Working Paper: CEPR ID: DP16046
Authors: Michael Ehrmann
Abstract: Inflation targeting is implemented in different ways – most often by adopting point targets, by having tolerance bands around a point target, or by specifying target ranges. Using data for 20 economies, this paper tests whether the various target types affect the anchoring of inflation expectations at shorter horizons differently. It tests two contradictory hypotheses, namely that targets with intervals lead to (i) less anchoring, e.g. because they provide more flexibility to the central bank, or (ii) better anchoring, because they are missed less often, leading to an enhanced credibility. The evidence refutes the first hypothesis, and generally finds that target ranges or (in some cases) tolerance bands outperform the other types. However, the effects partially depend on the economic context and no target type consistently outperforms all others. This suggests that there are some benefits to adopting intervals, but the central bank can anchor inflation expectations also by other means.
Keywords: inflation targeting; inflation expectations; point target; tolerance band; target range
JEL Codes: E52; E58; E31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| target ranges (Y10) | anchoring of inflation expectations (E31) |
| tolerance bands (C46) | anchoring of inflation expectations (E31) |
| point targets (C54) | anchoring of inflation expectations (E31) |
| credibility enhancement hypothesis (D83) | anchoring of inflation expectations (E31) |
| target ranges (Y10) | credibility enhancement (D83) |
| missing a target range (E61) | credibility (D83) |
| target ranges (Y10) | forecaster disagreement (C53) |
| inflation expectations (E31) | responsiveness to realized inflation (E31) |
| flexibility hypothesis (D29) | anchoring of inflation expectations (E31) |