Abstract

Statistics South Africa’s Quarterly Labour force Survey (QLFS)/Labour Market Market Dynamic Series (LMDS) public release data contain pre-populated imputations for bracket and non-response labour market earnings. Up to now it has not been possible to assess how these imputations were derived. This work sheds insights into the current hot deck imputation used by Statistics South Africa and how it produces different estimates of distributional statistics which are sensitive to the approach currently used which underestimates key economic indicators such as income distribution and income percentiles with implication of economic policies such as the minimum wage. An alternative earnings series is presented with implications for researchers to be aware of these limitations when using the public release wage data. This research is important for users of labour statistics and labour market researchers.