Key points are not available for this paper at this time.
The authors analyze the substitution bias in Laspeyres-type price indexes, using 101 c ommodities. They construct the tightest theoretical bounds on the cos t-of-living index using nonparametric methods and also construct supe rlative price indexes. Using nonparametric methods, the authors find homothetic preferences are consistent with the data. Sensitivity test s indicate that this result is not vacuous. Under this hypothesis, th e bias is between 0.22 and 0.14 percent per year. Superlative indexes imply a bias of about 0.18 percent. This estimate is somewhat larger than found in earlier studies. Commodity aggregation is found to be a major contributor to measured substitution bias. Copyright 1988 by The Econometric Society.
Manser et al. (Fri,) studied this question.