Eye of the beholder? Art buyers discount female artists

Erasmus School of Economics

A study of more than a million art auction sales across 40 years has revealed that works by women artists attract prices that are nearly 50% lower than those for works by men, on average. Even when auction results of “stars” such as Leonardo da Vinci were not taken into consideration, female art pieces were still sold at a large discount, the researchers found. Their research shows that this discount has nothing to do with talent or thematic choices. It is solely because of the gender of the artists.

The research lookes at sales of artworks by 67,000 individual artists. Even when sales over $1 million – the Rembrandts and the Picassos – were taken away, there was still a 29% discount for female artists, the researchers say, which are Renée Adams from UNSW, Roman Kräussl from Luxembourg School of Finance, Marco Navone from the UTS Business School and Patrick Verwijmeren from Erasmus School of Economics.

Their results reveal the struggle and cultural bias facing female artists in getting recognition and true compensation for their work. The research examined 1.5 million auction transactions in 45 countries between 1970 - 2013 and found a 47.6% price gap between male and female artists across all sales. To understand the influence of cultural perceptions of gender, the researchers looked at auction prices in conjunction with measures of gender inequality around the world, including the number of women in parliament and women’s enrolment in higher education. They found that the price gap for female artists was greater in countries and years with greater gender inequality and the price gap decreased as gender inequality decreased, suggesting that cultural attitudes towards women are a key factor.

“The art auction market can have a profound influence on artists’ careers, and art markets are discounting paintings by women on the basis of gender,” the researchers say. To further understand these reasons behind the price difference, the researchers conducted two experiments. In the first experiment the researchers asked 1000 people to guess the gender of the artist for 10 works, half of which were painted by women, then rate how much they liked the paintings on a scale of 1-10. This allowed them to measure whether perceived gender might affect a person’s appreciation of the work. The results showed that participants were unable to accurately guess the gender of the artist by looking at the painting. However, artworks perceived to be painted by women were rated lower by participants who are male, affluent and who visit art galleries.

In the second experiment the researchers showed 2000 people a painting with either a male or a female artist’s name randomly assigned to the artwork, and asked them to rate the painting. To avoid associating fake artist names with real paintings, the researchers ‘created’ the paintings using an artificial intelligence application that converts a photo into a painting. Most people rated the artworks equally, however one category, wealthy individuals who visit art galleries frequently – those who would typically buy art at auction – gave the painting a lower rating when it was associated with a female name. According to the researchers, these results reveal the struggle female artists face to have their art valued on equal footing with male artists, and suggest that policies to reduce gender inequality may improve outcomes for female artists.

Furthermore, the research exposes the potential sample selection bias of some art indices, which employ repeat sales to calculate annualized returns for various segments of the art market. Artworks that have significant price appreciation are more likely to result in higher returns relative to the population at large, leading to a sample selection bias that must be adjusted accordingly, yielding a much less profitable investment profile. The dataset consisted of 2.3 million paintings sold between 1960 and 2010. All identifiable repeat sales were sought out, leaving out indeterminate matches, buy-ins, and other outliers, resulting in 32,928 paintings with a total of 69,103 repeat sales. Of this population, less than two percent have been sold more than once in the 50-year period. From the other 98 percent, many paintings remain in private estates and have been gifted to museums. Many other paintings could have been sold privately, but even then, less than two percent is a very small percentage, especially during a period that is seen as one of the longest bull markets in art history. Another reason could be that some of the paintings have not been reoffered because they would incur a loss. This loss could be the result of high transaction costs, which questions the profitability of art as an investment. 

More information

Patrick Verwijmeren is Professor of Corporate Finance at Erasmus School of Economics. His research focuses on the optimal capital structure for businesses, the influences of short sales and hedge funds and recent developments in corporate governance. His main specialism is business financing through convertible bonds.

Click here for the entire research paper in the Sydney Morning Herald, 21 January 2018.

You can find an article about the research in The Economist, 12 May 2019, attached.

You can find an article about the research in Belgian newspaper De Standaard, 23 May 2019, attached.

You can find an article about the research in the Italian Elle, 17 June 2019, here.

Read more about investing in art and the research in Artsy, 24 June 2019, here