From the introduction:
The paper estimates survival over time in two broad ways. The measure I am most interested in employs auction results from the major and secondary auctions in the U.S., England and Europe during the period 1987 to 1997. I assume that an artist survives the test of time if his work shows up at auction 50 or more years after his work appeared in either the Paris, Armory or World’s Fair exhibitions. Auction sales indicate market survival because they provide evidence that there is still sufficient interest in the artist among collectors to support a secondary market in his works. To anticipate the empirical analysis, it turns out that the majority of the artists in the exhibitions did not appear at auction during the 1987 to 1997 period. Moreover, there are vast differences among the artists who did survive in the number and value of works sold at auction. For many of the surviving artists, s few works were sold at auction over the ten year period for very low prices. At the other extreme, a small number of artists accounted for the bulk of auction sales. These differences are clearly of interest and are addressed in the paper. For example, the records of the three exhibitions provide me with bibliographical and other information on the individual artists—including some measures of their reputations at the time of the exhibitions—which I can relate to survival differences.
The other estimate of survival relies on non-market factors that, for simplicity, I call “critical assessment” indicators. Here I gathered data from both the on-line edition of the 34 Volume Grove Dictionary of World Art published in 1996 (the print edition costs $8,800) and a Website called “AskArt.com” that specializes in American art.