Herding in Money Demand

Big Players influence the demand for money in ways not easily captured by traditional monetary theory. Most monetary economists have been in search of a well-specified model that will allow them to track accurately the demand for money and predict its future movements. By the early 1970s, a general consensus had emerged among economists about the functional form of money demand. This form seemed highly stable in the period after the Second World War. 2 At about the same time the Federal Reserve System began a long series of changes in its policies and procedures. As argued below, these changes produced greater Fed activism and discretion. Shortly thereafter money demand estimates became unstable, and the emerging data began to expose shortcomings in the accepted models. Most economists came to believe these problems with tracking money demand were a product of misspecification of the models. Since the 1970s, the game has been one of discovering which measure of income or wealth to use, which measure of risk, and which measure of liquidity, as well as some way to account for advances in payment technologies and institutional change. Yet, no model to date has been robust over time. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.