Gold and inflation: Expected inflation effect or carrying cost effect?

• Author: Chi‐Wei Su,Zhi‐Xin Liu,Yingying Xu,Jaime Ortiz
• Published date: 01 December 2019
• Date: 01 December 2019
• DOI: http://doi.org/10.1111/infi.12347

DOI: 10.1111/infi.12347

ORIGINAL ARTICLE

Gold and inflation: Expected inflation effect or

carrying cost effect?

Yingying Xu

1

|

Zhi‐Xin Liu

2

|

Chi‐Wei Su

3

|

Jaime Ortiz

4

1

Donlinks School of Economics and

Management, University of Science and

Technology Beijing, Beijing, People's

Republic of China

2

School of Economics and Management,

Beihang University, Beijing, People's

Republic of China

3

School of Economics, Qingdao

University, Qingdao, People's Republic of

China

4

Vice Provost Global Strategies and

Studies, University of Houston, Houston,

Texas

Correspondence

Jaime Ortiz, Vice Provost Global

Strategies and Studies, University of

Houston, E.W. Cullen Bldg. Suite 101.

4302 University Dr., Houston, TX 77204.

Email: jortiz22@uh.edu

Funding information

National Natural Science Foundation of

P.R. China, Grant/Award Number:

No.71873014

Abstract

This study examines whether the expected inflation

effect hypothesis adequately explains the causal rela-

tionship between inflation expectations and gold re-

turns. A bootstrap full‐sample Granger causality test

shows that gold returns cause inflation expectations

rather than the reverse. To account for possible

structural changes, we apply bootstrap subsample

Granger causality tests with 60‐month windows. The

results suggest that both professional forecasters' and

consumers' inflation expectations have negative effects

on gold returns in some but not all sample periods,

contradicting the expected inflation effect hypothesis.

No causality is found in other periods, consistent with

the carrying cost hypothesis that the expected gain from

gold due to higher inflation is offset by its carrying cost.

Holding gold will not necessarily hedge against inflation

because gold returns do not necessarily correlate with

inflation expectation. Therefore, the carrying cost

hypothesis more accurately explains the relationship

between gold returns and inflation expectation than the

expected inflation effect hypothesis does.

KEYWORDS

carrying cost hypothesis, expected inflation effect hypothesis,

gold returns, inflation expectation

JEL CLASSIFICATION

E31, E44, G44

International Finance. 2019;22:380–398.wileyonlinelibrary.com/journal/infi380

|

© 2019 John Wiley & Sons Ltd

1

|

INTRODUCTION

Financial crises provide a strong motivation for investors to seek out safe‐haven assets.

Conventional wisdom holds that the price of gold varies with the general price index, and

consequently, gold continues to hedge against the risk of inflation, even after the collapse

of the Bretton Woods system in 1973 (Aye, Chang, & Gupta, 2016; Long, Li, & Li, 2016).

Fisher (1930) stated the fundamental relationship between gold and inflation as follows:

the expected nominal asset return is a combination of expected return and expected

inflation rate. Thus, a rise in inflation expectation increases gold returns and subsequently

causes higher inflation. The empirical relationship between gold and inflation has been

widely analysed since the 1970s, but the results are mixed and inconclusive. Whereas

Wang, Lee, and Thi (2013) and Bampinas and Panagiotidis (2015) argue that gold is a

reliable hedge against inflation, Tully and Lucey (2007) and Ghazali, Lean, and Bahari

(2015) provide evidence to the contrary. Some scholars attribute the divergent results to

the use of different models and data in these studies (Hoang, Lahiani, & Heller, 2016).

Regardless, whether expectations of higher inflation consistently increase gold returns has

rarely been examined. This article provides convincing evidence for unpredictable

relationships between gold returns and inflation expectations.

Thelinkagebetweengoldreturnsandinflation expectation is closely tied to the supply

of and demand for gold. Generally, the supply of gold is relatively inelastic and stable,

owing to its limited supply and production capacity due to the logistics of establishing new

mines and the difficulty of the extraction process (Beckmann & Czudaj, 2013).

Furthermore, central banks always maintain passive stocks of gold regardless of

fluctuations in actual gold prices (Aizenman & Inoue, 2012). However, the demand for

gold is sensitive to business cycles and economic uncertainties (Baur & McDermott, 2010).

On the one hand, theoretically, an increase in inflation expectation would stimulate

demand for gold, either to hedge against the expected decline in the value of money or in

anticipation of rising gold prices. The resultant purchasing pressure would raise the price

of gold, and therefore, inflation expectation would serve as a leading indicator of gold

returns. On the other hand, Blose (2010) argues that the positive linkage between gold

returns and inflation expectation may be counteracted if higher inflation expectations

increase the carrying cost of gold by causing higher interest rates. The increased carrying

cost would offset any speculative profit from investing in gold over the inflationary period.

Thus, the carrying cost hypothesis predicts no change in gold returns when inflation

expectation changes. The interaction between the expected inflation effect and carrying

cost might nullify any correlation between inflation expectation and gold returns.

Among previous studies of the relationship between gold returns and inflation

expectation, Chua and Woodward (1982) showed that expected and unexpected inflation

rates were significant explanatory factors for gold returns from 1975 to 1980 for the United

States, though not for Canada, Germany, Japan, Switzerland, or the United Kingdom.

Using unexpected changes in the consumer price index (CPI) as the measure of changes in

inflation expectation, Adrangi, Chatrath, and Raffiee (2003) found that gold returns were

positively correlated with inflation expectation and concluded that investing in gold may

be a reliable inflation hedge over both the short and long runs. Consistent with the

expected inflation effect hypothesis, Le Long, De Ceuster, Annaert, and Amonhaemano

(2013) concluded that in Vietnam, nominal gold returns vary in one‐to‐one correspon-

dence with inflation expectation. Bampinas and Panagiotidis (2015) examined the

XU ET AL.

|

381