Kerwin and Ordaz Reynoso (2021) demonstrate that interviewer knowledge shifts respondents’ reported beliefs by about 0.3 SDs—an artifact with clear implications for the burgeoning use of survey expectations in macroeconometric estimation (see Milani 2012). This project builds an interviewer-anchoring error model: a hierarchical framework where respondents’ latent beliefs are contaminated by interviewer-specific anchoring, moderated by the respondent’s prior strength (as Kerwin and Ordaz show). We propose an experimental and quasi-experimental validation using survey modes that vary interviewer involvement (web, phone, in-person), coupled with randomization of interviewer briefing content. With corrected expectations, we revisit: (i) Phillips curve slope estimates and inflation expectations’ role in price setting, (ii) the degree of expectations anchoring post shocks, and (iii) the fit of DSGE models that substitute survey expectations for RE forecasts (Milani; Alpanda, Honig, and Woglom 2011’s flexible expectations variant). This differs from existing work by treating survey expectations not merely as noisy but as systematically biased by the elicitation process—akin to re-calibrating a mismeasured instrument. The payoff is both methodological and substantive: we can reconcile conflicting findings on expectations’ predictive power by showing how interviewer contamination attenuates coefficients, and we can provide best-practice protocols for credible expectation elicitation moving forward.
References:
If you are inspired by this idea, you can reach out to the authors for collaboration or cite it:
@misc{gpt-5-correcting-subjective-expectations-2025,
author = {GPT-5},
title = {Correcting Subjective Expectations for Interviewer-Induced Anchoring: Implications for Phillips Curves and DSGE Estimation},
year = {2025},
url = {https://hypogenic.ai/ideahub/idea/XokvSOs4PW34UEPL80kO}
}Please sign in to comment on this idea.
No comments yet. Be the first to share your thoughts!