Art as an Investment and the Underperformance of Masterpieces

Abstract

This paper constructs a new data set of repeated sales of artworks and estimates an annual index of art prices for the period 1875-2000. Contrary to earlier studies, we find art outperforms fixed income securities as an investment, though it significantly under-performs stocks in the US. Art is also found to have lower volatility and lower correlation with other assets, making it more attractive for portfolio diversification than discovered in earlier research. There is strong evidence of underperformance of masterpieces, meaning expensive paintings tend to underperform the art market index. A further study reveals that the underperformance could be consistent with overbidding at auctions. The evidence is mixed on whether the "law of one price" holds in the New York auction market.(JEL G14, Z10) 1 Department of Finance and Department of Operations Management, Stern School of Business, New York University, 44 West 4th Street, New York, NY 10012-1126. We have benefited from helpful discussions with Will Goetzmann, Robert Solow and Larry White on art as an investment. We are grateful to Jennifer Bowe, Jin Hung and Loan Hong for able research assistance. We would also like to thank Mathew Gee of the Stern Computer Department for his tireless efforts in rationalizing our database. We also wish to thank John Ammer, John Campbell, Victor Ginsburgh, Robert Hodrick, Burton Malkiel, Tom Pugel and seminar participants at Temple University and University of Southern California for helpful comments and John Campbell for his latent variable model algorithm. All errors are ours.© Jianping Mei and Michael Moses.9/24/01 1 Art as an Investment and the Underperformance of Masterpieces AbstractThis paper constructs a new data set of repeated sales of artworks and estimates an annual

Old master paintings, export veto and price formation: an empirical study

“…For the determination of price indices, at least four approaches have been developed: i) indices which reflect experts' personal judgements (like the Sotheby's Art Index); ii) indices based on the repeat sales regressions (see, among others, Locatelli-Bley and Zanola 1999; Pesando and Shum 1999;Mei and Moses 2003); iii) indices based on the average painting methodology, with its refinement of the representative painting (see Candela and Scorcu 1997); iv) indices based on hedonic regressions (see, among others, Buelens and Ginsburgh 1993;Chanel 1995;Agnello 2002).…”

Investment in early American art: the impact of transaction costs and no-sales on returns

“…Mei and Moses (2002) using the RSR method find considerably higher nominal and real returns in over the period 1900-1986 for all art (American, Impressionist and Old Masters) with relatively lower standard deviation (2002: 1657). Mei and Moses used auction catalog entries for their data, and if an entry noted a prior sale, the earlier sale price was recorded.…”

The impact of the Droit de Suite in the UK: an empirical analysis

“…Given our database, we have proceeded with the matching method described above. 20 This methodology was developed by Bailey et al (1963) and used by Case and Shiller (1987) and Hosios and Pesando (1991) for the real estate market, and subsequently used by Goetzmann (1993), Pesando (1993), Mei and Moses (2002), Goetzmann and Spiegel (2003) and Beggs and Graddy (2008) for the art market. 21 The average return per year is calculated as exp(l) 1/13 , where l is the sum of the coefficients on each period.…”

“…r 2 /2) 1/13 , where r 2 is the variance in the error term of the repeat sales regressions (2) above. Goetzman (1993) and Mei and Moses (2002) use the coefficient on the second stage Case and Shiller regressions as an estimate of r 2 , but this requires an i.i.d. assumption on the errors of the repeat sales regressions.…”

“…However, the volatility of the art market was very high during this period due to the run-up in prices in 1989 (seeAshenfelter and Graddy 2003 or Mei andMoses 2002) and the volatility of Contemporary Art, even relative to Impressionist and Modern Art, was especially high during the end of the 1990s. Furthermore, due to taste changes for art over time, cross-country and cross-market comparisons are more likely to be valid than comparisons over 10-year time periods.…”