Monetary Policy and Stock Prices: Does the “Fed Put” Work When It Is Most Needed?

• Author: Chen Gu,Alexander Kurov
• Date: 01 December 2016
• DOI: http://doi.org/10.1002/fut.21790
• Published date: 01 December 2016

Monetary Policy and Stock Prices:

Does the “Fed Put”Work When

It Is Most Needed?

Alexander Kurov* and Chen Gu

Moststudies of the effectof monetary policyon asset prices usethe event study methodologywith

daily data. Theresulting estimates sufferfrom bias due to omitted variablesand endogeneity of

policy decisions. We provide evidence that this bias becomes so large during the 2007–2008

financialcrisis that it reverses thesign of the estimated stock marketresponse to monetary news,

leadingto an erroneousconclusion thatinterest rate cutsare bad news for stocks. We alsoexamine

thestock market reactionto monetary policyduring the zero lowerbound period. The resultsshow

asignificant bias in daily event study estimatesof the stock market response to news about the

futurepath of monetary policy.© 2016 Wiley Periodicals,Inc. Jrl Fut Mark 36:1210–1230,2016

1. INTRODUCTION

We contribute to the literature on the relation between monetary policy and equity prices.

Obtaining accurate estimates of the responsiveness of stock prices to monetary policy is

important for investors and monetary policy makers. Most previous studies looking at the

response of stock returns to monetary policy use the event study methodology, regressing

daily stock returns on unexpected changes in the policy rate. For example, Bernanke and

Kuttner (2005) show that an unexpected 25-basis-point cut in the fed funds target rate leads

to a 1% increase in the level of stock prices on average.

1

Using event study regressions with daily data to estimate the response of stock returns to

changes in monetary policy produces biased estimates. There are two reasons for that. First,

monetary policy is endogenous and reacts to stock returns (e.g., Rigobon & Sack, 2003).

Second, both monetary policy and stock returns react simultaneously to economic news.

Rigobon and Sack (2004) show that the bias in event study estimates is generally modest.

Alexander Kurov is an Associate professor of Finance in the Department of Finance, West Virginia University,

Morgantown, West Virginia. Chen Gu is a doctoral student in Finance in the Department of Finance, West

Virginia University, Morgantown, West Virginia. We thank Arabinda Basistha, Alexandros Kontonikas, Ron

Spahr, Harry Turtle, participants at the 2015 Financial Management Association Conference, seminar

participants at West Virginia University, Robert Webb (the editor) and two anonymous referees for helpful

comments and suggestions. Errors or omissions are our responsibility.

JEL Classification: E44, E52, E58, G14, G18

*Correspondenceauthor, Departmentof Finance, College of Business andEconomics, West Virginia University,P.O.

Box 6025, Morgantown,West Virginia 26506. Tel: 304-293-7892,Fax: 30-293-3274, e-mail: alkurov@mail.wvu.edu

Received September 2015; Accepted March 2016

1

Other examples of studies in this area include Basistha and Kurov (2008), Ehrmann and Fratzscher (2004), and

Wongswan (2009).

The Journal of Futures Markets, Vol. 36, No. 12, 1210–1230 (2016)

© 2016 Wiley Periodicals, Inc.

Published online 20 May 2016 in Wiley Online Library (wileyonlinelibrary.com).

DOI: 10.1002/fut.21790

However, this bias becomes larger if the variance of economic shocks increases, the variance

of stock return innovations increases or the response of monetary policy to the stock market

becomes stronger. We argue that all of these conditions are likely to exist during periods of

crisis, potentially leading to incorrect inferences regarding the responsiveness of stock prices

to monetary shocks.

Kontonikas, MacDonald, and Saggu (2013) argue that a structural shift in the relation

between monetary policy actions and stock returns occurred during the financial crisis of

2007–2008. Using event study methodology with daily data, they show that the relation

between unexpected changes of fed funds target rate and stock returns became positive

during the crisis, suggesting that interest rate cuts were viewed by the market as a signal of

worsening economic outlook. Florackis, Kontonikas, and Kostakis (2014) and Gregoriou,

Kontonikas, MacDonald, and Montagnoli (2009) find similar results for the relation between

monetary policy changes and stock returns in the U.K. Based on these results, Florackis et al.

(2014) and Kontonikas et al. (2013) conclude that conventional monetary policy becomes

ineffective during the crisis when the policy rate is close to the zero lower bound.

Many investors believe that the Federal Reserve will support financial markets,

including the stock market, in periods of market stress. The Fed considers stock market

conditions when making policy decisions for two reasons. First, stock prices influence

economic activity. For example, higher stock prices increase household wealth, leading to

higher consumer spending. Higher stock prices also reduce cost of capital for firms, spurring

investment. Second, stock returns contain timely information about the state of the

economy. The Fed’s pattern of reacting to market declines by easing monetary policy has

been labeled the “Fed put,”often named after the current Fed chairman, such as “Greenspan

put”or, more recently, the “Bernanke put.”

2

The finding of prior studies that stock prices

decline in response to unexpected interest rate cuts during the financial crisis suggests that

the “Fed put”becomes ineffective when it is most needed.

Using intraday futures data to mitigate endogeneity and omitted variable biases, we

show that, if anything, the negative relation between monetary shocks and stock returns

becomes stronger during the crisis, indicating that interest rate cuts remain good news for

stocks. These results are supported by estimates from daily time series regressions designed to

control for endogeneity and omitted variable bias. Therefore, monetary policy remains a

useful tool for alleviating market stress in periods of crisis.

2. SAMPLE SELECTION AND METHODOLOGICAL ISSUES

2.1. Sample Selection

We examine a sample period from January 1994 to October 2015. Our sample period starts in

1994 because that is when the Federal Open Market Committee (FOMC) started to

announce its decisions immediately after the meeting. Before 1994, policy changes had to be

inferred from the open market operations following FOMC meetings. G€urkaynak, Sack, and

Swanson (2007) argue that since 1994 the fed funds futures rate became a better predictor of

the future target rate, allowing for more precise estimation of monetary policy shocks.

After the FOMC meeting of December 16, 2008 the fed funds target rate has been at the

zero lower bound. In the program known as quantitative easing, the Fed used large-scale bond

purchases to influence longer-term interest rates. The Federal Reserve also influenced

medium-term yields by providing increasingly detailed guidance about the future path of the

fed funds target rate. Therefore, we divide the sample period into two sub-periods. The

2

See, for example, Poole (2008) for a good discussion of the issues involved.

Monetary Policy and Stock Prices 1211