Quantifying survey expectations: A critical review and generalization of the Carlson–Parkin method

This paper provides a critical review of the popular Carlson–Parkin (CP) quantification method using household-level data from the University of Michigan’s Survey of Consumers. We find strong evidence against the threshold constancy, symmetry, homogeneity, and overall unbiasedness assumptions of the CP method. To address these violations, we generalize the CP method using a hierarchical ordered probit (HOPIT) model. By comparing the quantified inflation expectations with quantitative expectations obtained from the same set of households directly, we show that the generalized model performs better than the CP method. In particular, when the CP unbiasedness assumption is replaced by a time-varying calibration, the resulting quantified series is found to track the quantitative benchmark well, over diverse time periods.