Is art a good investment?
Abstract:
– Establishing a reliable index of art market prices faces the problem of selection bias, which was only recently raised (2013) in academic literature.
– There is consensus that: (1) the profitability of art (paintings) is more volatile than that of the stock market; (2) the profitability of art is weakly correlated with that of the stock market; (3) in the long term, art is less profitable than the stock market (around 3% vs. 9%);
– Investing in a diversified portfolio of paintings (regardless of segment) would only be rational for an investor with a diversified asset portfolio (other than art);
– However, contemporary pictorial art and works by star artists significantly outperform the pictorial art market as a whole, but on fundamentals that do not appear to be sustainable.
Art, by its very nature, belongs more to the realm of aesthetics than to the commercial sphere. In reality, this is not always the case: in Switzerland and Luxembourg, some works are simply stored in vaults at a constant temperature and humidity, waiting to be resold and… moved to another vault. The acquisition of works of art at high prices by the super-rich (Paul Gauguin’s painting » Quand te maries-tu? » was sold for a record price of $300 million in a private sale) also raises questions about its commercial dimension: compared to stock market investments, is art a good investment?
1. A price index that is difficult to establish
The first difficulty is establishing the evolution of art prices. This difficulty is specific to illiquid and heterogeneous assets: Picasso paintings are traded once a year and are all different, while Microsoft shares are traded every day on the stock market and are identical.
The first work in this area dates back to the early 1960s[1]: once auction data is available, how can a price index for paintings be established, given the uniqueness of each painting? There are two main methods, the limitations of which demonstrate the difficulty of constructing such an index.
The first consists of comparing the price of a work that has been sold several times at auction over time. The advantage of this method is that it allows the impact of a work of art’s characteristics on its sale price to be taken into account perfectly. The disadvantage is that only works of art that have been resold are taken into account. However, a resale may be motivated by an increase in value, which introduces a bias into this method. The first study on this issue, which is relatively recent (2013), even estimates that the overvaluation of the profitability of art as a result is in the order of 50%[2].
The second method uses econometrics to separate the value that reflects the characteristics of the work from its price index component. The price index component of a work would reflect, in particular, its aesthetic value, which varies over time according to fashion trends. The disadvantage of this method is the assumption that a few variables can explain very significant differences in intrinsic value, typically between a « star » work and a work that is not.
These methods have since been systematized by specialized research firms, such as Artprice, which maintain their own indices.
2. Pictorial art is less profitable than the stock market but has the advantage of being weakly correlated with it
The evolution of the stock market and painting prices between 2003 and 2013 (see graph below) illustrates three widely accepted observations:
– The (quarterly) profitability of pictorial art is more volatile than that of the stock market.
– The profitability of art is weakly correlated with that of the stock market, with a « beta » of around 0.2-0.3. Beta is a coefficient that indicates how the profitability of a given asset varies on average in relation to the profitability of the market (the stock market, in practice). A beta of 0.2 means that when the stock market rises by 1%, the price index for pictorial art rises by an average of 0.2%.
– Over the long term, art is less profitable than the stock market. The S&P 500 stock market index has appreciated by an average of 9% per year since its creation in 1923, compared to around 3% for pictorial art over several centuries.
Graph 1: Evolution of prices for pictorial art and the stock market between 2003 and 2014
Sources: Artprice, BSI Economics
« Less profitable, » « more volatile » but « less correlated »: what conclusions can we draw?
3. Investing in pictorial art: diversifying or riding the contemporary art bubble
Investment funds specializing in works of art allow investors to invest in a diversified portfolio of paintings[5]. Does financial theory justify such an investment?
According to the Capital Asset Pricing Model (CAPM), established by Nobel Prize winner William Sharpe in 1970, the expected return on an asset by an investor with a diversified asset portfolio depends solely on the non-diversifiable risk of that asset, measured by its beta. A ten-year history of the quarterly rate of return on paintings (Artprice index) and the S&P 500 allows us to calculate a beta of nearly 20%, reflecting the relatively low sensitivity of painting returns to stock market returns. With a beta of 20%–30%[6], a direct application[7] of the CAPM suggests that the expected rate of return on a diversified portfolio of paintings should be in the range of 3–4%[8] per year, which roughly corresponds to the rate of return on pictorial art over several centuries.
For an investor who has to choose between investing everything in a diversified stock portfolio and investing everything in a diversified portfolio of pictorial art, the choice should (without hesitation) be the stock market, as the return on the stock market is higher and less volatile than that on paintings. In the absence of diversification, the entire risk (volatility of rates of return), and not just the non-diversifiable risk (« beta »), would need to be compensated.
Investing in pictorial art (in a diversified manner) is therefore justified for an investor who has assets other than works of art, the intuition being that the low correlation between the profitability of pictorial art and the market as a whole reduces the risk of such a portfolio, thereby making the relatively low profitability of art acceptable. Given the high transaction costs, which can reach 25% of the value of a work, such an investment is justified in the long term.
However, the performance of pictorial art over several centuries is neither representative of its more recent performance in the context of contemporary globalization, nor representative of the performance of contemporary art.
Several studies show that prices stagnated or fell between 1900 and 1960/1970, with declines precipitated by World War I and the 1929 crisis. Since the turn of the 1970s, the price index has increased tenfold, or approximately 5-6% per year, driven by contemporary art.
Within contemporary pictorial art, it is the works of the most highly rated artists that have seen the greatest increase in price. Over the period 2000-2008, the index for contemporary pictorial art more than doubled, while that for artists in the top 100 (in terms of sales) almost quadrupled. For example, Picasso’s painting » Les Femmes d’Alger » sold for $179 million in 2015, compared to $32 million in 1997.
Is there a bubble in contemporary art? Since art has no intrinsic value in the financial sense (we cannot therefore talk about a price difference in relation to an intrinsic value, which defines a bubble), we talk about a bubble in a phenomenological sense, based on the analysis of price movements. A study[10] published earlier this year in the Journal of Empirical Finance concludes unequivocally that there has been a bubble in contemporary pictorial art, among other areas, since 2011.
What could trigger a sharp price correction? And first of all, what would motivate those willing to pay millions for a painting? Art has the advantage, for some fortunes, of being an opaque market. In concrete terms, a work of art can be sold at auction without the identity of the buyer and seller being revealed. This opacity is a godsend for concealing[11] capital flight from emerging countries, particularly China, which has become the driving force behind the contemporary art market. From this point of view, the economic situation seems unfavorable, due in particular to the slowdown in growth in China, the economic crisis in Russia and Brazil, a possible rise in Fed rates (leading to a reallocation of capital to the US bond market) and the likely lifting of capital controls in China (reducing the need to hide capital).
Figure 2: Evolution of painting prices since 1900
Sources: The Economics of Aesthetics and Record Prices for Art since 1701, Christophe Spaenjers, William Goetzmann, and Elena Mamonova (2015)
Conclusion
From a financial theory perspective, investing in a diversified portfolio of paintings makes sense for investors who already have diversified assets. The intuition, formalized by the CAPM, is that the significantly lower return on art (3% per year over several centuries) compared to stocks (9% per year) is offset by the reduction in portfolio risk, due to the relatively low correlation between the return on art and the stock market (beta of around 0.2 – 0.3). This is, more or less, the consensus on this issue, which some researchers (Kraüssl, Lehnert, and Martelin) have recently questioned, highlighting the importance of selection bias in the calculation of art market price indices, a problem that real estate also faces.
The outlook seems bright for this market if we believe that the price momentum that began in the early 1970s (5-6% per year) will continue. This is even more true for works by star artists (from the « top 100 »), the most profitable segment of the art market. Are these price increases based on sustainable fundamentals? The 1.2 million works stored in the vaults of Geneva’s free ports give cause for doubt.
Sources:
-Report: The Art Market in 2014, Artprice.
-Website of Professor Aswath Damodaran (Stern School of Business, New York University)
Auctions and the Price of Art, Orley Ashenfelter and Kathryn Graddy, 2003
-The Economics of Aesthetics and Record Prices for Art since 1701, Christophe Spaenjers, William Goetzmann, and Elena Mamonova, 2015
-Accounting For Taste: Art and the Financial Markets Over Three Centuries, William N. Goetzmann, 1993
-Art and Money, William N. Goetzmann, Luc Renneboog, and Christophe Spaenjers, 2010
-Does it Pay to Invest in Art? A Selection-corrected Returns Perspective, Arthur Korteweg, Roman Kräussl, and Patrick Verwijmeren, Review of Financial Studies, 2013.
-Is There a Bubble in the Art Market?, Roman Kräussl, Thorsten Lehnert, and Nicolas Martelin, The Journal of Empirical Finance, 2016.
-When the contemporary art bubble bursts…, Kenneth Rogoff, Les Echos, September 10, 2015.
[1]Rush, Richard H. 1961. Art as an Investment
[2]Arthur Korteweg, Roman Kraüssl, and Patrick Verwijmeren calculate an average annual return on art over the period 1972-2010 of 6.5% after correction for bias, compared to 10% before correction.
[3]Average annual rate of return of 9.5%, according to Aswath Damodaran.
[4]3.3% over the period 1780–1960 according to Robert Anderson (1974), 3.2% over the period 1716–1986 according to William N. Goetzmann (1993).
[5]Example: funds managed by The Fine Art Fund Group.
[6]Beta of 31.5% calculated by Pesando over the period 1977–1991.
[7]Using a risk-free rate of 2.2% and a risk premium of 6.1%, figures regularly updated by Aswath Damodaran (Stern School of Business, New York) on his website.
[8]Expected rate of return = risk-free rate + beta x risk premium.
[9]Does it Pay to Invest in Art? A Selection-corrected Returns Perspective, Arthur Korteweg, Roman Kräussl, and Patrick Verwijmeren, Review of Financial Studies, 2013.
[10]Is There a Bubble in the Art Market?, Roman Kräussl, Thorsten Lehnert, and Nicolas Martelin, The Journal of Empirical Finance, 2016.
[11]China, for example, prohibits its citizens from taking more than $50,000 out of the country per year.
