional part of an art purchase?

Overall it might be interesting to discuss in some more detail the relation of the economic price and the aesthetic value of a piece of art. The different market failures that are present in the market for art might contribute to the problematic relation of aesthetic value and economic price. If the market would suffer less from concentrated market power, asymmetrical information, bandwagon effect, volatility of prices then the prices might better correspond to aesthetic value. Nonetheless it has to be kept in mind that both aesthetic value and economic price are different entities that remain hard to compare. Aesthetics cannot be ordered ordinal whereas prices are perfectly combinable. Even though it has been shown that prices are not always objectively determinable, it must be apparent that aesthetics is surrounded by a much more subjective sphere.

22 Despite these differences Abbing rightly identifies interdependency between aesthetics and pricing. He argues that aesthetics is a result of social values of which the economic value is one. This means that through the social sphere a relation between market and aesthetics is present and thus both must be investigated. So, we must remember that the field of cultural economics (as we might call it) is a difficult combination, as it ‘attempts to integrate within itself differing approaches, namely those of both the economists and the non-economists.’ (Svoboda, 2011).

23

8.

Bibliography

Abbing, H. 2002. Why are Artists Poor? The Exceptional Economy of the Arts. Amsterdam: Amsterdam University Press.

Adler, Moshe. 1985. "Stardom and Talent." The American Economic Review 75 (1): 208-212.

Ashenfelter, O. and K. Graddy. 2003. "Auctions and the Price of Art." Journal of Economic Literature 41 (3): 763-786.

Backhouse, R. E. 2002. The Penguin History of Economics. London: Penguin Books Ltd.

Bamberger, A. 1998-2011. He Kids, It’s Bubble Time! http://www.artbusiness.com/orwxb.html Batty, D. 2013. "Hirst Weg Bij Uberkunsthandelaar." De Volkskrant, 08-01, 9.

Baumol, William J. 1986. "Unnatural Value: Or Art Investment as Floating Crap Game." The American

Economic Review 76 (2): 10-14.

Bonus, H. and D. Ronte. 1997. "Credibility and Economic Value in the Visual Arts." Journal of Cultural

Economics 21: 103-118.

Buelens, N. and V. Ginsburgh. 1993. "Revisiting Baumol's 'Art as Floating Crap Game'." European

Economic Review 37: 1351-1371.

Coase, R. H. 1972. "Durability and Monopoly." Journal of Law & Economics 15: 143-149. Crow, K. 2011. "The Gagosian Effect." The Wall Street Journal, 04-01.

Cultuurredactie. 2013. "Superrijken Drijven Prijzen Topkunst Op." NRC Handelsblad, 17-05, 21. Ekelund Jr, R. B., R. W. Ressler, and J. K. Watson. 2000. "The 'Death-Effect' in Art Prices: A

Demand-Side Exploration." Journal of Cultural Economics 24: 283-300.

Essen, J. 2012. "Kunsthandel Krijgt Code Tegen Prijsafspraken." De Volkskrant, 12-12, V8.

Frey, B. S. 1997. "Art Markets and Economics: Introduction." Journal of Cultural Economics 21 (3): 165-173.

Frey, B. S. and R. Eichenberger. 1995. "On the Rate of Return in the Art Market: Survey and Evaluation." European Economic Review 39: 528-537.

Gerard-Varet, L. 1995. "On Pricing the Priceless: Comments on the Economics of the Visual Art Market." European Economic Review 39: 509-518.

Ginsburgh, V. and Ph Jeanfils. 1995. "Long-Term Comovements in International Markets for Paintings." European Economic Review 39: 538-548.

24 Goetzmann, W. N. 1993. "Accounting for Taste: Art and Financial Markets Over Three Centuries."

The American Economic Review 83 (5): 1370-1376.

Goetzmann, W. N. and M. Spiegel. 1995. "Private Value Components, and the Winner's Curse in an Art Index." European Economic Review 39: 549-555.

MacDonald, G. M. 1988. "The Economics of Rising Stars." The American Economic Review 78 (1): 155-166.

Mankiw, G. and M. Taylor. 2010. Economics. South- Western Sengage Learning

Rosen, Sherwin. 1981. "The Economics of Superstars." The American Economic Review 71 (5): 845-858.

Svoboda, F. 2011. "In Search of Value: Vienna School of Art History, Austrian Value Theory and the Others." The Journal of Socio-Economics 40: 428-435.

Thompson, D. 2008. The $12 Million Stuffed Shark. London: Aurum Press Ltd.

Vogel, C. 2013. "Christie's Contemporary Art Auction Set Record at $495 Million." New York Times, 15-05.