Empirical work documents substantial disagreement in inflation expectations obtained from survey data. Furthermore, the extent of such disagreement varies systematically over time in a way that reflects the level and variance of current inflation. This paper offers a simple explanation for these facts based on asymmetries in the forecasters' costs of over-and under-predicting inflation. Our model implies biased forecasts with positive serial correlation in forecast errors and a cross-sectional dispersion that rises with the level and the variance of the inflation rate.It also implies that forecast errors at different horizons can be predicted through the spread between the short-and long-term variance of inflation. We find empirically that these patterns are present in inflation forecasts from the Survey of Professional Forecasters. A constant bias component, not explained by asymmetric loss and rational expectations, is required to explain the shift in the sign of the bias observed for a substantial portion of forecasters around 1982.

In this paper, we examine the effect of having an inflation targeting framework on the dispersion of inflation forecasts from professional forecasters. We use a panel data set of 25 countries-including 14 inflation targeters-with 16 years of monthly information. We find that the dispersion of long-run inflation expectations is smaller in targeting regimes after controlling for country-specific effects, time-specific effects, the level and the variance of inflation, disinflation periods, and global inflation. On average, the full effect is not observed until the third year after implementation of inflation targeting. When we differentiate between developed and developing countries, the dispersion of inflation expectations after inflation targeting is smaller and statistically significant only in developing countries. Copyright (c) 2010 The Ohio State University.

RESUMENEn este documento se analiza el traspaso de movimientos en el tipo de cambio a diferentes índices de precios en México. El análisis se basa en un modelo de vectores autorregresivos (VAR) para datos mensuales de enero de 1997 a diciembre de 2010. Los efectos del traspaso se calculan utilizando los impulsos respuesta acumulados de un choque al tipo de cambio que se identifica mediante el método recursivo. Los resultados muestran que el traspaso a los precios de las importaciones es completo, pero que el grado de traspaso disminuye a lo largo de la cadena de distribución, de tal manera que el efecto en los precios al consumidor es menor a 20%. Además, se encuentra que el traspaso a los precios al consumidor parece disminuir considerablemente a partir de 2001, lo cual coincide con la adopción del esquema de objetivos de inflación por parte del Banco de México.

Combination of forecasts from survey data is complicated by the frequent entry and exit of individual forecasters which renders conventional least squares regression approaches infeasible. We explore the consequences of this issue for various combination methods in common use and propose a new method that projects actual outcomes on the equal-weighted forecast to adjust for biases and noise in the underlying forecasts.Through simulations and an application to inflation forecasts we show that the entry and exit of individual forecasters can have a large effect on the real time performance of conventional combination methods. The proposed projection works well in practice.

Empirical work documents substantial disagreement in inflation expectations obtained from survey data. Furthermore, the extent of such disagreement varies systematically over time in a way that reflects the level and variance of current inflation. This paper offers a simple explanation for these facts based on asymmetries in the forecasters' costs of over-and under-predicting inflation. Our model implies biased forecasts with positive serial correlation in forecast errors and a cross-sectional dispersion that rises with the level and the variance of the inflation rate.It also implies that forecast errors at different horizons can be predicted through the spread between the short-and long-term variance of inflation. We find empirically that these patterns are present in inflation forecasts from the Survey of Professional Forecasters. A constant bias component, not explained by asymmetric loss and rational expectations, is required to explain the shift in the sign of the bias observed for a substantial portion of forecasters around 1982.