摘要：We use a monetary overlapping-generations model to discuss the cause and
durability of the marked fall in the volatility of inflation in recent decades.
In our model, agents have to forecast inflation, and they do so using two
"heuristics." One is based on lagged inflation, the other on an inflation target
announced by the central bank. Agents switch between those heuristics based on
an imperfect assessment of how each has performed in the past. The way the
economy propagates productivity shocks into inflation depends on the proportion
of agents using each heuristic. Movements in these proportions generate
fluctuations in small-sample measures of economic volatility. We use this simple
model of heuristic switching to contrast the performance of monetary policy
rules. We find that, relative to the rule that would be optimal under rational
expectations, a rule that responds to both productivity shocks and inflation
expectations better stabilizes the economy but does not prevent agents from
switching between heuristics. Finally, we study the impact of introducing an
explicit inflation target, which can be used by agents as a simple heuristic,
into an economy that did not previously have one. Depending on the heuristics
agents have access to before the introduction of the target, this can result in
reduced inflation volatility.
