In a new report reviewing last year’s fall auction season, veteran art market analysts Michael Moses and Jianping Mei have said it was “the worst overall financial performance” for the art market this century. However, several art market experts have contested that assessment, telling ARTnews that they think the report is flawed by “sample selection bias” and gives a narrow, unnuanced view of the art market.
“For the first time this century, the mean of all individual returns was negative,” reads Moses and Mei’s report, “How Bad Was the Fall 2024 Auction Season? Hard to Believe, But Worse Than the Spring,” which was provided to ARTnews ahead of publication. “Fifty-two percent of the works sold produced a negative return.”
This latest report follows an analogous one released last year by the duo, who run JP Mei & MA Moses Art Market Consultancy, that similarly declared that the spring 2024 auction season was the worst in nearly 25 years, saying it was “financially as bad as it gets.” For the spring report, the duo analyzed about 50,000 repeat sales of artworks at Christie’s, Sotheby’s, and Phillips over the last 25 years. Only works first purchased at any worldwide auction from 1970 were included. The report used each repeat sale to compute the compound annual return (CAR) of the fluctuation in price over time between purchase and sale. According to the research, the mean return for repeat sale pairs of artworks last spring was almost zero, the lowest since 2000.
For the fall sales, they adopted the same methodology and analyzed around 56,000 repeat sales. “For the first time this century the mean of all the individual semi-annual returns was negative (-0.1 percent),” the report says.
“I was very surprised by the fall results,” Moses told ARTnews in a recent interview. “I thought the art market might start to turn, but it clearly hasn’t quite hit the floor. The question becomes, like in stock parlance, are you attempting to catch a falling knife?”
Now retired, Moses was previously a professor at New York University’s Stern School of Business. Mei is a professor at Beijing’s Cheung Kong Graduate School of Business. Their new findings come after Christie’s recently announced its 2024 sales across both art and luxury had dropped by 8 percent compared to 2023, while Sotheby’s revealed this month that its 2024 consolidated sales were down 23 percent compared to 2023.
Clare McAndrew, an economist who pens the annual Art Basel and UBS Global Art Market Report, as close to industry-wide economic data as the art market has, told ARTnews that she is skeptical of Moses and Mei’s findings. Using only repeat sales, she said, is “sample selection bias.”
“You’re only accounting for works that have been sold twice, and you’re ignoring all the single sales, which eliminates a whole batch of contemporary and other material,” she said. “Say a work has been sold ten times, by galleries or privately before being sold at auction, [Moses and Mei’s report] doesn’t take this into account, either. There’s a whole swathe of sales that are not included in this methodology.”
McAndrew suggested that only surveying results at the three major auction houses—Christie’s, Sotheby’s, and Phillips—is further sample selection bias. Artworks in the report’s dataset for repeat sales could have been sold at other auction houses numerous times.
“There’s nothing wrong with the technique used [by Moses and Mei], but it doesn’t necessarily explain how the art market is going,” McAndrew said. “They’re making out that everything is bad, that’s just not necessarily true.”
New York-based art advisor Adam Green, who presents the ArtTactic podcast, told ARTnews that the report shouldn’t be taken at face value. “While market analyses, such as the one conducted by Mei Moses, provide useful insights into overall trends, the art market is highly nuanced,” he said. “Many distinct sub-markets exist not only by collecting category but also at the artist level, making overarching assessments only somewhat relevant. Although the overall market softened this past year, the full picture is more complex.”
When asked about McAndrews’ suggestion of sample selection bias, Moses responded, “We analyzed almost 1,000 artists who have at least 10 repeat sales pairs each, so it’s a pretty broad selection.”
When asked if only using repeat sales from the top three auction houses gave too narrow a picture of the art market, Moses likened that focus to stock market analysts focusing on the S&P 500, and added that sales at those houses counts for 95 percent of the dollar volume of public art sales worldwide.
“The most important factor is to try and add transparency to the market,” Moses said. “And the only place you get true transparency is at auction and/or sales that have tax associated with them. Other than that, we can talk about the market and total revenue and things like that, but it really doesn’t say much about the financial returns available in the art market.”
Christie’s, Sotheby’s, and Phillip’s declined to comment on the report.
