News, Noise, and Fluctuations: An Empirical Exploration

Abstract

We explore empirically models of

News Shocks in the Data: Olympic Games and Their Macroeconomic Effects

“…Most of the existing evidence for the presence of anticipation in macroeconomics is indirect. Anticipation effects present serious challenges to empirical research (see also Blanchard, L'Huillier, and Lorenzoni ).…”

“…They find that output, investment, and hours fall slightly after a news shock while consumption increases. Blanchard, L'Huillier, and Lorenzoni () show that the use of structural identification assumptions in a VAR setting is sensitive to the informational assumptions made and that SVAR shocks may be perfectly anticipated by the consumers. These authors point to the adoption of more structural models for the identification of news shocks.…”

Macroéconomie et information imparfaite

“…Au lieu de cela, lorsque les agents économiques n'observent pas les chocs structurels en temps réel, l'économètre, confronté aux mêmes données que les agents économiques, peut ne pas être en mesure d'identifier correctement les chocs (Blanchard et al, 2013). Pour pouvoir identifier correctement les chocs structurels, l'ensemble d'information de l'économètre doit donc être plus large que l'ensemble d'information des agents économiques 4 .…”

“…Certaines de ces intuitions ont été appliquées à l'étude empirique des effets des chocs de nouvelles informations sur l'innovation technologique et des fluctuations non fondamentales du cycle économique (voir par exemple Barsky et Sims, 2012 ;Blanchard et al, 2013et Forni et al, 2013 ; à l'effet des chocs conventionnels de politique monétaire (Hubert, 2017 ;Hubert et Maule, 2016 ;Miranda-Agrippino et Ricco, 2017), des chocs non conventionnels de politique monétaire (Andrade et Ferroni, 2017), et des chocs budgétaires (Ricco, 2015 ;Ricco et al, 2016).…”

Information rigidities and the news-adjusted output gap

“…A novel decomposition method is then described to investigate the agents' use of this new information, separating out that part which is expected to have a permanent e ect on output from that part expected to have more short-lived e ects. The methods are illustrated using data from the U.S. Survey of Professional Forecasters over the last 45 years and their usefulness is illustrated through the construction of a`news-adjusted' output gap measure that is purged of the e ects of output movements that are known 1 See for example, Ball et al (2005) and the recent work by Blanchard et al (2013), Kulish and Pagan (2013) and Lorenzoni (2010), among others.…”