Munich Personal RePEc Archive

Expectations' Dispersion & Convergence towards Central Banks' IR forecasts: Chile, Colombia, Mexico, Peru & United Kingdom, 2004-2014

Barrera Chaupis, Carlos (2016): Expectations' Dispersion & Convergence towards Central Banks' IR forecasts: Chile, Colombia, Mexico, Peru & United Kingdom, 2004-2014.

[img]
Preview
PDF
MPRA_paper_85410.pdf

Download (937kB) | Preview

Abstract

The study evaluates the effect of both the publication of Inflation Report (IR)’s forecasts and the subsequent media diffusion efforts (made by 5 central banks) on (i) the dispersion of ‘fixed-event’ forecasts for inflation and real growth produced by the macroeconomic insiders of a country (and gathered by Consensus Economics, Inc.), as well as (ii) the distance between their median and the aforementioned official forecasts. The 5 central banks correspond to the monetary authorities in Chile, Colombia, Mexico, Peru and United Kingdom. Statistically testing the effects on the dispersion and distance uses a common sample of monthly forecasts from 2004 to 2014 and reach high specificity by using separate samples according to the forecasting horizon (short and medium ‘term’) and the macroeconomic uncertainty level (IR publication months are classified as either high- or low-uncertainty months). With a significance level of 10 per cent, the general results are that (a) increases and decreases in the dispersion can be attributed to either IR forecast publication or media diffusion; and (b) increases and decreases in the distance can be attributed to either IR forecast publication or media diffusion, although the number of increases in the distance is low relative to (a). Comment from the author: It would be interesting to add results for more countries. Specifically, I was planning to add Canada and New Zealand. However, in the case of New Zealand, the corresponding series from Consensus Economics, Inc. is actually not available near Peru for the whole sample (the nearest one is actually located at the British Library!). There exists a critique addressing the econometric approach: it is related to the idea of causality and the need to use the difference-in-difference approach (this implies the need to include data from non-inflation-targeting countries). I am totally satisfied with the paper, though. In a nutshell, I consider more important to address the issue as if I were a medicine doctor wondering about whether the temperature is normal, high or low for the specific cases of 5 individuals instead of digressing about what is "normal temperature" for (say) 40 individuals.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.