Thesis: The first chapter questions the conventional intuition that a high concentration of income at the top of the distribution should promote the emergence of rational asset bubbles. I use an OLG model with financial fictions and heterogeneous agents that differ in terms of savings rate, portfolio choices and skills. I show that a high concentration at the top promotes the emergence of asset bubbles if and only if those asset bubbles are illiquid or financial markets are arbitrage-free. Instead, if asset bubbles are liquid and liquid assets pay a premium under illiquid assets, a low concentration promotes the emergence of asset bubbles. The second chapter studies the circumstances under which asset bubbles are expansionary in an OLG-New Keynesian that includes capital. I show that secular stagnation is a necessary but not sufficient condition. Indeed, asset bubbles stimulate investment, consumption and output if and only if there's a strong shortage of aggregate demand. Finally, the third paper shows that "standard" New Keynesian models make puzzling predictions when aggregate demand is chronically deficient they predict a secular boom, and seeks to understand how those models must be adjusted to analyze secular stagnation. I emphasize the crucial role of the long run elasticities of asset demand and supply with respect to the output gap in general equilibrium; and I also connect the secular boom to other puzzling predictions of the New Keynesian model.