Thesis
Essays on expectation formation
- Abstract:
-
This thesis is composed of three independent papers that model patterns of heterogeneity in inflation expectations and price-setting.
In the first paper, I present stylized facts from the Survey of Professional Forecasters that suggest the inflation expectation formation process changes over the term of the forecast. I develop a model to capture these features: forecasters have (1) a cost for inacccuracy, for which they pay the penalty when the variable is realized and (2) a cost for inconsistency (changing a forecast from quarter to quarter), for which they pay the penalty when the adjustment is reported. Due to heterogeneous time discounting, the inconsistency cost is more highly but differentially weighted at long horizons, which induces the observed stylized facts.
In the second paper, I estimate a sector-level Rational Inattention (RI) price-setting model. I estimate the model separately on two regimes (pre-Volcker and Great Moderation) to investigate predictions from RI theory about price-setting reactions to a change in volatility. I find that the model is not capable of reproducing price level dynamics, and that the attention capacity has low predictive power across regimes. Given these caveats, there is some evidence that attention capacity increased across the sample.
In the third paper, I consider two kinds of uncertainty: the mean and the variance of the prior distribution of a stochastic variance. I build a model with strategic complementarity and provide comparative statics welfare results. Higher mean-uncertainty in the private signal always decreases welfare, and higher mean-uncertainty in the public signal sometimes decreases welfare. However, higher variance-uncertainty in the private signal always increases welfare, and higher variance-uncertainty in the public signal sometimes increases welfare. This finding contributes to the debate as to whether this class of model supports limited transparency among policy makers.
Actions
Access Document
- Files:
-
-
(Preview, pdf, 919.0KB, Terms of use)
-
Authors
- DOI:
- Type of award:
- DPhil
- Level of award:
- Doctoral
- Awarding institution:
- University of Oxford
- UUID:
-
uuid:327d208f-4dce-4b41-b920-85b93f469e87
- Deposit date:
-
2018-05-17
- ARK identifier:
Terms of use
- Copyright holder:
- Lauderdale, K
- Copyright date:
- 2017
If you are the owner of this record, you can report an update to it here: Report update to this record