The Optimal Degree of Monetary-Discretion in a New Keynesian Model with Private Information
This paper considers the optimal degree of discretion in monetary policy when the central bank conducts policy based on its private information about the state of the economy and is unable to commit. Society seeks to maximize social welfare by imposing restrictions on the central bank’s actions over time, and the central bank takes these restrictions and the New Keynesian Phillips curve as constraints. By solving a dynamic mechanism design problem we find that it is optimal to grant “constrained discretion” to the central bank by imposing both upper and lower bounds on permissible inflation, and that these bounds must be set in a history-dependent way. The optimal degree of discretion varies over time with the severity of the time-inconsistency problem, and, although no discretion is optimal when the time-inconsistency problem is very severe, our numerical experiment suggests that no-discretion is a transient phenomenon, and that some discretion is granted eventually.
How much flexibility should society allow a central bank in its conduct of monetary policy? At the center of the case for flexibility is the argument that central bankers have private information (Canzoneri, 1985), perhaps about the economy’s state or structure, or perhaps about the distributional costs of inflation arising through heterogeneous preferences (Sleet, 2004). If central banks have flexibility over policy decisions, then this gives them the ability to use for the public’s benefit any private information that they have. However, if central banks face a time-inconsistency problem (Kydland and Prescott, 1977), then it may be beneficial to limit their flexibility. Institutionally, many countries have balanced these competing concerns by delegating monetary policy to an independent central bank that is required to keep inflation outcomes low and stable, often within a stipulated range, but that is otherwise given the freedom to conduct policy without interference. Inflation targeting is often characterized as “constrained discretion” (Bernanke and Mishkin, 1997) precisely because it endeavors to combine flexibility with rule-like behavior.