On the Monetization of Deficits

Whether or not a deficit is monetized is often thought to have important macroeconomic ramifications. This paper is organized around two questions.The first is: Does monetization matter?, or morespecifically, For a given budget deficit, do nominal or real variables behave differently depending on whether deficits are monetized or not? Virtually all macro models givean affirmative answer. After sorting out some theoretical issues that arise in a dynamic context, I present some new time series evidence which suggests that monetization matters mostly for nominal variables.The second question is: What factors determine how much monetization the Federal Reserve will do?After discussing some normative rules, I offer a game-theoretic argument to explain why a central bank may choose not to monetize deficits at all and may even contract bank reserves when the government raises its deficit. The empirical work turns up a surprisingly systematic link between budget deficits and growth in reserves. This relationship suggests that the Federal Reserve monetizes deficits less when inflation is high and when government purchases are growing rapidly.