Monetary Policy When Preferences Are Quasi-Hyperbolic

We study discretionary monetary policy in an economy where economic agents have quasi-hyperbolic discounting. We demonstrate that a benevolent central bank is able to keep inflation under control for a wide range of discount factors. If the central bank, however, does not adopt the household's time preferences and tries to discourage early consumption and delayed-saving, then a marginal increase in steady state output is achieved at the cost of a much higher average inflation rate. Indeed, we show that it is desirable from a welfare perspective for the central bank to quasi-hyperbolically discount by more than households do. Welfare is improved because this discount structure emphasizes the current-period cost of price changes and leads to lower average inflation. We contrast our results with those obtained when policy is conducted according to a Taylor-type rule