LIQUIDITY, MONETARY POLICY, AND UNEMPLOYMENT: A NEW MONETARIST APPROACH

• Date: 01 May 2019
• Author: Mei Dong,Sylvia Xiaolin Xiao
• DOI: http://doi.org/10.1111/iere.12374
• Published date: 01 May 2019

INTERNATIONAL ECONOMIC REVIEW

Vol. 60, No. 2, May 2019 DOI: 10.1111/iere.12374

LIQUIDITY, MONETARY POLICY, AND UNEMPLOYMENT: A NEW

MONETARIST APPROACH∗

BYMEI DONG AND SYLVIA XIAOLIN XIAO1

University of Melbourne, Australia; Peking University, China

We discover a consumption channel of monetary policy in a model with money and government bonds. When

the central bank withdraws government bonds (short-term or long-term) through open market operations, it

lowers returns on bonds. The lower return has a direct negative impact on consumption by households that hold

bonds and an indirect negative impact on consumption by households that hold money. As a result, firms earn

less profits from production, which leads to higher unemployment. The existence of such a consumption channel

can help us understand the effects of unconventional monetary policy.

1. INTRODUCTION

We develop a model with money and government bonds to study how a change in the

supply of government bonds through open market operations (OMOs) affects consumption

and unemployment through a consumption channel. Conventional monetary policy generally

targets some short-term interest rates by conducting OMOs. During the recent Great Recession,

targeted short-term interest rates in several advanced economies have been cut close to zero.2

This limits a central bank’s ability to further lower the short-term interest rate to stimulate the

economy. Instead of targeting short-term interest rates, the central banks of the United States,

Japan, and some European countries all conducted unconventional monetary policy by either

purchasing long-term government bonds or other government-guaranteed private securities in

financial markets. The goal is to directly lower long-term interest rates in financial markets.

An important feature of conventional and unconventional monetary policy is that they are

essentially about adjusting the supply of government bonds through OMOs. Hence, it is neces-

sary for a model to have money and bonds in order to understand the effects of such policy. A

few recent papers in monetary theory including Williamson (2012) and Rocheteau et al. (2018)

consider multiple assets, but most of these models do not have unemployment. Given that one

∗Manuscript received February 2016; revised August 2017.

[Correction added on 8 March 2019, after first online publication: Sylvia Xiaolin Xiao was added as one of the

corresponding author.]

1We thank Guido Menzio and three referees for insightful comments. We also thank Chris Edmond, Pedro Gomis-

Porqueras, Susumu Imai, Benoit Julien, Chao Gu, Tim Kam, Gordon Menzies, Guillaume Rocheteau, Jeff Sheen,

Chi Chung Siu, John Wooders, Stephen Willamson, Randall Wright, Yajun Xiao, and seminar participants at Deakin,

Monash, RMIT, Tsinghua, Adelaide, Melbourne, UNSW, Queensland, UTS, UW-Madison, St. Louis Fed, the 26th

Australian PhD Conference in Economics and Business, Sydney Macro Reading Group, and the 2014 Chicago Fed

Summer Workshop on Money, Banking, Payments, and Finance for helpful comments. Dong acknowledges financial

support under Australian Research Council’s DECRA scheme (project number DE120102589). Xiao acknowledges

the hospitality and support when visiting UW-Madison, St. Louis Fed, and Melbourne, and the funding support from

Dan Searle Fellowship for her postdoc at UW-Madison. Please address correspondence to: Mei Dong, Department of

Economics, the University of Melbourne, 111 Barry Street, Carlton, 3053, Victoria, Australia. Phone: 61-3-8344-1209.

Fax: 61-3-83446899. E-mail: mei.dong@unimelb.edu.au. Sylvia Xiaolin Xiao, Guanghua School of Management, Peking

University, Beijing, China, 100871. Phone: +86 10 6275 7492. E-mail: sylvia.xiao@gsm.pku.edu.cn.

2In 2008, the U.S. Federal Reserve cut the Federal Funds Rate to zero. This zero-lower-bound problem is also

observed in Japan and some European countries. As early as in 1995, the Bank of Japan cut the short-term interest

rate almost to zero, which lasted until 2016. After the Great Recession, the European Central Bank cut the short-term

target rate to zero.

1005

C

(2018) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

1006 DONG AND XIAO

of the U.S. Federal Reserve’s mandated objectives is the achievement of “full employment,”

we integrate a labor search model into a microfounded monetary model with multiple assets.

There are two key elements of our model. The first is that money and bonds are valued

by households because they can facilitate transactions in the goods market. The coexistence

of money and bonds makes the model suitable to consider OMOs as central banks’ swaps of

government bonds and money. Monetary policy can affect households’ portfolio decisions if

the policy changes the relative return of these assets. The second key element is that the labor

and goods markets are connected. Firms bring production from the labor market for sale in

the goods market. Households can purchase goods for consumption in the goods market using

money or government bonds. This link between the labor and goods markets provides a channel

through which monetary policy affects unemployment.

Our model builds on Berentsen et al. (2011). We first add short-term government bonds in

addition to money. The central bank can adjust the supply of short-term bonds as a monetary

policy instrument, that is, OMOs. We find that different cases of monetary equilibrium exist

depending on the relative supply of short-term bonds. OMOs can affect the economy only

when the supply of bonds is scarce, but not too scarce. In this case, the return on bonds is

higher than that on money. Households that have access to bonds use only bonds to trade

in the goods market, and households that do not have access to bonds use money. When the

central bank reduces the supply of government bonds by purchasing bonds, the price of bonds

increases and the short-term interest rate decreases. For households that use bonds, the lower

interest rate directly induces them to hold fewer bonds and consume less. As a result, firms’

profits from selling to these households decrease. In the labor market, lower profits discourage

firms from entering and raise unemployment. In the goods market, households face fewer

trading opportunities, and this will lower the marginal benefit of holding money. This general

equilibrium effect indirectly makes households that use money hold less money and consume

less, which further lowers firms’ profits and raises unemployment.

We focus on the effects of OMOs through a consumption channel. That is, the change in the

supply of bonds affects the return on bonds and consumption by bond holders. This in turn can

affect labor market outcomes, which further affects the return from holding money. Therefore,

real balances also respond to OMOs. Our model provides a clear transmission channel of

OMOs to the real side of the macroeconomy. A permanent decrease in the supply of bonds

has a negative impact on employment. The conventional view is that a central bank’s purchase

of government bonds would lower interest rates and thus stimulate investment. Our model

does not have this investment channel. As this investment channel has been identified in recent

studies such as Rocheteau and Rodriguez-lopez (2014), we argue that the consumption channel

is complementary to the investment channel.

During the Great Recession, several central banks chose to purchase long-term government

bonds or other government-guaranteed private securities. In order to address the effects of

this unconventional policy, we extend the basic model by adding long-term government bonds.

We consider the long-run effects of unconventional policy where the central bank changes

the supply of long-term government bonds. When the short-term interest rate is close to zero,

the central bank can buy or sell long-term government bonds to directly adjust the long-term

interest rate. Through the consumption channel, unconventional policy lowers the long-term

interest rate and households’ consumption. As a result, equilibrium unemployment increases.

We find that a positive lower bound on the long-term interest rate exists because long-term

bonds are less liquid than short-term bonds.

We contribute to the New Monetarist literature by providing a framework to analyze how

monetary policy, especially OMOs, affects unemployment. We find that OMOs may not affect

consumption and unemployment when the supply of government bonds is either too low or too

high. When OMOs have a real effect, the consumption channel indicates that the central bank’s

withdrawal of either short-term or long-term government bonds lowers consumption and raises

unemployment. Moreover, there exists a positive lower bound on the long-term interest rate if

a central bank changes the supply of long-term bonds.