Monetary News Shocks

• Date: 01 October 2020
• Author: HASHMAT KHAN,NADAV BEN ZEEV,CHRISTOPHER GUNN
• DOI: http://doi.org/10.1111/jmcb.12686
• Published date: 01 October 2020

DOI: 10.1111/jmcb.12686

NADAV BEN ZEEV

CHRISTOPHER GUNN

HASHMAT KHAN

Monetary News Shocks

We pursue an empirical strategy to identify a monetary news shock in the

U.S. economy.We use a monetary policy residual, along with other variables

in a vector autoregression (VAR), and identify a monetary news shock as

the linear combination of reduced-form innovations that is orthogonal to the

current residual and that maximizes the sum of contributions to its forecast

error varianceover a finite horizon. Real GDP declines in a persistent manner

after a positive monetary news shock. This contraction in economic activity

is accompanied by a fall in inflation and a rapid increase in the nominal

interest rate.

JEL codes: E32, E52, E58

Keywords: monetary news shocks, monetary policyresidual, federal funds

rate, forward guidance, DSGE models.

OVER AT LEAST THE PAST 30 years there has been a signifi-

cant effort in macroeconomics to identify the causal effects of monetary policy on

economic activity by studying the impact of shocks to central bank rate decisions.1

These shocks are interpreted as unanticipated or “surprise” deviations of a central

bank’s policy rate from that implied by a systematic policy rule. Yet, recent evidence

suggests that at least some of these policy deviations may have not been surprises

Wethank two anonymous referees for detailed comments and suggestions. We also thank Kosuke Aoki,

Paul Beaudry, Alex Cukierman, Zvi Eckstein, Giovanni Gallipoli, Francisco Gonzalez, Zvi Hercowitz,

Victoria Hnatkovska, Anna Kormilitsina(discussant), Jesse Perla, Franck Portier, Yaniv Yedid-Levy, and

participants at presentations at the Canadian Macro Study Group Meetings (Kingston, Canada, 2016),

Bank of Israel, Money Macro and Finance Conference (Bath, 2016), Canadian Economic Association

Meetings(Toronto, 2015), Mid-WestMacro Meetings (Rochester, 2015), Interdisciplinary Center Herzliya,

Universit´

e de Sherbrooke, University of Tokyo, Vancouver School of Economics, and the University of

Windsor for helpful comments.

NADAV BEN ZEEV is at the Department of Economics, Ben-Gurion University of the Negev, Israel

(E-mail: nadavbz@bgu.ac.il). CHRISTOPHER GUNN is at the Department of Economics, Carleton University,

Ottawa, Canada (E-mail: chris.gunn@carleton.ca). HASHMAT KHAN is at the Department of Economics,

Carleton University, Ottawa, Canada (E-mail: hashmat.khan@carleton.ca).

Received February 27, 2017; and accepted in revised form May 21, 2019.

1. See Ramey (2016) for a detailed overview.

Journal of Money, Credit and Banking, Vol. 52, No. 7 (October 2020)

C

2019 The Ohio State University

1794 :MONEY,CREDIT AND BANKING

after all (see, e.g., Gurkaynak, Sack, and Swanson 2005, Campbell et al. 2012, and

Nakamura and Steinsson 2015). Rather, a portion of federal open market commit-

tee (FOMC) rate decisions appear to be anticipated by private agents in advance of

the formal rate-setting meetings. While a handful of recent papers have examined

the impact of anticipated components of rate changes on both asset markets and on

economic activity, relatively little work has been done to identify anticipated shocks

using the structural vector autoregression (SVAR) framework.

The main objective of our paper is to fill this gap. Weidentify anticipated monetary

policy shocks, or monetary news shocks, and determine their effectson the U.S. econ-

omy during the Greenspan–Bernanke era of Federal Reserve Chairmanship. These

shocks represent pure changes in expectations about the nonsystematic component of

future monetary policy, orthogonal to nonsystematic policy in the present. Building

on the news shocks view (Beaudry and Portier 2006), our approach is in the spirit of

the empirical literature that provided a benchmark for evaluating the effects of sur-

prise monetary policy shocks in theoretical New Keynesian models (e.g. Christiano,

Eichenbaum, and Evans 1999).2We see our paper as a first step toward developing

this empirical benchmark for monetary news shocks.

Our approach involves first constructing a monetary policy residual (MPR) that

measures deviations from an estimated interest rate rule that tracks the observed

federal funds rate well during 1988-2007.3Next, we propose a restriction to identify

monetary news shocks using the maximum forecast error variance (MFEV) approach

within the SVAR framework similar to Barsky and Sims (2011), Francis et al. (2014),

and Ben Zeev and Khan (2015). We identify a monetary news shock as the linear

combination of reduced-form innovations orthogonal to current policy residual that

maximizes the sum of contributions to the policy residual’s forecast error variance

over a finite horizon. The information set that we consider includes credit spreads

and interest rate futures in the VAR. In our baseline specification, we also impose

orthogonality with respect to credit spreads to ensure that monetary news shocks are

distinct from credit supply shocks, as measured by the excess bond premium variable

from Gilchrist and Zakrajˇ

sek (2012).

A potential source of monetary news shocks is the practice of forward-guidance

through which a central bank provides information about the future course of mon-

etary policy via formal statements or releases (Rudebusch 2008, den Haan 2013,

Svensson 2015). One of its forms—the “Odyssean” forward-guidance— which im-

plies a commitment to a future action, is viewed as a type of a monetary news shock

(Campbell et al. 2012). Yet, it need not be limited to that. Informal comments by

policymakers to the media outside of regularly scheduled meetings, implicit com-

munication implied by changes in the structure or membership of monetary policy

2. Other examples on identifying unanticipated monetary shocks include Bernanke and Mihov (1998),

Romer and Romer (2004), and Barakchian and Crowe (2013).

3. The baseline sample start date is due to the availability of federal funds futures contract data from

1988. We choose 2007 as the sample end date to first establish results without the influence of a variety

of considerations in the post-2007 data (such as the financial crisis and the Great Recession [2008–09],

Quantitative Easing, and the zero-lower-bound).