- Luc Renneboog
Unprecedented high prices commanded for an artist’s work at auction are frequently reported by the popular press. A record sale at US$450.3 million (with a hammer price of US$400 million) for the Salvator Mundi painting (allegedly) by Leonardo da Vinci was auctioned by Christie’s on November 15, 2017. This auction record exceeded the previous auction records of the same year (namely, Twelve Landscape Screens by Qi Baishi, and Untitled by Jean-Michel Basquiat) by more than 300%. The media not only report on absolute price records but also on records by renowned artists and the breakthroughs of artists whose work reaches thresholds of US$1 million, $5 million, or multiples thereof for the first time.
In the light of the prices above, one could wonder why records continue to be broken. A first reason is that the fundamental value of an art object is difficult to determine. The value of an asset is the sum of the discounted future cash flows, but given the absence of a rental market for art, this method cannot be used and one is largely reliant on past prices to set this number. Furthermore, even if one believes that an art object is a store of value and is likely to retain its value in the future, a collector may still pay a higher price because owning an art object may have personal utility. This utility could be called an “aesthetic dividend,” a term which alternatively captures the value assigned to the ability to enjoy the art object (e.g. a painting on one’s wall) on a daily basis; a “dividend on cultural capital” as owning top art may give the collector access to global cultural or business elites; or a “showing-off dividend” when media attention triggered by very high prices caresses one’s ego.
Second, for a large part of the auction art market the supply is inelastic as most artists (with exception of late modernist and contemporary ones) are deceased. The availability of objects by deceased artists on the market decreases following acquisitions by museums and long-term collectors. This scarcity augments prices. Thus, prices are primarily driven by demand. Art prices increase when new art collectors and investors arrive in the market. For instance, Russian collectors discovered the art market about 2000 and Chinese investors drove prices up around 2005 (Renneboog and Spaenjers 2010). The entry into the art market of new types of collectors (e.g. younger people) with different tastes could induce excessive publicity about specific art schools or artists (Pénasse, Renneboog, and Spaenjers 2014).
A third reason for record prices is that bubbles are inherent in the art price formation (Pénasse and Renneboog 2017; see also Art market bubbles and crashes). Short selling is not possible in the art market, as it is in liquid stock markets. An investor who has negative information about a firm or is pessimistic about its future performance can short this stock. The investor borrows the stock from a bank and immediately sells it, expecting a future price decline. Some time later, they purchase the stock at a lower price and return it to the bank, thus making a profit. In this manner, negative information is immediately reflected in stock prices. Due to illiquidity, this cannot be the case in the art market and, therefore, the art market mainly moves with positive information. When pessimism reigns, the trade in art itself often declines as sellers wait for better times. In addition, when some artists are out of vogue their sales are excluded from the secondary market (the auction market) such that this type of price pressure is not visible in the public art market. Therefore, negative information on art has a lower or a delayed impact on art prices. When optimism is prevalent for longer periods or new art-loving clienteles enter the market, the market may boom until the bubble bursts. A large shock to the art market—comparable to the Salvator Mundi auction—occurred in 1987 with the sale of van Gogh’s Vase with Fifteen Sunflowers at Christie’s London for the then record price of US$83.6 million (inflation-adjusted; the original or nominal price was US$39.7 million). While many records were broken in the second half of the 1980s, the so-called Japanese bubble marked a paradigm shift and created a very large art bubble (Hiraki, Ito, Spieth, and Takezawa 2009). This was a time of very high demand for French art—French Impressionist and Post-Impressionist painting in particular—by Japanese businesses and collectors. It initiated an upward trend for art prices that would endure for the decades to come (with the exception of the price correction in the early 1990s). By the middle of the 2010s, the record benchmark moved up to US$185 million in the auction market (Picasso’s Les femmes d’Alger) and US$310 million for private sales (Willem de Kooning’s Interchange). Occasionally, the bubble of the overheated art market bursts with strong price corrections, usually induced by an economic shock: the Japanese bubble deflated in 1990–1991 following the crisis in the property market; the strong art market of 1998–1999 suffered from the collapse of the dot.com bubble in 2000–2003; and the high prices in 2004–2006 were negatively affected by the great recession of 2008–2010 (Pénasse and Renneboog 2017).
Fourth, art prices are also affected by spillover effects from other markets. An increase in share prices is followed in about a year by an increase in art prices because high capital gains spillover to the art market (Renneboog and Spaenjers 2013). In a study spanning three centuries, Goetzmann, Renneboog, and Spaenjers (2011) also reveal that large income inequality positively affects the art prices; when high-net-worth individuals (the top 0.1% of the income distribution) earn a larger part of the wealth, the demand for art rises and the prices go up.
The art market also suffers from some institutional problems that affect art price formation. First, the market is not very transparent. Only about half the art market operates via auctions (the results of which are publicaly available) and the other half of sales are private transactions (most of which are never made public). Some parties are able to manipulate prices. For instance, it is known that some galleries who represent specific artists buy their artists’ work at auctions. These purchases are made in order to set high prices that can serve as benchmarks for future reference or to protect the value of the gallery’s stock of paintings of those artists (in case the work is in danger of not reaching the reservation price at the auction and hence remains unsold). Given that the identity of a purchaser at an art auction is never released (unless he or she chooses to do so), it is hard to identify artists who have benefited from this practice of price support. Second, the market is illiquid, with auctions not taking place continually but at specific times of the year, often clustered in the spring and autumn. Also, illiquidity is sometimes artificially created; galleries can ration the work of the artists they represent in case of high demand in order to create interest and then charge higher prices. Third, auction houses try to convince art collectors to sell top artworks, around which auctions can be built and marketed. In order to do so, they can offer the sellers minimum price guarantees provided by external parties (collectors interested in buying the object who then share the auction house’s commission on the sale) or by the auction house itself (Graddy and Hamilton 2017). Fourth, auction house estimates of the value of the art object can be manipulated; auctioneers can inflate these estimates to increase transaction prices and there is research that has detected evidence of such anchoring of art prices (Beggs and Graddy 2009).
While record prices are eye-catching and may induce collectors and speculators to enter the art market, they may be put on the wrong footing for several reasons. First, even when the media report a record price for an artist, it may not necessarily be a real record. For instance, at a Christie’s auction on February 28, 2017, the hammer price of Gauguin’s landscape Te Fare (La Maison) was US$25 million (yielding a cash flow to the seller of US$22 million after subtracting the commission). Still, the seller had paid US$85 million for the same painting in a private transaction nine years earlier. Thus, the “record” price in 2017 actually obscured a massive loss. At the same auction, Magritte’s Le domaine d’Arnheim sold for US$12.7 million (hammer price), a precipitous drop from the purchasing price of US$43.5 million seven years earlier. These examples serve as a warning that, while records in the upper percentiles of the art market may be continually broken, these numbers do not reveal much about the returns that an art collector, a speculator, or an (uninformed) art lover can earn in the market. Record prices are not a foolproof indicator of an artist’s value or the direction of the market. Auction catalogs usually provide details about the provenance of an art object, but past prices and returns are not included, which implies that the burden is on potential investors to collect past pricing information. Renneboog and Spaenjers (2013) report that diversified art investors with a long-term investment horizon (from 1957 until 2007) obtain an annual real return of 4% (and about 8% nominally).
Second, the record prices reported in the media are mostly raw numbers in that they are not corrected for vast transaction costs (and costs of transportation and insurance), which reduce the net value of the object for the seller. The commissions for a round-trip (a purchase and subsequent sale) depend on the total value of the art object, but average about 25%. This stands in marked contrast with the liquid secondary equity and bond markets, where transaction costs are only a tiny fraction of those in the secondary art market. The commissions in the primary art market, namely those charged by the galleries, amount to 50%. Hence, large commissions account for a part of the record prices.
A third caveat on record prices relates to the fact that they do not reveal anything about the risks of the underlying art investment. The risk (the volatility of returns) of art is very high—much higher than that of other financial asset classes. Furthermore, the riskiness depends on the type of art. The more recent the artist or school of art, the greater the risk of long-term popularity and value retention and the higher the probability for strong price declines.
All in all, record prices are unlikely to be driven by the fundamental value of art objects, but rather are affected by the psychological effects of owning and enjoying art. Record prices may not only be a premonition of art market busts, but they can also hide negative returns and obscure high transaction costs. The art market can be manipulated by market participants, such as auction houses, by means of anchoring or price guarantees.
- Beggs, A., and Graddy, A. “Anchoring Effects: Evidence from Art Auctions.” American Economic Review 99, 3 (2009): 1027–1039.
- Hiraki, T., Ito, A., Spieth, D., and Takezawa, N. “How Did Japanese Investments Influence International Art Prices?” Journal of Financial and Quantitative Analysis 44, 6 (2009): 1489–1514.
- Renneboog, L., and Spaenjers, C. “The Iconic Boom in Modern Russian Art.” Journal of Alternative Investments 13, 3 (2010): 1–14.
- Goetzmann, W., Renneboog, L., and Spaenjers, C. “Art and Money.” American Economic Review 101, 3 (2011): 222–226.
- Renneboog, L., and Spaenjers, C. “Buying Beauty: On Prices and Returns in the Art Market.” Management Science 59, no. 1 (2013): 36–53.
- Pénasse, J., Renneboog, L., and Spaenjers, C. “Sentiment and Art Prices.” Economics Letters 112, no. 3 (2014): 432–434.
- Graddy, K., and Hamilton, J. “Auction House Guarantees for Works of Art.” Journal of Economic Behavior & Organization 133, 1 (2017): 303–312.
- Pénasse, J., and Renneboog, L. “Speculative Trading and Bubbles: Origins of Art Market Fluctuations,” CentER Discussion Paper Series No. 2014–068 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2523854 (accessed May 30, 2018).