This paper investigates the price determinants and investment performance of art. We apply a hedonic regression analysis to a new data set of more than one million auction transactions of paintings and works on paper. Based on the resulting price index, we conclude that art has appreciated in value by a moderate 3.97% per year, in real U.S. dollar terms, between 1957 and 2007. This is a performance similar to that of corporate bonds---at much higher risk. A repeat-sales regression on a subset of the data demonstrates the robustness of our index. Next, quantile regressions document larger average price appreciations and higher volatilities in more expensive price brackets. We also find variation in historical returns across mediums and movements. Finally, we show that measures of high-income consumer confidence and art market sentiment predict art price trends.

This paper was accepted by Wei Xiong, finance.

/pdf/buying-beauty-on-prices-and-returns-in-the-art-market-3gydf7j5uq.pdf

Buying Beauty: On Prices and Returns in the Art Market

This paper investigates the price determinants and investment performance of art. We apply a hedonic regression analysis to a new data set of more than one million auction transactions of paintings and works on paper. Based on the resulting price index, we conclude that art has appreciated in value by a moderate 3.97% per year, in real U.S. dollar terms, between 1957 and 2007. This is a performance similar to that of corporate bonds--at much higher risk. A repeat-sales regression on a subset of the data demonstrates the robustness of our index. Next, quantile regressions document larger average price appreciations (and higher volatilities) in more expensive price brackets. We also find variation in historical returns across mediums and movements. Finally, we show that measures of high-income consumer confidence and art market sentiment predict art price trends.

This paper considers the question of how to measure and interpret the financial return to investing in collectibles. We review various methodologies for creating priced indexes and then discuss studies that explicitly calculate a rate of return to some set of collectibles. While most collectibles appear to yield positive real returns, the majority embody more risk and yield lower financial returns than stocks. Other characteristics of collectibles prices, such as covariance with other asset prices, are also examined.

https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.13.4.193

Measuring Returns on Investments in Collectibles

In this study, 37 605 paintings by 60 well-known Australian artists sold at auction over the period 1973–2003 are used to construct a hedonic price index The attributes included in the hedonic regression model include the name and living status of the artist, the size and medium of the painting and the auction house and year in which the painting was sold The resulting index indicates that returns on Australian fine-art averaged 7 per cent over the period with a standard deviation of 16 per cent The hedonic regression model also captures the willingness to pay for perceived attributes in the artwork, and this shows that works by McCubbin, Gascoigne, Thomas and Preston and other artists deceased at the time of auction, works executed in oils or acrylic, and those auctioned by Sotheby's or Christie's are associated with higher prices [ABSTRACT FROM AUTHOR]

https://papers.ssrn.com/sol3/Delivery.cfm/ecor_237.pdf?abstractid=725897&mirid=5

Financial Returns and Price Determinants in the Australian Art Market, 1973-2003

In this study, 37 605 paintings by 60 well-known Australian artists sold at auction over the period 1973–2003 are used to construct a hedonic price index. The attributes included in the hedonic regression model include the name and living status of the artist, the size and medium of the painting and the auction house and year in which the painting was sold. The resulting index indicates that returns on Australian fine-art averaged 7 per cent over the period with a standard deviation of 16 per cent. The hedonic regression model also captures the willingness to pay for perceived attributes in the artwork, and this shows that works by McCubbin, Gascoigne, Thomas and Preston and other artists deceased at the time of auction, works executed in oils or acrylic, and those auctioned by Sotheby's or Christie's are associated with higher prices. [ABSTRACT FROM AUTHOR]

/pdf/financial-returns-and-price-determinants-in-the-australian-2hc1zc3ty2.pdf

Financial Returns and Price Determinants in the Australian Art Market, 1973–2003*

Past studies on art investment, generally have found that returns are low and risk is high. In this study we find that the return to art investment is more in line with traditional investments and thus the cost for consumption associated with art seems fairly small. Employing a large sample of paintings by 66 American artists sold at auction between 1971–1992, average returns are found to be over 9 per cent and 3 per cent in nominal and real terms, respectively. The model employs a log linear price regression estimated by pooled cross section and time series data, and allows rates of return as well as hedonic values for various painting and auction attributes to be estimated. These include size, media, age of execution, authenticity of the work, and auction month and house. In addition, rates of return are differentiated by artist, time period of investment, price range and genre of the painting. The findings indicate significant sensitivity of rates of return to these data stratifications.

Renée K. Pierce

Papers

Financial returns, price determinants, and genre effects in American art investment