Expectations play a central role in macroeconomics. How they are modelled not only shapes predictions about the decisions of economic agents, but also influences the equilibrium properties of the model. This thesis consists of three self-contained chapters, each contributing to the broad theme of expectations and equilibrium in macroeconomics. The first chapter studies the heterogeneous attention choices of households and firms. Using the rational inattention framework, I show that households find it optimal to pay more attention to supply shocks, because these shocks most affect their real income, while firms optimally pay more attention to demand shocks due to their larger impact on profits. The model reconciles survey evidence on heterogeneous beliefs across households and firms, generates state-dependent Phillips curve slopes, and shows that central bank communication that fails to consider the heterogeneous attention choices can backfire. The second chapter, co-authored with Guido Ascari, moves to the financial markets. Investors are assumed to possess limited memory and to form expectations in a time-varying way. The former guarantees a tendency to revert to fundamentals. The latter induces momentum in asset prices and is motivated by a novel empirical observation about a time-varying mapping from price-dividend ratio to return expectations in survey data. Using the simulated method of moments, we show that the model matches a host of asset-pricing features, and generates survey-consistent subjective investor beliefs. The third chapter, co-authored with David Murakami and Ivan Shchapov, shows that the presence of an occasionally binding constraint from the effective lower bound (ELB) in New Keynesian models often leads to multiple or no equilibria. The problem stems from a strong feedback loop between expectations (of inflation and output) and current outcomes at the ELB. We show that simple fiscal policy rules can introduce additional stabilising forces that dampen this loop, thereby ensuring the existence and uniqueness of an equilibrium.