A good monetary policy rule for stabilizing the economy involves using simple three-parameter rules that are very effective at minimizing fluctuations in inflation, output, and interest rates. These rules smooth the interest rate response to shocks and use feedback from anticipated policy actions to stabilize inflation and output. It is better to react to a multi-period inflation rate rather than the current quarter rate, and targeting the price level instead of the inflation rate involves only small additional stabilization costs. These findings hold true even with uncertainty in parameters and models, and when nominal interest rates cannot go below zero. However, if expectations are not influenced by policy, the performance of efficient policies may worsen.