Quality selection in the consumption basket of urban and rural households: the comparison of the Hackman’s two-step method and the approach of Shankowiler and Yen

It is argued that application of the inverse of mill's ratio as a correction factor regression in Heckman’s two-step method leads to inconsistent estimates and the resulted elasticities are not reliable. To avoid inconsistency of sample selection bias, Shonkwiler & Yen (1999) introduce an alternative method. In this paper results of the two-stage Heckman method (1979) are compared with the estimated elasticities by Shonkwiler and Yen.  We use Iranian households' survey data to estimate demand elasticities for meat. Interestingly, the results of both methods are similar and there are no inconsistencies in the two-stage Heckman method. Both of the approaches show that, economic factors have greater impact on the quality of demand for meat than demographic factors. Also, estimated coefficient for all variables have the similar sign by the methods, but the magnitude of the resulted effects are larger for Shonkwiler and Yen estimators.