Affective Sensitivity and Inflation Expectations across the Income Distribution

Households do not form inflation expectations with information alone; emotions matter too. This paper introduces an affective channel into the limited-attention model of expectation formation. A household’s economic mood—the emotional component of consumer sentiment that macroeconomic fundamentals cannot explain—shapes how firmly it holds on to its past inflation beliefs. Attention to new information works as in the benchmark. We estimate the model on Michigan Survey microdata from 1978 to 2025. A more positive mood weakens belief persistence in both income groups. Low-income households, however, are about two and a half times as sensitive as high- income households. The affective channel substantially improves the fit of the model and lowers long-run expected inflation below the limited-attention benchmark. The results are robust across estimators, measures of mood and demographic groups, and in an extension where mood also shapes how current inflation is read. For monetary policy, the households most exposed to inflation are also those whose expectations are most sensitive to their emotional state.