The more you know, the better: A Heckman repeat-sales price index

The Repeat-Sales (RS) model controls quality by using the prices of the same items in various time periods. However, this methodology suffers from sample selection bias. To address this issue, a Heckman two-stage procedure has been applied to a sample of Picasso paintings from 1990 to 2016. Contrary to the literature — which argues sample selection bias comes from buy-ins — in this paper, bias is modelled to capture information on items sold only once. The research field frequently uses the standard (RS) index, yet empirical evidence shows systematic differences between the standard RS index and the Heckman RS index. These differences are relevant when buyers make decisions on art investments.