In 2025, it depends. Particularly if you are a Republican/Lean Republic — then it’s almost zero…

Here’s the relationship of different groups disaggregated by partisan affiliation, against news sentiment (lexographically defined from newspapers), as measured by the SF Fed.

Figure 1: U.Mich Sentiment – Democratic/Democratic leaning (blue, left scale), and SF Fed News Sentiment index (tan, right scale). October News Sentiment is through 10/26. Source: U.Michigan, and SF Fed.

Figure 2: U.Mich Sentiment – Independent (chartreuse, left scale), and SF Fed News Sentiment index (tan, right scale). October News Sentiment is through 10/26. Source: U.Michigan, and SF Fed.

Figure 3: U.Mich Sentiment – Republican/Republican leaning (red, left scale), and SF Fed News Sentiment index (tan, right scale). October News Sentiment is through 10/26. Source: U.Michigan, and SF Fed.

The sensitivity of sentiment to news sentiment is 0.88 for Democratic/Lean Democratic, 0.66 for independents (both coefficients statistically significant at conventional levels, and adj-R² of 0.70 and 0.82, respectively). For Republican/Lean Republican, the coefficient is -17, and is not statistically significant. The adj-R² is 0.02. In other words, measured news sentiment is unrelated to Republican/Lean Republican economic sentiment.

One interesting aspect of the results is that Republican/Lean Republican expectations are positively associated with news sentiment, while current conditions are negatively associated (the worse the news, the better the assessment of current conditions). This is not true for Democratic/Lean Democratic or Independents. Here’s a scatterplot.

Figure 4: U.Mich Current conditions – Republican/Republican leaning vs. SF Fed News Sentiment index, 2025M01-2025M10. Source: U.Michigan, and SF Fed.