Doubly heterogeneous monetary spillovers

• Published date: 01 August 2022
• Author: Nihar Shah
• Date: 01 August 2022
• DOI: http://doi.org/10.1111/infi.12410

Received: 24 December 2020

|

Accepted: 21 January 2022

DOI: 10.1111/infi.12410

ORIGINAL ARTICLE

Doubly heterogeneous monetary spillovers

Nihar Shah

2560 Rutherford Court, Fremont,

California, USA

Correspondence

Nihar Shah, 2560 Rutherford Court,

Fremont, CA 94539, USA.

Email: nihar.shah@post.harvard.edu

Funding information

Harvard University; Bank of Canada;

Rumsfeld Foundation

Abstract

Monetary spillovers are heterogeneous in two ways:

how central banks generate them and how countries

receive them. First, the Fed is mostly unique in its

ability to affect other countries' financial markets,

among ten developed central banks. This is noteworthy

given the lack of data on other central banks' spillovers.

This paper makes public a novel data set of these ten

central banks' monetary shocks to support future

research. Second, the Fed affects recipient countries

in different ways, with the bonds and currencies of

countries with high‐interest rates reacting differently

than those of low‐rate countries. This can help shed

light on theories around the Fed's spillovers, and this

paper demonstrates how the exact pattern is

inconsistent with models in which developed central

banks react to the Fed.

KEYWORDS

high‐frequency identification, international financial markets,

monetary spillovers

JEL CLASSIFICATION

E44, E52, E58, F42, G14, G15

1|INTRODUCTION

Spillovers of the Federal Reserve's monetary policy into foreign financial markets have become

a fixture of the international financial system. Foreign central bankers routinely blame the Fed

for chaos in local markets, while Fed governors like Jerome Powell and Lael Brainard reiterate

their commitments to domestic objectives. In academic circles, Hélène Rey at the Jackson Hole

Symposium first identified the Fed‐driven ‘global financial cycle’which threatens to swamp

countries without capital controls, and the argument has since spurred a large literature.

International Finance. 2022;25:126–150.wileyonlinelibrary.com/journal/infi126

|

© 2022 John Wiley & Sons Ltd.

This paper explores two forms of heterogeneity in monetary spillovers: how the central

banks of developed countries generate them, and how the financial markets of developed

countries receive them. Using high‐frequency data and methodologies robust to market noise,

I first establish a novel fact on how currencies and bonds react heterogeneously to the

Fed's announcements. When the Fed tightens, the dollar appreciates more against currencies of

high‐interest rate countries (e.g., Australia) than against currencies of low‐interest rate

countries (e.g., Japan), and long‐maturity bond yields of high‐rate countries rise more than

those of low‐rate countries. Second, I establish that these effects are unique to the Fed. No other

central bank in my sample of 10 banks, apart from the European Central Bank, generates such

spillovers; and even then, the European Central Bank's transmission has very different

properties than the Fed's.

These facts are important for future research on spillovers, as successful models must

explain both the heterogeneity in how countries receive spillovers and in how countries

generate them. In support of this, I take two steps. First, this paper constructs the monetary

shocks emanating from the 10 central banks from public data, so that they can be freely shared

to support future work, which is especially important given the total lack of data beyond the

Fed and European Central Bank (ECB). Second, I illustrate the value of these findings using

one example: I argue against the popular theory in which foreign central banks are forced to

accommodate the Fed's actions. The asymmetries I identify in how currency and bond markets

react to the Fed are inconsistent under this theory and thus offer evidence against it.

The patterns in currency markets suggest that investors expect the central banks of low‐rate

countries to tighten most when the Fed tightens, while investors in bond markets suggest the

central banks of high‐rate countries will tighten most.

To illustrate the paper's contributions, consider an example. At 12:30 PM on January 25,

2012, the Fed announced its intentions to keep interest rates low until 2014. The surprise easing

affected foreign assets in heterogeneous ways, as 60‐min windows around the announcement

show in Figure 1. Yields on 10‐year Australian bonds immediately fell whereas yields on

10‐year Japanese bonds did not. Moreover, the dollar depreciated more against the Australian

dollar than against the yen, or equivalently, the Australian dollar appreciated against the yen.

As in this example, I first establish how currencies and bonds respond heterogeneously to

Fed announcements. Across nine developed countries—Australia, Canada, the Eurozone,

Japan, Norway, New Zealand, Sweden, Switzerland, and the United Kingdom—and

announcements from 2001 to 2016, I find the following: the dollar appreciates or depreciates

most against currencies in countries with high‐interest rates historically (e.g., Australia) and

least against currencies in countries with low‐interest rates historically (e.g., Japan) when the

Fed tightens or eases, respectively. At the same time, when the Fed tightens or eases, long‐

maturity bond yields from historically high‐rate countries rise or fall more than long‐maturity

bond yields from historically low‐rate countries, respectively; this is not driven by the zero

lower bound.

Moreover, the Fed is unique among other central banks in such regards. The central banks

of eight of the other nine countries generate almost no spillovers, except with small effects on

neighbours (e.g., the Reserve Bank of Australia on New Zealand assets). The European Central

Bank generates heterogeneous spillovers between countries inside and outside of continental

Europe—but in a different manner. When the ECB tightens or eases, the euro appreciates or

depreciates most against non‐European countries; but it is those same countries whose yields

move least. This is in contrast to the Fed, where the dollar appreciates or depreciates most

against the countries whose yields move most.

SHAH

|

127