Rank correction: a new approach to differential nonresponse in wealth survey data

This paper is concerned with the problem of modelling the tail of the wealth distribution with survey data in the context of differential nonresponse. In order to deal with the problem post data collection, it is standard practice to combine wealth survey data with observations from rich lists and then fit a Pareto tail. In contrast, our approach does not require information about individual wealth holdings from rich lists and is thus applicable in situations where such information is not available. Applying the procedure to wealth survey data (HFCS, SCF, WAS) yields estimates of top wealth shares, which are closely in line with estimates from the World Inequality Database and thus represent a likely improvement over the raw survey data.