Kathryn Graddy

We provide evidence for the behavioral biases of anchoring and loss aversion in paintings sold at auction. We find that anchoring is more important for items that are resold quickly and that the effect of loss aversion increases with the time that a painting is held. This evidence contributes significantly to the empirical evidence of the endowment effect: of increasing loss aversion with the length of holding. However, we do not find evidence that investors can take advantage of these behavioral biases.

In order to investigate the role of immigrant artists on the development of artistic clusters in U.S. cities, we use the U.S. Census and American Community Survey, collected every 10 years between 1850 and 2010. We identify artists, authors, musicians, actors, architects, and journalists, their geographical location and their status as a native or an immigrant. We look at the relative growth rate of the immigrant population in these occupations over a ten year period and how it affects the relative growth rate of native-born individuals in these artistic occupations. We find that cities that experienced immigrant artist inflows, also see a greater inflow of native artists.

Does creativity, on average, increase or decrease during bereavement? Dates of death of relatives and close friends of 33 French artists and 15 American artists were gathered from electronic sources and biographies, and information on over 15,000 paintings was collected from the Blouin Art Sales Index and the online collections of the Metropolitan Museum of Art, the Art Institute of Chicago, the National Gallery of Art, the J. Paul Getty Museum, and the Musée d’Orsay, including more than 12,000 observations on price. An event study indicates that there is no evidence that the death of a friend or relative makes an artist more creative, and there is some evidence that prices of paintings are significantly lower during the first year following the year of death of a friend or relative. Furthermore, paintings that were created during this bereavement period are less likely to be included in a major museum’s collection.

This paper addresses the question of whether auction guarantees change the bidding environment and whether they cause a change in price once the value of an item is taken into account. We analyze both the cases of third-party guarantees and in-house guarantees. We use two datasets: one of Christie’s and Sotheby’s Contemporary and Impressionist Evening Sales from January 2010 to February 2012 and another larger dataset consisting of all items auctioned at Christie's from 2001 to May 2011. While more expensive items are more likely to have guarantees, we find little empirical evidence to suggest an effect on price from the guarantee itself once the value of the item is taken into account.

Color plays an important part in modern life and influences our decision making process. However, little is known about how the different attributes of color, namely intensity and lightness, influence price. By analyzing auction data for paintings we can put a price on these attributes of color. Using a unique set of data for Contemporary artworks of Andy Warhol prints, we are able to observe the influence of intensity and lightness using RGB values as explanatory variables on prices achieved at auction. Controlling for other hedonic characteristics, our empirical results find significant evidence of intense colors fetching a premium over equivalent artworks which are less intense in color. Furthermore, darkness carries a premium over lightness.

Roger de Piles (1635–1709) was a French art critic who decomposed the style and ability of 58 different artists into areas of composition, drawing, color, and expression, rating each artist on a 20-point scale in each category. Based on evidence from two data sets that together span from the mid-eighteenth century to the present, this article shows that De Piles' overall ratings have withstood the test of a very long period of time, with estimates indicating that the works of his higher-rated artists achieved both greater returns and higher critical acclaim than the works of his lower-rated artists.

This article proposes an alternative specification for the second stage of the Case‐Shiller repeat‐sales method. This specification is based on serial correlation in the deviations from the mean one‐period returns on the underlying individual assets, whereas the original Case‐Shiller method assumes that the deviations from mean returns by the underlying individual assets are i.i.d. The methodology proposed in this article is easy to implement and provides more accurate estimates of the standard errors of returns under serial correlation. The repeat‐sales methodology is generally used to construct an index of prices or returns for unique, infrequently traded assets such as houses, art and musical instruments, which are likely to be prone to exhibit serial correlation in returns. We demonstrate our methodology on a data set of art prices and on a data set of real estate prices from the city of Amsterdam.

This paper measures the returns to investing in violins, using two different datasets. One dataset includes 337 observations on repeat sales of the same violins at auction and at dealer sales starting in the mid‐nineteenth century, and another dataset includes over 2,500 observations on sales of individual violins at auction since 1980. Overall, real returns for the dataset on repeat sales for the period 1850–2008 have been approximately 3.5%. Real returns to the overall portfolio of individual sales since 1980 have been about 3.3%. The price path has been stable with a slight negative correlation to stocks and bonds. (JEL D44, G11, L82)

We estimate a dynamic profit-maximization model of a fish wholesaler who can observe consumer characteristics, set individual prices, and thus engage in third-degree price discrimination. Simulated prices and quantities from the model exhibit the key features observed in a set of high quality transaction-level data on fish sales collected at the Fulton Fish Market. The model's predictions are then compared to the case in which the wholesaler must post a single price to all retailers. We find the added revenue the wholesaler receives from price discriminating to be small.

The failure of many paintings to sell in art auctions indicates the presence of reserve prices set by sellers. This paper examines the relationship between sale rates and price surprises over time in art auctions. Using data on contemporary and impressionist art, we show that while sale rates appear to have little relationship to current prices, there exists a strong positive relationship of sale rates to unexpected aggregate price changes, which is reminiscent of a Phillips curve. As a result, sale rates provide a useful quantity indicator of the strength of the art market. The data also indicate that sale rates revert to "normal" very quickly following a price surprise. We estimate an empirical model to measure normal sale rates. We also find evidence that the reserve price is set on average at about 70% of the auctioneer's low estimate, as published in the auction catalog.