Essays on the role of belief formation for asset prices and the macroeconomy

This thesis is structured around two main chapters which study the role of individuals' beliefs in affecting the asset markets and the macroeconomy. The first chapter of my thesis is based on a joint work with Paul Ehling and Christian Heyerdahl-Larsen in which we study asset prices and portfolio choice when agents learn from their own experience. We develop a general equilibrium overlapping generations economy with learning in which cohort specific experience drives beliefs about consumption growth and through that affects equilibrium outcomes. The main findings are that the young act as trend chasers, wealth shifts from young to old, the market price of risk is countercyclical and the risk free rate of return is pro-cyclical. Importantly, the model also produces a negative relation between mean forecast about consumption growth with future realized returns. The latter is consistent with a growing empirical literature that shows how survey expectations of returns negatively predict future returns across countries and financial instruments. In the second chapter, I explore how variations in information precision influence decision making through the formation of beliefs. In this regard, I develop a general equilibrium model with production in which the representative agent has incomplete information about productivity growth and learns about it through a noisy signal. The precision of the signal is allowed to change and its variations are a source of aggregate uncertainty which I call confidence risk. The key findings are that the model produces fluctuations in macroeconomic variables such as consumption and investment that are disconnected from fundamental shocks, that is, even when productivity growth is set at its unconditional average. The model also generates stock market crashes that are not driven by large moves in macroeconomic variables