مدلسازی تاثیر تغییر سازمان مبتنی بر پیش بینی در تورم ایالات متحده   Modeling the impact of forecast-based regime switches on US inflation

1. Introduction Lucas (1976) showed that macroeconometric models with constant parameters cannot be used for evaluating policy changes, since policy changes usually lead to behavioral changes by economic agents, which result in inconstant model parameters. It is well known that agents also react to macroeconomic forecasts. This suggests that unexpected economic forecasts may also lead to changes in the model parameters. Several theoretical and empirical studies have indicated this effect of forecasts. Theoretically, Fellner (1976) explained that the public’s expectations are prone to selfjustifying skepticism about policy makers, and policy makers react to that. Empirically, Givoly and Lakonishok (1979) found that serious upward revisions in financial earnings forecasts have significant effects on stock prices. Steiner,Großand Entorf (2009) showed that asset prices demonstrate an immediate reaction to returns in macroeconomic announcements. Moreover, they find that the reactions to positive news are faster than those to negative news. Sinclair, Gamber, Stekler, and Reid (2012) showed that forecast errors have an impact on the target interest rate set by the Federal Reserve Bank. Although the literature suggests that forecasts have an impact in various fields, this paper focuses on US inflation time series data. It is well known that the dynamic character of this series is affected by policy changes; see for example Cogley and Sargent (2002, 2005) and Primiceri (2005). Furthermore, inflation forecasts play an important role, since (i) policy makers react to forecasts due to the FED Volcker-regime inflation targeting (Clarida, Galí, & Gertler, 2000); and (ii) companies and consumers use inflation forecasts to decide upon future savings and expenditure levels. Carroll (2003) states that people update their expectations to public forecasts rather than to past inflation rates. Furthermore, economic theory also provides support for the impact of forecasts on the inflation rate. It is mainly mentioned as either the expectations trap (Christiano & Gust, 2000) or self-fulfilling expectations,where the public’s expectations of high inflation increase the actual inflation rate. Albanesi, Chari, and Christiano (2003) stated that ‘‘expectations of high or low inflation lead the public to take defensive actions, which then make accommodating those expectations the optimal monetary policy’’. Both the expectations trap before 1979 (Leduc, Sill, & Stark, 2007) and inflation targeting since the 1980s suggest that inflation forecasts play a key role.