Despite the centrality of party identification in U.S. politics, the effects of partisanship on public opinion remain elusive. In this article, we use monthly economic opinion data disaggregated by partisanship to evaluate the role of party identification on economic perceptions. Using both static and time-varying error correction models, we find strong evidence of partisan bias in the public’s assessment of the state of the economy, and importantly, this bias changes over time. This evidence of the changing influence of partisanship helps reconcile some of the different findings of individual and aggregate level opinion studies. We also examine how the time-varying influence of partisanship affects aggregate public opinion. Specifically, we show that the increased influence of partisanship has led aggregate economic perceptions to respond more slowly to objective economic information.