Signaling Effects of Monetary Policy

By Leonardo Melosi

We develop a DSGE model in which the policy rate signals to price setters the central bank’s view about macroeconomic developments. The model is estimated with likelihood methods on a U.S. data set that includes the Survey of Professional Forecasters as a measure of price setters’ inflation expectations. We find that the model fits the data better than a prototypical New Keynesian DSGE model because the signaling effects of monetary policy help the model account for the run-up in inflation expectations in the 1970s. The estimated model with signaling effects delivers large and persistent real effects of monetary disturbances even though the average duration of price contracts is fairly short. While the signaling effects do not substantially alter the transmission of technology shocks, they bring about deflationary pressures in the aftermath of positive demand shocks. The signaling effects of monetary policy have contributed (i ) to heightening inflation expectations in the 1970s, (ii ) to raising inflation and to exacerbating the recession during the first years of Volcker’s monetary tightening, and (iii ) to subduing inflation and to stimulating economic activity from 1991 through 2007.

The model’s premise is that the central bank is more informed than the rest of the economy. Thus anything the central bank does is a signal about the true state of the economy, and the other economic agents observe and act on it. This is welfare improving, and could be even more so if the central bank could be credibly announcing the true state of the economy, and I am not sure you need the model’s price rigidities to get this result. As recent evidence shows, it is not that easy for the central bank to be that informed. That would likely make the model’s results less pronounced, and if the central bank is really confused could lead to welfare losses. But these are my conjectures.


2 Responses to Signaling Effects of Monetary Policy

  1. M. H. says:

    The model uses the Calvo fairy for price rigidity. Another good reason to get rid of rigidity.

  2. Leonardo Melosi says:

    Two clarifications on the assumption of perfect information of the central bank and on the issue of nominal rigidities.

    1) The central bank does NOT have to be perfectly informed for the signaling effects of monetary policy to arise in the model. The three necessary ingredients are: (1) firms have incomplete information; (2) firms observe the interest rate set by the central bank (or a noisy signal thereabout); (3) it is common knowledge that the central bank’s information set is different from firms’. In fact, in the paper, the central bank does not signal any truth but just its own view about the macroeconomic fundamentals (i.e., the history of aggregate shocks). It should be noted that potential mistakes that the central bank makes in estimating the current inflation rate and the current output gap are captured by the variable \eta_{r,t}.

    2) The assumption of nominal rigidities is by no means crucial for any results of the paper. In fact, the estimated model features an average duration of price contract that is very short. The only reason for having nominal rigidities is to induce a bit of forward-looking behaviors in the price-setting problem. Importantly, the likelihood favors the assumption of dispersed information to explain the persistence in the macro data. It should be noted that dispersed information generates persistence in the model because the strategic complementarity in price setting makes it optimal for price setting firms not only to respond to their beliefs about the aggregate shocks but also to their higher-order beliefs. This mechanism was termed by Robert Townsend “forecasting the forecasts of others” and is capable to generate a high degree of persistence in DSGE model through the sluggish adjustments of average higher-order beliefs. See also the following link:

    The welfare effects of providing a public signal in an economy with dispersed information are analyzed in Morris and Shin “The Social value of Public Information” (AER, 2003). Their results apply to the model of this paper. Morris and Shin find that the provision of public signals may have ambiguous effects on welfare. In particular, a high degree of strategic complementarity among players may lead to a welfare loss because the provision of noisy public signals might crowd out private information of better quality. While I believe that the provision of the noisy policy signal in the estimated model of this paper is welfare improving, this exercise would make an interesting paper.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: