The informational content of unconventional monetary policy on precious metal markets

• DOI: http://doi.org/10.1002/for.2461
• Date: 01 January 2018
• Author: Stephanos Papadamou,Vasilios Sogiakas
• Published date: 01 January 2018

RESEARCH ARTICLE

The informational content of unconventional monetary policy on

precious metal markets

Stephanos Papadamou

1

| Vasilios Sogiakas

2

1

Department of Economics, University of

Thessaly, Volos, Greece

2

Adam Smith Business School, University of

Glasgow, UK

Correspondence

Vasilios Sogiakas, Adam Smith Business

School (Economics), University of Glasgow,

Gilbert Scott Building, Glasgow G12 8QQ,

UK.

Email: vasilios.sogiakas@glasgow.ac.uk

Abstract

This paper investigates the informational content of unconventional monetary poli-

cies and its effect on commodity markets, adopting a nonlinear approach for model-

ing volatility. The main question addressed is how the Bank of England, Bank of

Japan, and European Central Bank's (ECB's) announcements concerning monetary

easing affect two major commodities: gold and silver. Our empirical evidence based

on daily and high‐frequency data suggests that relevant information causes ambigu-

ous valuation adjustments as well as stabilization or destabilization effects. Specif-

ically, there is strong evidence that the Japanese Central Bank strengthens the

precious metal markets by increasing their returns and by causing stabilization

effects, in contrast to the ECB, which has opposite results, mainly due to the hetero-

geneous expectations of investors within these markets. These asymmetries across

central banks' effects on gold and silver risk–return profile imply that the ECB

unconventional monetary easing informational content opposes its stated mission,

adding uncertainty in precious metals markets.

KEYWORDS

conditional correlation, GARCH models, unconventional monetary policy

1|INTRODUCTION

Recently there has been a debate in the literature regarding

the relationship between conventional monetary policy con-

duct and commodity prices dynamics. Among others,

Christiano, Eichenbaum, and Evans (2005), Mallick and

Sousa (2012), Sousa (2010), and Jawadi, Sousa, and Traverso

(2015) provide empirical evidence suppor ting the view of a

strong negative impact of monetary contractions on the

aggregate commodity price index. This argument is also

underlined by Belke, Bordon, and Hendricks (2010a, 2014)

and Belke, Orth, and Setzer (2010b), according to which

global liquidity and low interest rate regimes are valuable

indicators of commodity price inflation. Lastrapes and Selgin

(2001) argue that the relationship between precious metal

prices and monetary policy was weakened after the mid‐

1990s; however, Hammoudeh, Nguyen, and Sousa (2015)

identify a positive impact of US monetary policy contraction

on the aggregated commodity prices over the prefinancial cri-

sis period, though their findings are commodity sector spe-

cific. The precious metals reaction is more elastic, leading

to prices fall on impact, but then turns to a positive one

between two and six quarters, highlighting their role as finan-

cial instruments. In this direction Batten, Ciner, and Lucey

(2010) argued that precious metals cannot be considered as

a single asset class, due to the fact that among the precious

metals only the volatility of gold is affected by monetary

variables.

Over the last two decades, forming inflation expectations

via monetary policy announcements has been a typical strat-

egy for a central banker in developed countries. Central bank

communication has been a useful tool in forming these

expectations. Frankel and Hardouvelis (1985) are among

the first who document a relationship between monetary

Received: 21 September 2016 Revised: 14 January 2017 Accepted: 26 January 2017

DOI: 10.1002/for.2461

16 Copyright © 2017 John Wiley & Sons, Ltd. Journal of Forecasting. 2018;37:16–36.wileyonlinelibrary.com/journal/for

policy news and commodity prices. According to Frankel's

(1986) theoretical model, news about US monetary policy,

such as (unexpected) interest rate changes, is expected to play

an important role in commodity prices. In this line, Hayo,

Kutan, and Neuenkirch (2012) provide evidence that

expected target rate changes and communications decrease

commodity volatility, whereas target rate surprises and unor-

thodox measures increase it. Nevertheless, the intensity of

such results is found to be medium during the recent crisis.

Monetary policy communication is found to dampen com-

modity spot prices according to Tai, Chao, Hu, Lai, and

Wang (2014). Moreover, Hove, Mama, and Tchana (2015)

argue that volatility in emerging economies' market commod-

ities is effectively confronted, if the central bank targets on

CPI inflation. By focusing on China's economy, Klotz, Lin,

and Hsu (2014) indicate a positive causality from exchange

rate volatility to industrial metals price uncertainty. Addi-

tional evidence is found in Papadamou and Markopoulos

(2014), who reveal a unidirectional volatility transmission

from the British pound and euro currencies to gold and silver

precious metals. Moreover, Khalifa, Miao, and Ramchander

(2011) investigated the relationship between the returns of

precious metals and concluded that the normality assumption

of precious metals distributions is not valid and that high‐fre-

quency data (intraday) could provide better forecasts com-

pared with lower‐frequency data (daily). Furthermore, Baur

and Tran (2014) concluded that there is a close relationship

between gold and silver returns in the long term, which is

affected significantly during distress periods. Finally,

Narayan, Narayan, and Sharma (2013) and Batten, Ciner,

Lucey, and Szilagyi (2013) underline the importance of using

commodities within profitable trading strategies, though the

performance of these strategies could be affected signifi-

cantly during distressed periods, as was the case within the

recent financial crisis.

The first research hypothesis of our paper is the examina-

tion of potential effects of central banks' announcements on

gold and silver returns using daily and intraday data. Our sec-

ond research hypothesis focuses on the examination of poten-

tial stabilization or destabilization effects of central banks'

announcements on gold and silver volatility series using,

again, daily and intraday data. Finally, we examine the

dynamics of the correlation between gold and silver returns

due to central banks' announcements using both daily and

intraday data.

The main contribution of our study to the existing litera-

ture lies in the evaluation of the effects of the unconventional

monetary policy announcements on the risk–return profile of

the precious metals and on their correlation dynamics by uti-

lizing daily and intraday frequency data. Although there are

an increasing number of papers recently focusing on the

effect of quantitative easing on bond and currency markets,

1

little has been done concerning the role of unconventional

monetary policy announcements on precious metals markets.

This paper tries to fill this gap, adding to existing literature on

the interaction between monetary policy and commodities by

focusing on two precious metals, gold and silver, which are

attractive assets in a typical asset allocation strategy.

Our empirical evidence suggests that the effect that the

informational content of unconventional monetary policy

has on the precious metals markets is country specific. The

Bank of Japan (BoJ) affects positively metals returns, damp-

ening their volatility, in contrast to the European Central

Bank (ECB), which causes destabilization effects mainly

due to the heterogeneous expectations of investors within

these markets. These findings imply that the ECB unconven-

tional monetary‐easing informational content opposes its

stated mission, adding uncertainty in precious metals mar-

kets. The rest of the paper consists of the data and the econo-

metric methodology, the presentation of the empirical

findings, and a concluding discussion of the paper.

2|DATA AND RESEARCH

METHODOLOGY

For the purposes of our analysis we use prices for the two

major commodities, that is, gold and silver, on a daily

2

and

on an intraday (hourly) frequency. The sample period under

investigation spans from January 2009 to July 2015, covering

a period

3

where unconventional monetary policy actions are

undertaken by the Bank of England (BoE), the Japanese Cen-

tral Bank, and the ECB after the global financial crisis. The

role of interest rates as an effective monetary policy tool

was diminished through the recent financial crisis, leading

central banks to act as lenders of last resort, providing liquid-

ity in the markets by directly purchasing assets (quantitative

easing) and shifting from short‐term to long‐term funding

to banks (credit easing).

The announcements that are examined in this study

relate to the following actions: (i) the enhancement of mon-

etary easing by the Japanese Central Bank; (ii) the

increases in size of the asset purchase program by the

BoE; and (iii) the long‐term refinancing operations

(LTROs) with allotted amount greater than 100 billion

euros by the ECB. Following previous research by

Kenourgios, et al. (2015a) we tested a series of announce-

ments by each central bank covering the m onetary easing

strategies. More specifically, these include but not limited

to the following: (a) announcement declaring the enhance-

ment of monetary easing by the BoJ; (b) announcements

stating the size of asset purchases by the BoE; and (c) the

long‐term refinancing operations (LTROs) announcements

by the ECB with allotted amount greater than 100 billion

euros.

4

These announcements are captured by time

dummies that consider several lead–lag structures. The first

PAPADAMOUAND SO GIAKAS 17