North American Journal of Economics and Finance 34 (2015) 187–214
Contents lists available at ScienceDirect
North American Journal of
Economics and Finance
Review
What do scientists know about inflation
hedging?
Stephan Arnold, Benjamin R. Auer ∗
University of Leipzig, Department of Finance, Germany
a r t i c l e
i n f o
Article history:
Received 8 June 2015
Received in revised form 28 August 2015
Accepted 31 August 2015
Available online 30 September 2015
JEL classification:
G10
G11
G15
Keywords:
Hedge
Inflation
Stocks
Gold
Fixed income
Real estate
a b s t r a c t
In this article, we give an overview of the state of scientific knowledge on inflation hedging. Specifically, we distill the results of
several decades of research analysing the relationship between
major asset classes (common stocks, gold, fixed income securities, real estate) and inflation. Even though previous studies have
brought forth important facts characterising the interplay of asset
returns and inflation rates (e.g., time-dependency, asymmetry,
outlier-sensitivity and a tendency towards long-term but limited
short-term inflation protection), there is still no consensus on the
subject because sample, data and methodology issues preclude
strict comparison of most studies. Thus, from a synthesis of the
insights gained from our review, we also outline possible directions
for future research that may help to establish consensus among
researchers.
© 2015 Elsevier Inc. All rights reserved.
Contents
1.
2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
Some basic concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
2.1.
Inflation measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
2.2.
Inflation hedge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
2.3.
Fisher hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
∗ Corresponding author at: University of Leipzig, Department of Finance, Grimmaische Straße 13-15, 04109 Leipzig, Germany.
Tel.: +49 341 97 33 672; fax: +49 341 97 33 679.
E-mail address: auer@wifa.uni-leipzig.de (B.R. Auer).
http://dx.doi.org/10.1016/j.najef.2015.08.005
1062-9408/© 2015 Elsevier Inc. All rights reserved.
188
3.
4.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
2.4.
Extended Fisher hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
2.5.
Further extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
Asset returns and inflation: a synthesis of empirical evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
Common stocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
3.1.
Gold . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198
3.2.
Fixed income . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202
3.3.
Real estate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
3.4.
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
1. Introduction
Nearly 40 years ago, Lintner (1975) argued that ‘few matters are of more serious concern to
students of finance and to members of the financial community than the impacts of inflation
on our financial institutions and markets and its implications for investment policy’. This view
that inflation is one of the predominant financial concerns for both academics and practitioners originated in the heyday of American inflation in the 1970s and is just as important today
as it was then. Inflation forecasts now indicate new increases in price levels in the near future,
driven by higher food and energy costs resulting from a reverse of recent influences (e.g. depressed
oil prices) that contributed to a phase of rather low US inflation (see Barclays Research Centre,
www.wealth.barclays.com). Inflation acceleration in emerging markets is likely to be larger than in
developed countries because of the much greater importance of food in their economies and their
stronger rates of growth (see Amenc, Martellini, & Ziemann, 2009). As a result of these trends and
the knowledge about historic impacts of inflation on the economy, inflation hedging has become a
concern of vital importance not only for private investors, who see inflation as a direct threat to their
purchasing power, but also for pension funds, whose pension payments are indexed to consumer
prices.
In previous decades, scientific studies have examined the interplay between asset returns and
inflation rates in order to identify assets that can protect investors from inflation. Typically, the media
attribute an inflation-hedging ability to common stocks, gold, fixed income securities and real estate.
However, does empirical evidence support this view? Do these assets hedge both expected and unexpected components of the rate of inflation? Can they provide inflation protection in the short run and
the long run? Is the hedging effectiveness stable over time and economic states, or does it depend on
the sample under analysis?
Because each study has its own answer to these and related questions, there is no consensus on
whether these assets can hedge against inflation. This is because studies differ in their data sources,
sample period and frequency, country coverage and/or econometric methodology. In order to systematise the evidence, this article offers a comprehensive overview of what academics know about
inflation hedging. That is, we summarise important characteristics of asset–inflation relationships scientific research has identified. This will allow us to answer the questions stated above and identify gaps
in the literature that may be filled by future research. We will not discuss the macroeconomic prerequisites or econometric methodology used in the studies reviewed here; we will, however, assess the
direct inflation hedging properties of single assets.1 We focus on common stocks, gold, fixed income
securities and real estate because these assets have received the most significant attention in the
literature.2
1
Detailed descriptions of all methods mentioned in our review can be found in Tsay (2005) and Greene (2008).
There are of course other potential inflation hedges such as private equity or infrastructure. However, so far there is no data
of sufficient quality, time-span and frequency that allows reasonable econometric analysis (see Rödel & Rothballer, 2012). This
is not true for assets like general commodities or hedge funds but, in comparison to our four asset classes, they have received
only little attention in the inflation hedging literature. Thus, the few existing studies (summarized by Rödel, 2012) do not allow
similarly deep overall conclusions.
2
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
189
The remainder of our article is organised as follows. Section 2 reviews the concepts that are
required to study whether an asset can hedge against inflation. We define the terms inflation and
inflation hedging and discuss variants of the Fisher hypothesis linking asset returns and inflation. This
is because most research on inflation hedging is based on the latter hypothesis. Section 3 discusses the
most influential peer-reviewed articles and working papers. For each of the traditional asset classes
common stocks, gold, fixed income securities and real estate, we screen the literature for answers to
the question of whether they are effective inflation hedges. For comparative purposes, we supplement
our discussion with tables documenting the differences in the time-frames, datasets and methodologies used in the studies and indicating their results (hedge, no hedge, mixed results). Finally, Section
4 concludes and reveals gaps in the literature that might be interesting for future research.
2. Some basic concepts
2.1. Inflation measurement
Because inflation is defined as ‘a process of continuously rising prices, or equivalently, of a continuously falling value of money’ (see Parkin, 2008), a natural way to measure inflation is to use growth
rates of representative consumer price indices (CPIs). Even though there are significant conceptual
shortcomings with this approach (e.g., lagged announcements, international differences in measurement and the fact that CPI estimates are unlikely to constitute an adequate representation of the
relevant price changes for a particular investor), CPIs have become the most frequently used proxies
for inflation (see Tables 1–3). A smaller number of studies uses production price indices (PPI), wholesale price indices (WPI), retail price indices (RPI) or GDP deflators. Depending on the data frequency
used in empirical studies, either monthly, quarterly, bi-annual or annual period-to-period percentage
changes are used as inflation rates.
2.2. Inflation hedge
There are three possible definitions of inflation hedging (see Bodie, 1976). In the first one, a security
is called an inflation hedge if it eliminates or at least reduces the possibility that the real return on
the security will fall below some specified floor value such as zero. A second definition measures the
hedging effectiveness of a security as the proportional reduction in the variance of the real returns
on a default-free bond attainable by combining the security and the bond. Finally, the last definition,
which has been used in almost all empirical studies of inflation hedging, states that an asset is an
inflation hedge if its real return is independent of the rate of inflation, implying a positive correlation
between the nominal return of the hedging asset and inflation. A correlation of 1 is called a perfect
hedge because price increases are perfectly compensated by corresponding asset returns. If an asset
does not provide a perfect hedge, a stable positive return-inflation relation can still make the asset
valuable because, in this case, adequate hedge ratios theoretically allow its use as an effective hedge.
However, because hedge ratio-based hedging can induce high transaction costs for retail investors
(see Bekaert & Wang, 2010), high magnitudes of co-movement are of higher practical use than small
ones being levered.
2.3. Fisher hypothesis
Fisher (1930) was the first who formally stated the hypothetical relationship between asset returns
and inflation. He noted that the nominal interest rate can be expressed as the sum of an expected real
return and an expected inflation rate. Fama and Schwert (1977) point out that the proposition that
expected nominal returns contain market assessments of expected inflation rates can be applied to all
assets. Thus, if the information available at time t − 1 is processed efficiently, the market will set the
price of any asset j so that the expected nominal return Et−1 (rjtn ) on the asset from t − 1 to t is the sum
of the appropriate equilibrium expected real return Et−1 (rjtr ) and the best possible assessment of the
expected inflation rate Et−1 (t ) from t − 1 to t. Formally, this means that
Et−1 (rjtn ) = Et−1 (rjtr ) + Et−1 (t )
(1)
190
Table 1
Stocks as a hedge against inflation.
Period (frequency)
Region (data)
Inflation proxy
Econometric methodology
Jaffe and Mandelker
(1976)
Bodie (1976)
1953–1971 (monthly)
CPI, WPI
OLS regression
CPI
OLS regression
Nelson (1976)
1953–1974 (monthly)
CPI
OLS regression
Fama and Schwert
(1977)
Gultekin (1983a)
1953–1971 (semi-annual, quarterly, monthly)
USA (Lawrence Fisher
index)
USA (portfolios of NYSE
common stocks)
USA (Scholes index, S&P
500 index)
USA (portfolios of NYSE
common stocks)
26 countries (IMF stock
market indices)
CPI
OLS regression
CPI
OLS regression
Gultekin (1983b)
1952–1979 (semi-annual, annual)
USA (Dow Jones Industrial
Average index, S&P 500
index)
CPI, WPI
OLS regression
Boudoukh and
Richardson (1993)
1802–1990 (annual)
USA, UK (individual
portfolios)
CPI
Cochran and Defina
(1993)
1947–1989 (quarterly)
USA (S&P 500 index)
Consumption
deflator
OLS and
instrumental-variable
regressions, GMM
Cointegration techniques
(ADF), ECM
Solnik and Solnik
(1997)
1958–1996 (monthly)
USA, Germany, France,
Netherlands, UK,
Switzerland, Japan, Canada
(MSCI indices)
CPI
OLS and
instrumental-variable
regressions, GMM, joint
testing of all countries
Ely and Robinson
(1997)
1957–1992 (quarterly)
16 industrialised countries
(IMF stock indices)
CPI, PPI, GDP
deflator
Cointegration techniques
(ADF, KPSS, Johansen),
VECM
Lothian and Simaan
(1998)
1973–1994 (annual)
23 OECD countries (IMF
stock indices)
CPI, CLI
OLS regression (with long
period arithmetic averages
of log differences)
Schotman and
Schweitzer (2000)
–
–
–
Nine scenarios with
parameter estimates based
on previous studies.
Lothian and McCarthy
(2001)
1790–2000 (annual)
14 OECD countries (IMF
stock indices)
CPI, CLI
Cointegration techniques
(ADF, PP)
Kolari and Anari (2001)
1953–1998 (monthly)
USA, Canada, UK, France,
Germany, Japan (major
stock indices)
CPI
Cointegration techniques
(ADF, Johansen), VAR
models
1953–1972 (annual, quarterly, monthly)
1947–1979 (monthly)
Hedge?
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Study
1922–1997 (annual)
USA, Denmark (S&P
Composite index,
Copenhagen stock
exchange portfolio)
CPI, GDP deflator
VAR model for one-period
returns and inflation
Spyrou (2004)
1989–2000 (monthly)
10 major emerging
markets (Datastream total
stock market indices)
CPI
OLS regression
Ahmed and Cardinale
(2005)
1919–2002 (annual, monthly)
USA, UK, Japan, Germany
(GFD stock indices)
CPI
OLS regression,
cointegration techniques
(Johansen), VAR, Granger
causality
Luintel and Paudyal
(2006)
1955–2002 (monthly)
UK (Financial Times all
share index and industry
sub-groups)
RPI
Cointegration techniques
(Johansen, ADF, KPSS)
Alagidede and
Panagiotidis (2010)
1980–2007 (monthly)
Egypt, Kenya, Morocco,
Nigeria, South Africa,
Tunisia (IMF stock indices)
CPI
OLS regression,
cointegration techniques
(PP, Breitung, KPSS), VECM
Kim and Ryoo (2011)
1900–2009 (monthly)
USA (Dow Jones Industrial
Average index, S&P 500
index)
CPI
Cointegration techniques
(ADF), threshold VECM
Rödel (2012, 2014)
1950–2010 (annual)
45 countries (GFD stock
indices)
CPI
Panel regression with
country fixed effects,
cointegration techniques
(KPSS)
Abbreviations: ADF: augmented Dickey–Fuller test; CLI: cost-of-living index; CPI: consumer price index; ECM: error correction model; GDP: gross domestic product; GFD: global financial
data; GMM: generalised method of moments; IMF: International Monetary Fund; KPSS: Kwiatkowski–Phillips–Schmidt–Shin test; MCSI: Morgan Stanley Capital International; NYSE:
New York stock exchange; OLS: ordinary least squares; PP: Phillips–Perron test; RPI: retail price index; VAR: vector autoregression; VECM: vector error correction model; WPI: wholesale
and indicate evidence for and against an inflation hedging property, respectively, whereas
represents mixed or inconclusive results.
price index. The symbols
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Engsted and Tanggaard
(2002)
191
192
Table 2
Gold as a hedge against inflation.
Period (frequency)
Region (data)
Inflation proxy
Econometric methodology
Chua and Woodward (1982)
1975–1980 (monthly, semi-annual)
CPI
OLS regression
Brown and Howe (1987)
1975–1983 (monthly)
Canada, Germany, Japan,
Switzerland, UK, USA
(London PM fixing)
USA (Wallstreet Journal
gold prices, COMEX gold
futures)
CPI
OLS regression
Jaffe (1989)
1971–1987 (monthly, quarterly)
CPI
OLS regression
Mahdavi and Zhou (1997)
1970–1994 (monthly, quarterly)
CPI
Cointegration techniques (ADF, PP, KPSS), VECM
Taylor (1998)
1914–1996 (monthly)
USA (Handy/Harmann
and Knight-Ridder
Financial)
CPI
OLS and LAD regression, Huber M-estimation,
cointegration techniques (DF, ADF)
Ghosh et al. (2004)
1976–1999 (monthly)
USA, Worldwide
(unspecified)
CPI, RPI
Cointegration techniques (ADF, Johansen)
Levin et al. (2006)
1976–2005 (monthly)
USA, Worldwide
(unspecified)
CPI
Cointegration techniques (DF, Johansen), VECM
McCown and Zimmerman
(2006)
1970–2003 (monthly)
USA (Wallstreet Journal
gold prices)
CPI
CAPM-based multifactor model, cointegration
techniques (DF, KPSS, Johansen)
Worthington and Pahlavani
(2007)
1945–2006 (monthly)
USA (GFD gold prices)
CPI
Cointegration techniques with endogenous
structural breaks
McCown and Zimmerman
(2007)
Blose (2010)
1970–2006 (monthly)
USA (Wallstreet Journal
gold prices)
USA (London PM fixing)
CPI
CAPM-based multifactor model, correlation
analysis
Non-linear regression
Rubbaniy et al. (2011)
1985–2010 (monthly)
Baur (2011)
1979–2010 (monthly)
Wang et al. (2011)
1971–2010 (monthly)
USA, Japan (London PM
fixing)
CPI
Linear and non-linear cointegration techniques,
threshold VECM, causality tests
Beckmann and Czudaj (2013)
1970–2011 (monthly)
CPI, PPI
Markov-switching VECM
Erb and Harvey (2013)
1975–2011 (monthly)
USA, UK, Euro Area,
Japan (London PM fixing)
USA (London PM fixing)
CPI
Descriptive analysis, OLS regression
Batten et al. (2014)
1985–2012 (monthly)
USA (London PM fixing)
CPI
Kalman filter regressions, cointegration
techniques (ADF, KPSS, Johansen,
Saikkonnen–Lütkepohl), ECM, Granger causality
1988–2008 (monthly)
USA (London PM fixing,
gold stocks)
USA (London PM fixing)
Germany (London PM
fixing)
USA (London AM fixing)
CPI
CPI
CPI
Hedge?
Asymmetric CAPM-based multifactor model,
cointegration techniques (ADF, KPSS, Johansen)
Asymmetric GARCH model
Abbreviations: AM: ante meridiem; APT: arbitrage pricing theory; ADF: augmented Dickey–Fuller test; CAPM: capital asset pricing model; COMEX: (New York) Commodities Exchange;
CPI: consumer price index; DF: Dickey–Fuller test; ECM: error correction model; GARCH: generalised autoregressive conditional heteroscedasticity; GFD: global financial data; KPSS:
Kwiatkowski–Phillips–Schmidt–Shin test; LAD: least absolute deviation; OLS: ordinary least squares; PM: post meridiem; PP: Phillips–Perron test; RPI: retail price index; VECM: vector
and indicate evidence for and against an inflation hedging property; respectively, whereas
represents mixed or inconclusive results.
error correction model. The symbols
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Study
Table 3
Real estate as a hedge against inflation.
Period (frequency)
Region (data)
Inflation proxy
Econometric methodology
Tarbert (1996)
1978–1995 (semi-annual, quarterly)
UK (commercial property)
CPI
Cointegration techniques (CDW, ADF),
Granger causality
Barkham et al. (1996)
1982–1994 (monthly)
UK (Richard Ellis all property
index)
RPI
Cointegration techniques (ADF, PP, KPSS,
Johansen), VECM, Granger causality
Matysiak et al. (1996)
Chatrath and Liang (1998)
1964–1993 (annual, quarterly)
1972–1995 (monthly, quarterly)
UK (commercial property)
USA (equity, mortgage and
hybrid REITs; all NAREIT)
RPI
CPI, treasury bill
rates
Cointegration techniques (ADF, Johansen)
OLS regression, cointegration techniques
(DF, PP, Johansen)
Ganesan and Chiang (1998)
1984–1994 (quarterly)
Hong Kong (office,
commercial, industrial,
residential real estate,
property stocks)
CPI
OLS regression, cointegration techniques
(DF)
Anari and Kolari (2002)
1968–2000 (monthly)
USA (new and existing home
prices)
CPI minus
housing costs
Recursive OLS regression, Cointegration
techniques (ARDL)
Glascock et al. (2002)
1970–1995 (monthly)
CPI
Adrangi et al. (2004)
1972–1999 (monthly)
USA (NAREIT total return
index)
USA (equity and mortgage
REITs; both NAREIT)
CPI
Cointegration techniques (PP), VECM,
Granger causality
OLS regression with Chow tests,
cointegration techniques (DF, ADF, PP,
Johansen)
Simpson et al. (2007)
1981–2002 (monthly)
USA (publicly traded equity
REITs from CRSP)
CPI
Fixed-effect pooled regression
Le Moigne and Viveiros (2008)
1973–2007 (annual, quarterly)
Canada (apartment,
industrial, office, retail, and
mixed use real estate)
CPI
OLS regression, cointegration techniques
(ADF, PP)
Hoesli et al. (2008)
1977–2003 (quarterly)
USA, UK (private and public
commercial real estate)
CPI, RPI
Cointegration techniques (ADF, PP), ECM
Amenc et al. (2009)
1973–2007 (quarterly)
USA (NAREIT total return
index)
CPI
Cointegration techniques (ADF, Johansen),
VECM, VAR, Monte Carlo simulation
Hardin et al. (2012)
1980–2008 (monthly)
USA (REIT stock prices,
CRSP/Ziman databases)
CPI
OLS regression, VAR model
Lee and Lee (2012)
1972–2007 (monthly)
USA (NAREIT total return
index, CRSP/Ziman databases)
CPI
OLS regression, cointegration techniques
(DOLS, PP, Johansen)
Park and Bang (2012)
2002–2010 (quarterly)
Korea (commercial real
estate)
CPI
OLS regression, cointegration techniques
(DOLS, DF, ADF, Johansen), VECM, Granger
causality
Obereiner and Kurzrock (2012)
1992–2009 (monthly)
Germany (real estate funds,
special funds, real estate
stocks)
CPI
Cointegration techniques (ADF, PP,
Johansen)
Hedge?
193
Abbreviations: ADF: augmented Dickey–Fuller test; CDW: cointegrating Durbin–Watson statistic; CRSP: Center for Research in Security Prices; CPI: consumer price index; DF: Dickey-Fuller
test; DOLS: dynamic ordinary least squares; ECM: error correction model; KPSS: Kwiatkowski–Phillips–Schmidt–Shin test; NAREIT: National Association of Real Estate Investment Trusts;
OLS: ordinary least squares; PP: Phillips–Perron test; REIT: real estate investment trust; RPI: retail price index; VAR: vector autoregression; VECM: vector error correction model. The
and indicate evidence for and against an inflation hedging property, respectively, whereas
represents mixed or inconclusive results.
symbols
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Study
194
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Tests of the hypothesis that the expected real return and expected inflation vary independently
(and jointly that the market is efficient) require some way to quantify Et−1 (t ). For example, we
could derive an empirical measure of the expected inflation rate from (i) well-known theoretical
relations in the treasury bill market (see Fama, 1975; Fama & Schwert, 1977), (ii) survey measures (e.g.,
Consensus inflation forecasts), (iii) time-series models, and (iv) information contained in marketable
securities (e.g., inflation swaps and inflation-linked bonds) or (v) simply use realised inflation as a
rational forecast for future inflation (see Lintner, 1975; Rödel, 2012). Given a suitable measure, we can
estimate the regression model
rjtn = ˛j + ˇj E t−1 (t ) + εjt
(2)
Because a standard regression estimates the conditional expected value of the dependent variable
as a function of the independent variable, a ˇj estimate that is not significantly different from one
is consistent with the hypothesis that the expected nominal return of asset j varies in one-to-one
correspondence with the expected inflation rate and thus acts as a perfect hedge. Of course, because
of (1) this is also consistent with the hypothesis that the expected real return on the asset and the
expected inflation are unrelated (see Fama & Schwert, 1977).
2.4. Extended Fisher hypothesis
While Fisher (1930) incorporates only expected inflation Et−1 (t ), Fama and Schwert (1977) extend
(2) by also considering unexpected inflation t − Et−1 (t ), which leads to the regression equation
rjtn = ˛j + ˇj Et−1 (t ) + j [t − Et−1 (t )] + jt
(3)
In this framework, an asset j is a perfect hedge against expected inflation (unexpected inflation) if
ˇj ( j ) is not significantly different from one.3 A partial hedge is given for significant parameter values
between 0 and 1. Negative values of ˇj and j suggest that the asset acts as a ‘perverse hedge’ against
inflation. This means that an inflation hedge can be obtained only by shorting the asset.
2.5. Further extensions
Other scholars propose using a version of the Fisher model that takes taxes into account. Investors
are liable for paying income and capital gain taxes; this is why the nominal return rate should include
the effects of both taxes and inflation. Therefore, the elasticity factors in this tax-augmented version
of the Fisher model should be significantly above unity (see Luintel & Paudyal, 2006). Of course other
models can also capture the link between returns and inflation (see, for example, Bekaert & Engstrom,
2010; McCown & Zimmerman, 2007; Rubbaniy, Lee, & Verschoor, 2011). However, most studies of
inflation hedging rely on Fisher-type relations. They all claim that any type of asset should act as a
hedge against inflation because expected nominal returns should efficiently adjust to new information
about the level of goods prices.
3. Asset returns and inflation: a synthesis of empirical evidence
3.1. Common stocks
From a theoretical perspective, two important reasons make stocks good hedges against inflation
(see Lintner, 1975). First, equities represent claims against real assets whose values are expected
to keep pace with changes in purchasing power. Second, firms leverage their capital and are
net debtors on average. Thus, shareholders would benefit from unexpected inflation as the firm’s
3
This does not imply zero or small variance of real returns since inflation might explain only a small fraction of the variation
in nominal returns. The variance of the error term, which in this case is the variance of the real return, might be large relative
to the variance of inflation (see Fama & Schwert, 1977).
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
195
long-term commitments to pay fixed nominal amounts decrease in real value (the debtor-creditor
hypothesis).4
In recent years, the empirical analysis of stocks’ ability to protect investors from inflation-related
losses in purchasing power has received considerable attention in the academic literature. The most
important studies in the field are summarised in Table 1, which is limited to articles published
after 1970. Earlier studies, summarised by Roll (1972) and Nelson (1976), which resort either to
anecdotal evidence or use simple regression analysis, agreed on the proposition that the return
of common stocks co-moves with inflation in a one-to-one relation as predicted by the Fisher
model.
Bodie (1976) wrote a seminal study of inflation hedging. Using New York Stock Exchange (NYSE)
equity returns from the Center for Research in Security Prices (CRSP) with holding periods of one
month, three months and one year and the CPI from the US Bureau of Labour Statistics for 1953–1972,
he draws the surprising conclusion that equities appear to be a perverse inflation hedge since in the
short run their returns are negatively related to both expected and unexpected inflation. He concludes
that stocks must be sold short in order to be used as inflation hedges. This puzzling result is by no
means an isolated finding. Nelson (1976), Jaffe and Mandelker (1976), Fama and Schwert (1977) and
Gultekin (1983b) all confirm the negative relation between stock returns and the inflation rate in the
short run.
Jaffe and Mandelker (1976) analyse an extended sample covering almost one century of data,
from 1876 to 1971. Using consumer and wholesale prices for measuring inflation and the Lawrence
Fisher index (an equally weighted portfolio of all securities listed on the NYSE) for capturing
US stock returns, they show that stock returns appear to be independent of inflation over this
long period. However, they point out that this result may be biased by flawed data because
the CPI published by the Bureau of Labour Statistics has been of reasonable quality only after
a substantial upgrading of its sampling procedures in 1953. In a separate analysis of the period
from 1953 to 1971, Jaffe and Mandelker (1976) confirm that stock returns are significantly negatively related to inflation. In a study of 26 countries, Gultekin (1983a) also finds that there is a
consistent lack of positive relation between stock returns and inflation in most countries. Interestingly, these results hold irrespective of the three proxies he used for expected inflation: (i)
contemporaneous inflation, (ii) short-term interest rates, and (iii) expected and unexpected components of inflation decomposed by ARIMA models. Finally, the multi-asset study of Fama and
Schwert (1977) analysing the extended Fisher model for US common stocks, US government bonds
and bills, private residential real estate and labour income, concludes that there is no explanation for the negative relationship between stock returns and the inflation rate from 1953 to
1971.
A potential rationale for the anomalous behaviour of stock returns was provided four years
later. Fama (1981) argues that there is no direct interaction between inflation and stock returns
because the former simply acts as a proxy for real economic trends (the proxy hypothesis). On
the one hand, stocks benefit from higher expected economic activity. On the other hand, increasing inflation leads to lower economic activity due to the short-term non-neutrality of money.
Hence, changes in inflation signal changes in economic activity. The negative relation between stock
returns and inflation might therefore be spurious.5 Subsequent studies offered several alternative
explanations for the difference between theory (Fisher model) and practice (stocks perform poorly
during inflationary periods). Their main ideas are, for example, that the spurious relation between
the two variables may be linked to debt monetisation (see Geske & Roll, 1983), equilibrium processes in the monetary sector (see Kaul, 1987) or the role of a firm’s financing costs (see Lintner,
1975).6
4
However, unexpected inflation may also trigger shocks to aggregate output that can harm the hedging effectiveness of
stocks.
5
This proxy hypothesis has been both supported and rejected in several follow-up studies (see Cochran & Defina, 1993).
6
For additional explanations (including the so-called money illusion hypothesis) see the study of Wilcox (2012) which is
based on the findings of Modigliani and Cohn (1979).
196
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Another explanation of the puzzling relationship between stock returns and inflation is linked to
the properties of the analysed time series and the econometric approaches used for testing hedging
effectiveness. First, the Fisher effect describes a long-run phenomenon but most of the earlier studies
cover only small samples. Thus, more recent studies suggest that using longer time series (see Solnik &
Solnik, 1997) or cumulative stock returns (see Schotman & Schweitzer, 2000) can increase confidence
in the findings. Second, measurement error in inflation introduces classic errors in variables problem
into the standard OLS estimations of Fisher-type models which need to be addressed by instrumental
variable regression (see Boudoukh & Richardson, 1993). Third, stock returns and inflation rates exhibit
special time-series properties. Inflation is a slow-moving and persistent process, with much lower
variance than stock returns. This negatively influences correlation tests between the two variables
(see Madsen, 2007) because non-stationary variables introduce the problem of spurious regression,
which is the detection of significant relationships even though none exist (see Granger, Hyung, & Jeon,
2001). To circumvent this problem, recent studies search for cointegration between stock returns
and inflation.7 If the two variables are non-stationary but cointegrated, (2) may be interpreted as a
cointegrating regression, reflecting an equilibrium relationship between nominal interest rates and
inflation. Hence, a finding of cointegration combined with a slope parameter not significantly different
from one is supportive of a long-run Fisher relationship.8,9 Because cointegrating regressions only
consider long-term dynamics, they are usually supplemented by error correction models (ECM). They
have been proposed by Granger (1983) and Engle and Granger (1987) and are designed to capture
both short-term and long-term effects of one time series on another (see Cochran & Defina, 1993;
Tarbert, 1996). In other words, ECM model short-run transitory changes as deviations from the longrun relationship.
Studies published since the 1990s have explored these methodological issues and produced a more
diverse picture than those of the 1970s and 1980s. Boudoukh and Richardson (1993) used longer
time series. They collected almost 200 years of data – from 1802 to 1990 – for the US and the UK.
Instrumental variable regression with past inflation rates and short- and long-term interest rates as
instruments as well as OLS regressions show that long-horizon nominal stock returns are positively
related to both ex ante and ex post long-term inflation. The ex ante Fisher coefficient at the 5-year
horizon ranges from 0.38 to 2.07 in case of the instrumental variable regression depending on the
instrument. OLS regression shows an ex post Fisher coefficient of 0.52 over the sample period and
similar results for the two sub-periods from 1870 to 1990 and 1914 to 1990. This intriguing finding
of a positive relationship between stock returns and inflation in the long run has been confirmed by
Solnik and Solnik (1997) and Lothian and Simaan (1998) using panel data of eight and 23 advanced
economies, respectively. Similar to Boudoukh and Richardson (1993), Solnik and Solnik (1997) test the
Fisher model by means of an instrumental variable approach (with the 3-month bill rate, long-term
bond rate, past 3-month and 1-year inflation rates as instruments). The comparably short period of 38
years is econometrically compensated by a cross-section of eight major countries. They find that the
Fisher model holds at a 1-year holding period horizon. Even for horizons of less than 12 months the
Fisher model is not rejected. The magnitude of the coefficient lends stronger support at longer horizons,
though. Lothian and Simaan (1998) find that average stock returns co-move with inflation in the long
run in almost all 23 OECD countries examined. The two variables are also positively correlated across
countries.
Lothian and McCarthy (2001) also extend the time-series dimension and cover data for the US and
the UK from 1790 to 2000 and for 14 OECD countries (including the US and the UK) from 1945 until
7
Simply using first differences as regression variables could solve the problem of spurious regression but eliminates long-run
information that is crucial in measuring the Fisher effect (see Hendry, 1986).
8
However, traditional cointegration tests exhibit two major problems: First, they have low power to reject the null hypothesis
of non-stationarity for variables that are persistent but follow a stationary process (see DeJong, Nankervis, Savin, & Whiteman,
1992). Second, size distortions and moving-average components in the underlying data generating process might impair their
results (see Lütkepohl & Saikkonen, 1999; Perron & Ng, 1996; Schwert, 1987, 1989).
9
Note that, unfortunately, there is no consensus in the literature on whether inflation rates are non-stationary or stationary
(see Charemza et al., 2005; Culver & Papell, 1997). Therefore, studies on inflation hedging can differ in terms of which variable
(CPI growth rates or CPI levels) they use to test for a long-run relationship via cointegration methodology.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
hier
197
1999. Using cointegration tests, OECD panel data and the long series for the US and the UK provide
robust evidence that equity prices, in fact, keep pace with movements in the overall price level. For
shorter horizons, they find a negative relationship between the two quantities. In the same year, Kolari
and Anari (2001) tested the long-run Fisher effect for six major economies (the US, Canada, the UK,
France, Germany and Japan) for a time span comparable to Lothian and McCarthy (2001). They show
that the long-run elasticity of stock prices with respect to inflation ranges from 1.04 to 1.65, which
tends to support the Fisher effect. Furthermore, in an impulse-response analysis, the time path of
the response of stock prices to a shock in goods prices exhibits an initial negative response, which
turns positive over long horizons. This helps to reconcile previous short-run and long-run evidence
and reveals that stock prices have a long memory with respect to inflation shocks, so investors should
expect stocks to be a good inflation hedge over a long holding period.10
Schotman and Schweitzer (2000) explicitly cover the horizon sensitivity of the hedging property
of stocks. They develop nine theoretical hedging scenarios with different parameter values for inflation persistence, the Fisher effect, and the inflation beta of stocks. Assuming that these scenarios are
equally likely, they show that stocks provide a hedge against inflation if the investor’s investment
horizon is 15 years or longer. Institutional investors such as insurance companies or pension funds
with long-term liabilities often buy high duration investments to reduce the asset-liability mismatch
in portfolios. Thus, they are less concerned about the short-run relation between stock returns and
inflation. Schotman and Schweitzer (2000) note that the persistence of inflation is a crucial parameter
for determining an effective hedge ratio. They also highlight that the role of horizon effects for hedgers
is an important shortcoming of classic Fisher regression estimates. This is because within the Fisher
framework hedging coefficients are estimated for only one specific investment period.
Among the first authors to apply ECM in an inflation hedging context are Cochran and Defina (1993).
For US data from 1947 to 1989 they show that real stock returns are not independent of inflation. Hence,
US stocks do not provide a long-term hedge. On the contrary, expected and unexpected components
of the inflation rate have a negative effect on real stock returns. Ely and Robinson (1997) rely on a more
general error correction approach. They employ a vector error correction model (VECM) – a special
case of the vector autoregressive model (VAR) for variables that are stationary in their differences
– because it allows the simultaneous consideration of multiple variables (the inclusion of variables
that might play a role in the interaction of returns and inflation). While the error correction terms
in the VECM account for short-term deviations from long-term equilibrium relationships among the
variables, impulse response functions can be derived from the VECM to estimate the response of
returns and inflation to changes in other included variables (such as measures of monetary policy or
real output). With quarterly data from 1957 to 1992 for 16 industrialised countries, Ely and Robinson
(1997) find that stocks maintain their real value relative to goods prices and this conclusion generally
does not depend on the source of the inflation shock (real or monetary sector). The only exception
are the US where stocks do not maintain their value relative to goods prices following real output
shocks.11
In order to assess the hedging properties of US and Danish stocks, Engsted and Tanggaard (2002)
measure multi-period expected returns and inflation from a VAR model involving only one-period
variables. They find that Danish stock returns co-move with expected inflation in the long run but not
at short horizons. For US stocks, however, the relationship between expected returns and inflation is
quite weak at all horizons. Thus, in contrast to Boudoukh and Richardson (1993), the Fisher model
does not perform better as the horizon increases in the US case. However, it does so for Danish stocks.
Ahmed and Cardinale (2005) analyse data of different inflation regimes from 1919 to 2002 comprising the US, the UK, Germany and Japan. Although evidence for a cointegration relation is mixed –
depending on horizons, methods and countries – equities tend to be a partial inflation hedge, even at
long horizons. Equities appear to react asymmetrically to high or low inflation. In the short run (one
10
Knif, Kolari, and Pynnönen (2008) note that summing positive an negative inflation shocks tends to wash out or mute the
effects on inflation news on stock returns. They show that, depending on the economic state, positive and negative inflation
shocks can produce a variety of stock market reactions.
11
However, note that the analysis of Ely and Robinson (1997) is not fully equivalent to testing if stocks provide a hedge against
inflation because it is based on stock price indices not including dividends.
198
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
year), equity returns have been significantly higher at times of ‘normal’ inflation (up to three percent)
but lower in case of deflation or very high inflation (with the exception of Japan). Hence, equities do
not offer protection when it is most needed. The same point has been made by Kim and Ryoo (2011)
who conclude that stock returns and inflation show asymmetric adjustments to the long-run equilibrium depending on the period (high vs. low inflation regime). By means of a threshold VECM with two
regimes they allow for asymmetric adjustments of stock returns and expected inflation towards the
long-run equilibrium. Within a sample period of more than 200 years, US common stocks have been
a hedge against inflation since the early 1980s. This coincides with the decline in the volatility of US
inflation and other macroeconomic fundamentals, the ‘Great Moderation’. With more stable inflation
rates it is more likely that inflation expectations become more accurate, potentially leading to a better
hedging performance.
Luintel and Paudyal (2006) disaggregate approximately 50 years of UK data to challenge the Fisher
hypothesis for the country as a whole as well as for seven industry groups. The levels of UK stocks and
the RPI are cointegrated both in aggregate and disaggregate data. Accounting for taxes, goods price
elasticities are significantly above unity with the exception of the mineral extraction industry. Hence,
British stocks not only provide an inflation hedge in the long run but also compensate for the loss
in the real wealth of tax-paying investors. Finally, in a study of 45 developed and emerging markets,
Rödel (2012, 2014) makes the interesting finding that international equities are superior to domestic
equities as inflation hedges.
To summarise the current state of scientific knowledge, we can state that stocks appear to perform
poorly during inflationary periods in the short run since they are negatively related to changes in
the rate of inflation. This is confirmed by early studies based on OLS techniques as well as by more
recent ones adopting more sophisticated approaches. However, since the 1990s, several articles have
presented more favourable evidence on the relation between equity returns and inflation. Even though
there is still no consensus on the hedging properties of stocks, most of the recent studies find a longrun stock-inflation equilibrium implying that stocks do provide a hedge at horizons of at least five
years.
3.2. Gold
Especially in the aftermath of the 2008 financial crisis, gold has attracted significant attention in the
scientific community (see Lucey, 2011). This is mainly because gold has a variety of features that make it
a desirable asset (see Worthington & Pahlavani, 2007). First of all, gold’s intrinsic value neither depends
on prospective cash flows nor carries a default risk. Second, gold is universally acceptable and scarce.
Third, the relative inelasticity of the gold supply and the observed counter-cyclical demand qualify
gold as a store of value. Finally, and probably most importantly, a protection property is commonly
attributed to gold by many investors, individuals and the media.
There are four streams in the literature on the economic and financial aspects of gold: forecasting
the gold price, event studies and tests for the market efficiency hypothesis, determinants of the price
for gold and the capability of gold to act as a diversifier, hedge or safe haven.12 The latter group is of
particular interest for our review because it covers the questions of whether the price of gold moves
along with higher inflation or whether it is unaffected by the level of goods prices because of factors
such as carrying costs (see Blose, 2010).
Research on common stocks subsumes inflation protection under the term ‘hedge’. In the analysis
of interrelations between gold and other markets, Baur and Lucey (2010) have introduced more useful
12
For reviews of these literature strands, see Worthington and Pahlavani (2007) and O’Connor, Lucey, Batten, and Baur (2015).
More recent articles on the issue of modelling and forecasting gold prices are Shafiee and Topal (2010), Białkowski, Bohl, Stephan,
and Wisniewski (in press) and Baur, Beckmann, and Czudaj (2014). Charles, Darné, and Kim (in press) analyse gold market
efficiency, Tully and Lucey (2007) the economic determinants of gold prices. Baur and McDermott (2010), Coudert and RaymondFeingold (2011), Ciner, Gurdgiev, and Lucey (2013), Pullen, Benson, and Faff (2014), Beckmann, Berger, and Czudaj (2015) and
Lucey and Li (2015) investigate the role of gold as a hedge against risks other than inflation (such as market and currency risks,
interest rate changes or the oil price shocks). Spierdijk and Umar (2014) and Zhou (2014) focus on aggregate commodity futures
indices (including energy, industrial metals and precious metals) to evaluate their inflation hedging potential.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
199
operational definitions of hedging properties. They state that ‘a hedge is defined as an asset that is
uncorrelated or negatively correlated with another asset or portfolio on average’ whereas a diversifier
is ‘positively (but not perfectly) correlated with another asset or portfolio on average’ and a safe haven
is ‘uncorrelated or negatively correlated with another asset or portfolio in times of market stress or
turmoil’.13 Hence, a hedge can reduce losses on average but might exhibit a positive correlation when
protection is most needed. In contrast, safe haven assets do not co-move with other assets or indices
in crisis periods but provide tail risk insurance. They possess either positive or negative correlation
during calmer periods.14 Even though these definitions are of high practical importance, they have
found only limited attention in the analysis of the gold–inflation relationship (but highly frequent
use in the analysis of the relationship between gold and other assets). However, they are sometimes
implemented indirectly by considering different inflation regimes in which the gold–inflation relation
may show different characteristics.
The studies listed in Table 2 empirically examine the gold–inflation relation. In light of the rising
awareness of inflation risks in the 1970s, Chua and Woodward (1982) were among the first to extend
the inflation hedging literature on stocks to the gold market. Using monthly and bi-annual data of six
industrialised countries they find that, with the exception of the US, gold has been neither a complete
nor partial hedge against domestic inflation. However, the fact that gold was fixed at different levels
under the Bretton Woods system and trading was prohibited in the US before 1975 limits the investigation period (1975–1980) and the validity of this study. In a larger sample (1975–1983) for the US,
Brown and Howe (1987) find that, in the short run, hedged cash positions (the investor purchases gold
and simultaneously sells a future contract), do not keep up with inflation. Jaffe (1989) analyses the
returns of gold and gold stocks from 1971 to 1987 within a multi-asset framework and provides the
first evidence of gold being a diversifier. The returns are found to be independent of those of other
assets which suggests that a portfolio benefits from a share of gold in it. However, he adds that gold
should not be relied upon as an effective inflation hedge because of insignificant Fisher regression
results and low coefficients of determination (R2 ) between 0.4% and 5.8%.
These limited and even contradictory initial findings have been reexamined in follow-up studies
(mostly focusing on the US market) with longer time spans and improved econometric methodology.
Mahdavi and Zhou (1997) investigate the relationship between gold prices and the CPI by means of
an ECM.15 Controlling for economic and financial conditions (measured by the real gross domestic
product and short-term interest rates), they find that gold is a poor out-of-sample forecasting variable
for inflation. Potential explanations of this finding are linked to the facts that first, the short-term
movement in the price of gold is much more volatile than the general price level and that second,
the CPI basket tends to adjust more slowly to new information. In contrast to a broader basket of
commodities, gold was not cointegrated with the CPI from 1970 to 1994. Hence, the two variables
may drift apart not only in the short run but also in the long run which impedes using gold as an
inflation hedge.
With longer time series covering 82 years, Taylor (1998) examines the inflation hedging ability
of the four precious metals: gold, silver, platinum and palladium.16 The hypothesis of non-stationary
residuals (in regressions of monthly gold returns on ex post inflation rates) could not be rejected
for the entire period from 1914 to 1996 implying that gold is not a long-run hedge against inflation.
However, gold and the other precious metals were hedges before 1939 and around the second OPEC
oil crisis in 1979. Thus, hedging properties appear to vary over time, indicating the need to take into
account structural breaks when analysing the gold–inflation relation.
13
Baur and McDermott (2010) extend these definitions by distinguishing a strong hedge (safe haven), which is negatively
correlated with other assets, from a weak hedge (safe haven), which is uncorrelated with other assets on average (in times of
stress).
14
Lucey and Li (2015) present a list of assets, currencies and even whole countries that have all been referred to as safe havens
with respect to various risks.
15
The results of earlier studies analysing a cointegration relation are different depending on the stationarity test and sample
period (see Mahdavi & Zhou, 1997).
16
Jastram (2009) uses the most extensive sample studied so far by analysing the annual change of gold and goods prices in the
US and the UK from 1560 to 2007 and 1808 to 2007, respectively. Unfortunately, his study has been heavily criticised because
of its mere descriptive nature and other flaws (see Hautcoeur, 2010).
200
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Structural breaks occur for a variety of reasons: economic and financial crises, changes in policy,
regime shifts or institutional changes. In either case, classic cointegration tests can yield spurious
results because the presence of structural breaks makes it harder to reject the null of no cointegration.17
To avoid such bias, Worthington and Pahlavani (2007) explicitly consider the role of structural breaks
in cointegration testing. Employing the unit root test proposed by Zivot and Andrews (2002) and a
cointegration test proposed by Saikkonen and Lütkepohl (2000a, 2000b, 2000c), they estimate the
timing of significant breaks rather than assuming the timing to be exogenous. In addition, the sample
period from 1945 to 2006 is scaled down to the period after 1973 when market forces rather than
the fixed or two-tiered gold system determined gold prices. The most significant structural breaks for
gold (and inflation) were found in 1973 and 1979, coinciding with the first and second OPEC oil crises.
Cointegration between gold prices and the US CPI was found for different econometric specifications,
indicating that gold can indeed serve as an inflation hedge in the long run.
In order to explain the theoretical conditions under which gold hedges changes in goods prices,
Ghosh, Levin, MacMillan, and Wright (2004) present a model of the long-run determinants of the
gold price. They show that, in a competitive market with gold producers as profit maximisers, the
price of gold equals the marginal extraction cost which in turn equals the marginal cost of leasing
gold from central banks. As long as the costs associated with extracting gold rise at the general rate
of inflation, the price of gold will rise at the general rate of inflation and gold can be regarded as an
inflation hedge. Changes in the gold lease rate, the real interest rate, convenience yield, default risk,
the covariance of gold returns with other assets and the dollar/world exchange rate18 disturb this
equilibrium relationship and generate short-run price volatility. Using data from 1976 to 1999, Ghosh
et al. (2004) find empirical support for their model and a long-run relationship between the US RPI
and the gold price. The implied elasticity of this relationship is consistent with the view that gold is
a long-run hedge against inflation. Levin, Montagnoli, and Wright (2006) provide a further analysis
of the nature of short-run and long-run gold movements. With monthly data from 1976 to 2005
they show that gold co-moves with the US price level (but not with world inflation) in a statistically
significant one-to-one relationship. Their estimated ECM implies that gold reverses slowly to the
long-term equilibrium (it takes about five years to eliminate two-thirds of the deviation from this
equilibrium). The factors responsible for short-term deviations from the equilibrium are US inflation,
US inflation volatility and credit risk (positively related to gold prices) as well as the US dollar tradeweighted exchange rate and the gold lease rate (negatively related).
With US data from 1970 to 2003, McCown and Zimmerman (2006) examine the role of gold as a
diversifier or hedge. In the context of a CAPM-based multifactor model containing various risk factors (including market risk, default and term spreads, changes in industrial production, and changes
in inflation), gold yields a market beta that is statistically indifferent from zero indicating that an
investment in gold adds no systematic risk to an investor’s portfolio.19 Furthermore, gold (and silver)
prices and the CPI are found to be cointegrated. In a subsequent study over a similar period, McCown
and Zimmerman (2007) provide evidence of a positive correlation between gold and expected future
inflation (measured by the yield spread between nominal US Treasury securities and the US Treasury
Inflation Protected Securities). Correlation tends to rise as the horizon increases from one month to
five years. Besides noting that gold resembles a useful indicator of expected inflation, the authors
conclude that gold hedges inflation, especially at longer horizons.
While the studies discussed so far assume that the relationship between the price of gold and
inflation is symmetric, a new strand of the literature suggests that it should be asymmetric. That
is, a change in the rate of inflation has less or more effect on the price of gold depending on the
state of financial markets.20 Studies following this logic use threshold regressions, asymmetric GARCH
17
For references covering the econometric problem of unit root tests with endogenous structural breaks, see Worthington
and Pahlavani (2007).
18
The world exchange rate is the nominal major currencies Dollar index of the Federal Reserve Board.
19
A sub-period analysis shows that negative betas occur in the high-inflation phase of the 1970s whereas the betas are positive
between 1980 and 2006.
20
Not only the state of the financial markets but also the business cycle, mechanisms of the price adjustment or transaction
costs may cause asymmetries in the gold-inflation relation (see Wang et al., 2011).
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
201
models, non-linear cointegration tests or threshold VECM because theses econometric approaches can
account for asymmetry. Taking into account earlier evidence of unstable relations between gold and
inflation recent studies also supplement their empirical analysis by applying Kalman filter regressions
or Markov switching models.
Wang, Lee, and Thi (2011) apply linear and non-linear cointegration tests (as proposed by Enders
& Siklos, 2001) to analyse the inflation hedging ability of gold in the long run and threshold VECM
to also cover the short-run perspective. They find that, between 1971 and 2010, gold and the CPI in
the US (in Japan) show linear, symmetric (non-linear, asymmetric) cointegration. Hence, using gold
as an inflation hedge is almost absolutely effective in the US in the long run, however, only partially
effective in Japan. The hedging ability in the short run depends on the adjustment momentum between
prices (low- vs. high-momentum regime). In low-momentum regimes, gold cannot be a hedge against
inflation in either the US or Japan. However, under high-momentum regimes, gold returns are able to
hedge against inflation in the US, whereas price rigidity in Japan causes the price of gold to not fully
hedge against inflation in the short run.
While most studies focus on the US market, Rubbaniy et al. (2011) take a European perspective by
examining gold’s role as a hedge against different risk factors in Germany. They build on the work of
McCown and Zimmerman (2006) and confirm their finding that gold bears no market risk. Rubbaniy
et al. (2011) extend the approach of McCown and Zimmerman (2006) by using GARCH and threshold
GARCH specifications for the error terms of the multifactor model to account for volatility clustering
and by adding a market sentiment variable which allows testing of whether or not the estimated
factor coefficients are conditional on the state of the economy. The factor exposures indeed differ
during stable and crisis periods (negative sentiment coefficient for gold in times of crisis while it has
no predictive power in stable periods) implying that from 1985 to 2010 the gold price rises when
the economic sentiment index declines. Thus, gold positions act as safe havens in times of distrust.
Furthermore, the study shows that gold (but not silver and palladium) is cointegrated with the German
CPI and thus is the only metal providing a long-run inflation hedge to German investors.
Also extending the country focus beyond the US, Beckmann and Czudaj (2013) cover 41 years of
data for the US, the UK, the Euro area and Japan, and report further evidence that the gold–inflation
relation is characterised by instability. By means of Markov switching VECM the authors allow for
regimes since shifts in the adjustment parameters between short-term deviations from the longterm equilibrium are incorporated. They find that the hedging effectiveness with regard to the CPI is
stronger than for the PPI. Furthermore, it is stronger in the US and the UK than in the Euro area and
Japan. Because fluctuations in the price of gold converge towards the long-run equilibrium at varying
speeds depending on the regime, returning to the long-run relationship can take several years. Once
again, this illustrates that time horizons matter for inflation hedging.
Even though most of the studies discussed so far suggest co-movement of gold and inflation over
long time horizons, recent studies (see Batten, Ciner, & Lucey, 2014; Baur, 2011; Blose, 2010; Erb &
Harvey, 2013) are more skeptical about a positive relationship between gold and inflation, albeit for
different reasons. Blose (2010) proposes two theories concerning the impact of expected inflation on
the price of gold: the expected inflation hypothesis (EIH), which states that higher inflation expectations will increase the demand for gold and thus its price, and the carrying cost hypothesis (CCH),
which states that changes in inflation expectations will not affect gold prices because higher inflation
expectations cause interest rates to rise which implies that the costs of holding gold increase. The
data basis for testing these hypotheses consists of inflation forecasts (monthly Consensus estimates
published by the Wall Street Journal prior to the CPI announcement) as a proxy for expected inflation
and unexpected inflation measured by the difference between announced and forecasted inflation.
Using non-linear regressions and a sample of more than 20 years, Blose (2010) shows that surprises
in the CPI do not affect gold spot prices.21 In other words, if the price of gold changes, the changes
are not caused by changes in expectations about future inflation. Thus, investors cannot determine
market inflation expectations by examining the price of gold.
21
With this result, Blose (2010) confirms the findings of Adrangi, Chatrath, and Raffiee (2003) that unexpected inflation does
not affect gold prices. However, Adrangi et al. (2003) show that expected inflation does influence gold prices.
202
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Baur (2011) examines the properties of gold (hedge against exchange rate risks, interest rate
changes and inflation) in a model that captures simultaneous influences. He criticises that most of
the studies reporting evidence on these properties use simple regression frameworks and therefore
assume that other variables potentially affecting the price of gold either do not exist or are at least
stable over time. Using a multiple regression framework for US data from 1979 to 2010, Baur (2011)
finds that the evidence for gold as an inflation hedge is at best weak (in contrast to the US dollar and
commodity prices that are hedged by gold). The author first estimates the isolated impacts on gold
prices by means of simple regressions and, in a second step, uses multiple regressions to show that
the hedging ability of gold loses strength and significance when interactions among the independent
variables are controlled for.
Erb and Harvey (2013) observe that a single outlier in the data, namely the year 1980, might account
for the positive relationship between gold and inflation documented in earlier studies. After excluding
this outlier, little evidence can be found that gold was an effective hedge against unexpected inflation
from 1975 to 2011 (neither in the short run nor the long run). Moreover, gold does not always provide
tail risk insurance during highly inflationary periods (safe haven property).
Finally, changes in the US inflation series in the early 1980s also played a role in the relation between
gold and inflation (see Kim & Ryoo, 2011). Stock and Watson (1999, 2007) and Atkeson and Ohanian
(2001) find a significant structural break in US inflation around 1984 which coincides with the onset
of the ‘Great Moderation’. For this reason, Batten et al. (2014) restrict their analysis to 1985–2012
to examine the relation between gold price and CPI and arrive at three important findings based on
the most sophisticated statistical methodology considering all features of the relation identified in
the literature. First, they find no cointegration between the two variables. The evidence is robust to
different cointegration tests and structural breaks. Second, the Kalman filter regressions point out that
the sensitivity of the gold price to changes in goods prices declines in the 1990s and increases in the
2000s which is the underlying reason why the hypothesis of a stable and stationary, i.e. cointegrating,
relation is rejected. Third, Granger causality tests reveal that the inflation sensitivity of gold increases
with lower short- and long-term interest rates. This is consistent with the notion that monetary easing
fuels higher inflation expectations which then increase the demand for gold. Hence, investors seeking
protection against inflation should look at these underlying macroeconomic factors that generate
time-variation in the relation between gold prices and inflation in order to evaluate gold’s hedging
potential.
Summarising the evidence on the empirical relationship between gold and inflation, we can say that
most of the earlier studies in the field report evidence that gold prices and inflation are cointegrated
while more recent work published after the financial crisis in 2008 presents a less favourable picture.
The latter studies show that the link between gold and inflation is either non-existent, merely driven
by outliers or strongly time-dependent. In some periods, gold has outperformed inflation, while in
other periods gold has failed to match it.
3.3. Fixed income
To judge the inflation-hedging ability of fixed income securities, we can collect evidence from
two literature strands. The first strand concentrates on analysing the Fisher effect by testing the
relationship between interest rates, i.e., the returns of short-term sovereign debt (money market
instruments or bills with a maturity of less than one year) and inflation. Thus, they implicitly analyse
hedging ability and mainly neglect bond returns (see Mishkin, 1992). The second strand explicitly tests
for an inflation hedging ability and covers bills, standard bonds and inflation-linked bonds (ILB).22
We start with a brief discussion of the first strand of literature. In general, early studies find only
limited empirical support for the validity of the Fisher hypothesis for interest rates. Some studies
suggest and provide theoretical rationales for a Fisher coefficient significantly different from unity.
Mundell (1963) and Tobin (1965) argue for partial interest rate adjustment, i.e., ˇ < 1, where a certain
22
Because there are already literature reviews covering the first strand of literature and the second strand only consists of a
handful of important studies, this section does not require an overview table of relevant studies.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
203
level of anticipated inflation raises nominal interest rates by less than the rate of inflation. Inflation
reduces real wealth, which could lead to a portfolio shift out of nominal assets and into real assets,
which in turn puts pressure on real interest rates.23 Although this effect opposes the Fisher (1930)
theory, it is in line with his empirical observations because, he argues, that ‘when the cost of living is
not stable, the rate of interest takes the appreciation and depreciation into account to some extent,
but only slightly, and, in general, indirectly’. Carmichael and Stebbing (1983) even propose an inverse
one-for-one relationship between after-tax real interest rates and inflation. The conceptual antipodes
of an inverse Fisher effect are Darby (1975) and Taylor (1993). Taylor (1993) argues for monetary policy
rules rather than discretionary interventions of monetary authorities. In his opinion, nominal interest
rates tend to overestimate inflationary pressure which is why nominal rates may overshoot actual
inflation. In other words, he predicts a positive relation ˇ > 1. Darby’s (1975) tax-augmented version
also postulates Fisher coefficients above unity since investors must be compensated for their tax
burden. A one-percentage point increase in expected inflation would then increase nominal rates by
1/(1 − ) percentage points with being an investor’s marginal tax rate on nominal interest income.24
Several other early empirical studies examine inflation and interest rates in the context of inflation
forecasting and the efficient market hypothesis. As summarised by Fama (1975), early findings indicate that the market fails to forecast inflation rates. There are no relationships between interest rates
at time t and the subsequent rates of inflation. However, there is a clear link between current interest
rates and past inflation which would support the Fisher hypothesis. Analysing one- to six-month US
treasury bills, Fama (1975) presents evidence that the Fisher hypothesis holds in the short run. Since
real interest rates are constant from 1953 to 1971, variations in nominal rates arise from changes
in expected inflation. Unlike the majority of studies at that time, he does find relationships between
nominal interest rates and subsequently observed inflation rates. His findings support the notion of
an efficient capital market in the sense that nominal interest rates always incorporate and reflect all
relevant information about future inflation rates.25 There are at least two rationales why the unexpected component after inflation announcements should lead to a change in interest rates. Unexpected
inflation might trigger higher inflation expectations in the future (expected inflation hypothesis) or
monetary authorities are expected to tighten their policy in response to unexpected inflation (policy
anticipation hypothesis). Indeed, Urich and Wachtel (1984) show that short-term interest rates are
positively related to unexpected inflation announcements measured by changes in the PPI while this
effect is not observable in the CPI. Smirlock (1986) reports a significant positive response of long-term
bond markets to unexpected inflation. Moreover, the short- and long-term responses are immediate
which supports market efficiency.
Starting with Rose (1988), the empirical literature has tested whether interest rates and inflation
contain a unit root because stationary real interest rates imply that they are independent of inflation
and thus support the Fisher hypothesis. Neely and Rapach (2008) published a comprehensive review
of 29 studies covering the long-run properties of real interest rates; only four examine both ex ante
real rates (EARR) and ex post real rates (EPRR), and the remaining 25 studies analyse only EPRR.26 The
reviewed studies were published between 1988 and 2008 and use monthly or quarterly data. While
most of them focus on the US market, others mainly employ panel data from industrialised countries
such as a set of OECD countries or the G7.27 Although the evidence concerning (non-)stationarity of
interest rates and inflation is mixed, there are a few commonalties. First, the majority of studies fails to
reject a unit root for nominal interest rates as well as inflation. Second, EARR appear to be stationary.
23
Evans and Lewis (1995) present additional theoretical explanations such as liquidity premiums in general equilibrium
models or output shocks. Crowder and Hoffman (1996) provide references for empirical findings on the Mundell–Tobin effect.
24
Allowing for tax effects, the estimated coefficient should be in the range of 1.3–1.4 for a tax rate around 25% (see Summers,
1983).
25
Fama (1975) has been challenged mainly for statistical reasons such as low testing power (see Nelson & Schwert, 1977)
and dependency on the particular sample period (see Mishkin, 1984). Relaxing the stationarity assumption of Fama’s OLS
regressions, Garbade and Wachtel (1978) conclude that the real rate of interest varied over the sample period from 1953 to
1971.
26
Studies on real interest rate persistence after 2008 include Yoon (2010), Haug, Beyer, and Dewald (2011), Tsong and Lee
(2013) and Haug (2014).
27
The only exception is Lai (1997, 2008) who analyzes eight industrialised and developing countries each.
204
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
However, this conclusion should be treated with caution because of the limited number of reviewed
studies. Third, the findings with respect to EPRR are puzzling since a rejection of a unit root is found
by almost the same number of studies that present evidence of non-stationary EPRR. One possible
explanation for this observation is regime-dependent behaviour of EPRR (see Garcia & Perron, 1996).
Fourth, non-linear cointegration (see Christopoulos & León-Ledesma, 2007; Lai, 1997; Maki, 2003;
Million, 2004; Sun & Phillips, 2004) and the comprehension of structural breaks (see Bai & Perron,
2003; Caporale & Grier, 2000; Clemente, Montañés, & Reyes, 1998; Lai, 2004, 2008; Rapach & Wohar,
2005) have become popular in more recent studies and tend to support stationary EPRR and (at least
fractional) cointegration between nominal interest rates and inflation.
Turning to our second strand of literature that directly analyses inflation hedging with fixed income
securities, the seminal study by Fama and Schwert (1977) finds that US government bills and bonds
are a complete hedge only against expected inflation. Unexpected inflation has a strong negative
relationship with ex post returns. The magnitude of the effect increases with terms to maturity since
long-term bonds have more future periods that require adjustments in expected nominal returns.
Bekaert and Wang (2010) make the same point, stating that bonds fail to hedge unexpected inflation.
In a cross-section of 45 countries with annual government bond returns from 1970 to 2010, 17 out
of 19 statistically significant inflation betas of bond returns are negative, ranging from around 0 to
−3. The hedging coefficients improve as the horizon increases. Since substantial inflation shocks are
inimical to the portfolio returns of bond holders, it is no surprise that the bond markets of developed
countries outperform emerging economies which regularly suffer periods of high inflation.
The instability of the real rate of return on bills over long periods of time as well as the poor inflation
hedging characteristics of bonds gave rise to the new class of ILB which link principal payments and/or
coupon payments to a specific measure of inflation.28 The benefits of ILB are that (i) a new long-term
real safe asset exists, (ii) financial markets become more complete due to additional possible payoff
structures and improved real-time market information, and we have (iii) debt savings due to positive
inflation risk premiums, (vi) predictable real financing costs and (v) a reduction of the incentive for
governments to inflate their debt (see Garcia & Van Rixtel, 2007). In a portfolio context, ILB exhibit a
low correlation with other asset classes and a lower volatility of real returns (see Kothari & Shanken,
2004; Swinkels, 2012). However, some drawbacks of ILB are that the ILB market still accounts for only
a small proportion of government debt leading to poor liquidity compared to standard bonds and the
costs associated with issuing ILB are higher than for standard bonds. Moreover, a particular inflation
index may not match the inflation exposure of an investor because the index measurement may be
biased or lagged (see Bekaert & Wang, 2010; Campbell & Shiller, 1996).
Amenc et al. (2009) show that if short-term liability risk hedging is the sole focus in a framework where liabilities are indexed with respect to inflation, the optimal asset allocation consists of
investing 100% into an inflation-indexed bond portfolio. Brière and Signori (2012) examine the asset
return dynamics of cash, nominal bonds, ILB, equities, real estate and precious metals by means of
a VAR model for varying investment horizons from 1 month to 30 years. Despite the authors’ focus
on intertemporal portfolio decisions, they also report findings for individual assets. They show that
the inflation hedging properties of nominal bonds and ILB strongly differ depending on the regime
(1973 to 1990 or 1991 to 2010) and hedging horizon. In the first regime, nominal bond returns show
negative correlation coefficients with inflation up to −0.7 at all horizons, whereas ILB coefficients
become positive for horizons greater than five years. In the second regime, both types of bonds show
positive coefficients for horizons around eight to ten years. The study is complemented by an analysis
of shortfall probabilities according to which nominal bonds performed well with a probability of not
achieving the inflation target of 7% (first regime) and 0% (second regime) at 30-year horizons. This
performance may be explained by the significant fall in the inflation risk premium due to persistent
disinflation.
In short, the literature studying the interplay of interest rates and inflation has not been accompanied by a comparable quantity of studies that explicitly test the inflation hedging effectiveness of
28
For a comprehensive discussion of ILB (market development, issuance and information extraction from ILB), see Garcia and
Van Rixtel (2007).
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
205
fixed income securities. State dependent behaviour and structural breaks in time series of real interest rates impede the detection of cointegration between nominal interest rates and inflation. Bills
and bonds with fixed principal or coupon payments are vulnerable to inflation shocks. They show
negative correlation with inflation in the short run but have some potential as inflation hedges in the
very long run. ILB bonds technically protect investors against inflation. However, scarce liquidity and
limited available maturities confine their use for investors. ILB unfold their inherent advantages (low
correlation with other asset classes, low real return volatility) mainly in a portfolio context.
3.4. Real estate
Because of the shortcomings of ILB markets, most investors continue to rely on the indirect hedging
properties of traditional asset classes. Real estate is the last of these asset classes that we discuss in our
review. It is a heterogeneous asset class that comprises diverse investment vehicles. Research on the
inflation hedging properties of real estate has to differentiate between private and public real estate,
residential and commercial properties as well as indirect (securitised) and direct (income-producing)
means of participating in the real estate sector. The rationale for a potential hedging property, however,
does not differ. Real estate represents an underlying value and its returns derive from both income
(rents might be adjusted in line with the general price level) and capital appreciation (due to higher
demand, for example through real output growth). One determinant of the inflation sensitivity of
real estate is the lease structure including the length of the lease agreement, indexed escalations
and pass-through of expenses (see Huang & Hudson-Wilson, 2007; Le Moigne & Viveiros, 2008). The
identification of other major influences on the inflation hedging effectiveness of real estate is the
subject of current empirical work.
In the analysis of private and public real estate, we encounter the general problem that price
information on real estate markets is often deficient (in quality and quantity), because of a lack of
transparency or an insufficient number of trades which impedes market tracking and thereby research
in general. This is especially true of private real estate and the reason why researchers often resort to
appraisal-based indices. Appraisers base their estimates of the market value of a real estate property on
past and current information about fundamental variables and transactions. The problems associated
with an appraisal-based approach in which information must be extracted from noisy transaction
prices can be subsumed under the term ‘appraisal smoothing’.29 It occurs not only in the appraisal
process itself at the disaggregate (property) level, but also at the aggregate level due to the frequency
of aggregation (time lags). Examining the relationship between real estate returns and inflation on the
basis of appraisals may be biased by the fact that appraisers tend to adjust the property value estimate
made in t − 1 by an inflation factor when estimating the value of a property in t. Hence, it is hardly
surprising that positive coefficients are found when such returns are regressed on inflation. However,
there is still no consensus on whether return series should be ‘unsmoothed’ because the revision of
time series itself may produce misleading results regarding the inflation hedging ability of real estate
(see Le Moigne & Viveiros, 2008).
In general, direct real estate (ownership without financial vehicles) has several disadvantages, such
as large required fund outlays, lack of a central market, need for local market knowledge, low liquidity,
high transaction costs, maintenance expenditures and management requirements (see Wilson &
Zurbruegg, 2003). These drawbacks, in conjunction with the difficulties posed by low frequency data
and questions about the reliability of appraisal-based returns have drawn academics’ attention to
securitised real estate markets and have given rise to the introduction of real estate investment trusts
(REITs) that pool the funds of investors in order to invest in income-producing mortgages, real estate
properties, joint ventures and other hybrid structures. REITs can be subdivided into mortgage REITs,
which hold mortgages and construction loans, and equity REITs, which specialise in income-producing
properties (see Chen & Tzang, 1988). REITs and real estate stocks should be treated separately from the
discussion in Section 3.1 because (i) the underlying assets of REITs are primarily real estate, because
29
Geltner, MacGregor, and Schwann (2003) provide an overview of the estimation of market values by appraisers and the
informational efficiency of the public and private real estate markets.
206
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
(ii) although evidence suggests that REITs tend to behave more like common stocks than like real
estate, the equity component of REITs might be declining (see Adrangi, Chatrath, & Raffiee, 2004), and
because (iii) REITs possess stable cashflows due to high dividend payouts, a feature that differs from
stocks.30
Starting with Fama and Schwert (1977) the inflation hedging property of real estate has received
considerable attention in the literature. Because other scholars have already summarised the early
results, we have structured our review as follows. We first discuss the initial findings of Fama and
Schwert (1977) and summarise the results from the overviews of Simpson, Ramchander, and Webb
(2007) and Hoesli, Lizieri, and MacGregor (2008). We then review the most important recent studies
using more advanced statistical methodology (see Table 3).
Fama and Schwert (1977) use transaction-based real estate returns for the US market (the
home purchase price component of the CPI) and find that nominal real estate returns co-move
in a one-to-one relation with inflation. Specifically, they show that private residential real estate
can provide an effective hedge against both expected and unexpected inflation. Although the findings of follow-up studies differ with respect to the hedging coefficients, they show that private
real estate can be a good hedge against expected inflation.31 However, in most cases, the hedging
coefficients regarding the rate of inflation are significantly less than 1 and indicate only a partial hedge. Furthermore, they present evidence that even though private real estate offers hedges
against expected inflation, it fails to hedge against unanticipated inflation shocks (see Simpson et al.,
2007).
The question of whether these findings also hold for securitised forms of real estate and the
issue of data-related shortcomings of unsecuritised real estate introduced a new strand of literature using REIT returns. Interestingly, early findings for securitised real estate resemble those for
stocks from the late 1970s. REIT returns appear to be a perverse hedge against inflation (negative
or insignificant hedging coefficients), in sharp contrast to the partial hedging properties of unsecuritised real estate. However, most early studies of inflation hedging with REITs test the Fama
and Schwert (1977) model or its extensions based on ‘deficient’ OLS approaches.32 The negative or
insignificant relationship between REIT returns and inflation has been documented for various sample periods and countries and remained largely unexplained until more recent research based on
advanced econometric models (such as cointegration techniques) shed light on the subject. These
approaches that are able to differentiate between any long-run equilibrium and short-run dynamic
adjustments provided potential explanations including (i) spurious regression, (ii) differences in information processing between REITs and direct real estate, (iii) long- and short-run dynamics, (vi)
regime dependency and (v) asymmetric adjustments of REIT returns to inflation (see Hoesli et al.,
2008). However, also note that a long-term equilibrium relationship between real estate and inflation must not necessarily imply a hedge at long horizons because real estate may influence inflation.
This introduces endogenity biases whose problems have to be addressed by instrumental variable
estimation.
Turning to our review of studies using methodologies beyond simple OLS regressions, we start
with Tarbert (1996). Based on quarterly and semi-annual data from 1978 to 1995, he finds no evidence of a long-term hedging ability of British commercial real estate. Cointegration tests do not
yield evidence for any stable relationship between commercial property and inflation. This result
has been confirmed by Stevenson (1999) who employs both an OLS-based Fama and Schwert (1977)
framework to examine commercial and residential property in the UK (mixed evidence concerning
30
If a large number of REITs become included in stock indices that are negatively correlated with inflation in the short run,
this could weaken the hedging function of REITs, as the volatility from the stock market can overweigh the stability of REIT
dividend payments over time (see Hardin et al., 2012).
31
Appraisal-based returns, as used in Hartzell, Hekman, and Miles (1987), tend to yield higher hedging coefficients than
transaction-based returns or returns that are corrected for appraisal smoothing.
32
For example, Yobaccio, Rubens, and Ketcham (1995) propose a modified Fama and Schwert (1977) approach termed capital
asset pricing model under uncertain inflation (CAPMUI), which relates expected real returns to inflationary expectations. Tarbert
(1996) challenges their findings that are based on static regression approaches because they cannot capture slow adjustments
of the real estate markets to changes in inflation.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
207
inflation hedging) and cointegration analysis (neither residential nor commercial property is cointegrated with inflation between 1983 and 1995). However, Barkham, Ward, and Henry (1996) and
Matysiak, Hoesli, MacGregor, and Nanthakumaran (1996) provide contradictory results, implying
that British direct real estate does hedge inflation measured by the RPI in the long run. By means
of unit root, maximal eigen values and trace tests, Matysiak et al. (1996) show that while commercial
real estate does not hedge inflation in the short run, there is a long-term inflation hedging characteristic of direct real estate in contrast to property shares. Barkham et al. (1996) conclude that
real estate returns and inflation are cointegrated with persistent deviations from the equilibrium.
Furthermore, inflation appears to Granger-cause real estate but real estate does not Granger-cause
inflation.
The findings in countries other than the UK also tend to favour an equilibrium between real estate
returns and inflation over longer horizons. Anari and Kolari (2002), Park and Bang (2012) and Obereiner
and Kurzrock (2012) all find direct real estate to be cointegrated with inflation in the US, Korea and
Germany, respectively. Based on both autoregressive distributed lag (ARDL) models and recursive
regressions, Anari and Kolari (2002) find Fisher coefficients that are consistently above unity, ranging
from 1.08 to 1.42 in the US between 1968 and 2000. Although Korean direct real estate markets lack
reliable, long and transaction-based time series,33 Park and Bang (2012) show that, contrary to listed
real estate which provides a hedge only over longer horizons, appraisal-based commercial real estate
provides a hedge against inflation, both in the short run and in the long run. Finally, Obereiner and
Kurzrock (2012) examine three indirect real estate investment vehicles in Germany from 1992 to
2009: open-end funds, special funds and real estate stocks. The returns are almost independent of
inflation in the short run. However, causality tests and different cointegration models suggest that
indirect real estate provides a long-run hedge.
The studies of Ganesan and Chiang (1998), Glascock, Lu, and So (2000, 2002), Hoesli et al. (2008),
Le Moigne and Viveiros (2008) and Hardin, Jiang, and Wu (2012) advocate that the inflation hedging ability of real estate depends on the specific type of real estate asset, is regime-dependent or
(dis)appears over time. For example, the findings of Ganesan and Chiang (1998) based on the Fama
and Schwert (1977) framework differ widely depending on the type of real estate. Commercial and
residential real estate in Hong Kong possesses good hedging properties with respect to expected and
unexpected components of inflation in the sample period of 1984–1994, whereas other private real
estate and property stocks fail to hedge inflation in the short run. In the long run, only property stocks
were found to be cointegrated with inflation.
Glascock et al. (2002) consider three potential rationales for the negative relationship between
REIT returns and inflation in the literature that contradicts the alleged hedging property of direct real
estate. First, REITs might be better at processing information than the general real estate sector, thus
instantaneously reflecting future economic prospects. Second, the usage of data that contain more
recent rather than future information could bias unsecuritised real estate returns. Third, improper
econometric approaches may reverse the causal relationships between REIT returns and inflation.
Introducing monetary regimes, the authors show that the perverse inflation hedging property of REITs
is attributable merely to changes in monetary policy and therefore spurious. REIT returns seem to
anticipate (Granger-cause) changes in expected and unexpected inflation.
A similar approach that combines the proxy hypothesis framework with a long-run framework
(covering the linkages between inflation, asset returns as well as macroeconomic and monetary variables) has been taken by Hoesli et al. (2008) for US and UK data from 1977 to 2003. According to
their study, real estate returns are positively linked to expected inflation, but not to (unanticipated)
inflation shocks. The adjustment to changes in inflation is gradual, implying that real estate does not
always offer short-term protection.34 Inflation is positively linked to private and public commercial
33
While initially there has been much attention to American and British real estate markets, Asian markets have become
the focus of recent interest. This entails new obstacles such as a lack of privatised real estate markets and limited access to
investment opportunities. China, for instance, faced a gradual renunciation from a welfare-based approach to a market-based
approach not until the late 1990s (see Zhou & Clements, 2010).
34
This finding is in line with the unfavourable (moderate) short-run (long-run) hedging ability identified by Froot (1995),
Hoevenaars, Molenaar, Schotman, and Steenkamp (2008), Amenc et al. (2009) and Brière and Signori (2012) in a multi-asset
208
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
real estate once the impact of macroeconomic and monetary variables is considered, a result that
holds for the UK and the US. Real estate returns hedge expected inflation one-to-one whereas the
coefficients for unexpected inflation are either negative or insignificant. Private real estate appears
to outperform public market assets in the sense that privately-owned real estate exhibits significant
coefficients for both expected and unexpected inflation.
Like Glascock et al. (2002), Lee and Lee (2012) base their study on a US total return REIT index.
By means of dynamic OLS models (DOLS) they show that REITs serve as a hedge against expected
inflation only after a structural break in 1993, where a tax reform made large-scale investments in
REITs more desirable to institutional investors. Furthermore, they highlight that the hedging capability of REITs is driven by large-cap REITs but small-cap REITs fail to hedge inflation once isolated
from the influence of large REITs. Hardin et al. (2012) also take into account the structural break in
the US by splitting the sample period into two sub-periods (1980–1992 and 1993–2008) in order to
obtain robust evidence on real estate stocks and REITs. By employing dividend yield composition,
the authors show that, while both inflation illusion and hedging effects exist in REITs, the effect of
inflation illusion seems to dominate throughout the entire sample period, from 1980 to 2008. Their
results support the Modigliani and Cohn (1979) hypothesis that investors cannot quickly reconcile
the changes in discount rates and dividend growth rates associated with inflation into stock prices.
Thus, even though we might observe that REIT stock prices are negatively related to expected inflation, REITs compensate investors for the expected component of inflation via the dividend growth
rate.
Le Moigne and Viveiros (2008) examine how a variety of Canadian real estate types (by property
type and province) hedge against inflation based on unsmoothed real estate returns from 1973 to 2007.
They find that Canadian real estate acted as an inflation hedge over the sample period but also that
this inflation-hedging ability has disappeared since the mid-1980s. To check for robustness, the tests
are repeated for two subperiods (1973–1991 and 1992–2007) that coincide with the introduction of
an inflation targeting policy by the Bank of Canada aiming to maintain inflation in the 1–3% range. The
findings confirm the argument of Stevenson and Murray (1999) that real estate loses its protection
against inflation in regimes with low rates of inflation.
While Chatrath and Liang (1998) and Adrangi et al. (2004) support the traditional notion that
REITs do not hedge inflation (in contrast to direct real estate) with almost identical sample periods
from 1972 to 1995 and 1972 to 1999, respectively, Simpson et al. (2007) consider the asymmetric behaviour of REIT returns in explaining the alleged perverse inflation behaviour of REITs. Based
on monthly data for 195 publicly traded equity REITs (EREITs) between 1981 and 2002, they use
fixed-effect pooled regressions to allow for firm-specific differences via different intercept terms
for each cross-section of the analysis. The authors conclude that the response of EREIT returns to
inflation is asymmetric. In other words, when expected and unexpected inflation are separated
into positive and negative changes, results indicate that EREIT returns rise in response to both
increases and decreases in inflation. Rather than attributing the traditionally found negative relation to proxy effects or regime dependence, Simpson et al. (2007) claim that it is in fact this
asymmetric adjustment of returns in conjunction with monetary policy effects that explain the
puzzling behaviour. The findings of previous studies would then be nothing but ‘an artefact of the
methodology they employ, which implicitly assumes symmetrical responses of EREIT returns to inflation’.
In summary, empirical evidence indicates that even though direct real estate features severe datarelated and practical shortcomings, it is regarded as an (at least partial) inflation hedge by earlier
studies that rely mainly on static regression approaches. In contrast, early studies of REIT returns
have found a negative relation to inflation. With the advent of cointegration approaches in the real
estate literature in the 1990s, the evidence on the inflation-hedging properties of real estate assets has
become more ambiguous. Cointegration techniques as well as causality tests in more recent studies
perspective where the focus of interested in how an asset reduces inflation exposure of a portfolio when it is added to the
portfolio. It is supplemented by Bekaert and Wang (2010) showing that real estate consistently underperforms stocks in hedging
expected and unexpected inflation.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
209
have produced mixed results on whether direct and indirect real estate is capable of hedging inflation.
In particular, the choice of the sample period, inflation regime dependency, the type of real estate and
asymmetries in the reaction of real estate returns to inflation news appear to play a crucial role in the
interplay of real estate returns and the rate of inflation.
4. Conclusion
Recent research has directed considerable effort towards answering the important question of
whether investments in certain assets can shield investors from inflation risk. In this line of scientific
work, the basic feature of interest is the interplay between asset returns and inflation rates because
a positive relationship implies that asset returns (fully or partially) compensate a rising inflation
rate. Similar to other fields, studies evolve by application of more advanced econometric techniques,
by considering previously detected data features (such as structural breaks and time-dependency
of variables) or by extension to new countries or longer time-spans. This process naturally leads to
a wide variety of (potentially conflicting) results that require structuring in order to draw overall
conclusions.
Reviewing the literature on the inflation hedging properties of the four major assets (asset classes)
stocks, gold, fixed income securities, and real estate, we provide such a structure that allows the
detection of changing perceptions of the effectiveness of inflation hedging over time and to identify
gaps that might be addressed in future research. Overall, our review describes the state of scientific
knowledge concerning the link between asset returns and inflation.
Most of the recent studies, using more sophisticated statistical methodology than the seminal
studies in the field, find a long-run equilibrium between common stocks and inflation implying an
effective inflation hedge at horizons of five years or more. In contrast, gold, typically advertised by the
media as an ideal hedge because of its historically documented ability to rise in times of recession or
financial turmoil, does not seem to offer such protection. Studies published in the aftermath of the
financial crisis in 2008 report strong time variations in the gold–inflation linkage caused by structural
breaks, different strength of the link depending on the investment horizon or even argue that outliers
are the sole reason for historically documented positive relationships. As far as fixed income securities
are concerned, nominal bills and bonds cannot hedge against inflation shocks. Inflation-linked bonds
are designed to protect investors against inflation but scarce liquidity and limited available maturities
restrain investors from using them. Finally, while the overall picture suggests that real estate might be
an (at least partial) inflation hedge especially in the long run, several technical issues limit the validity
of previously published results. While the analysis of direct real estate suffers from severe data-related
and practical shortcomings, the more favourable indirect instrument of REITs also provides only a
mixed picture because, for example, state dependence and asymmetric price reactions play a crucial
role in the interplay of real estate returns and the rate of inflation.
Given that even the most recent studies have produced partially conflicting results, how should
future research proceed? To obtain a clear answer on the inflation-hedging question, we need a
full-scale analysis that covers all relevant dimensions and crucial impact factors identified in previous studies. First, it has to encompass all asset classes (stocks, gold, fixed income securities and
real estate) simultaneously in order to allow a direct comparison of results and identify which
asset can be regarded as the best hedge. As far as the analysis of common stocks is concerned,
it would be interesting to supplement the typical use of stock marked indices by the direct use
of highly liquid individual stocks because they are typical components of private investors’ portfolios and because this way index construction methodologies cannot distort results. Second, we
should focus on the safe haven property rather than a general hedging property because the latter offers protection only on average but not necessarily in times when it is most needed: in
phases of high inflation. Besides considering this inflation regime dependency, it is also important to take into account other kinds of structural breaks such as periods of fixed gold prices,
significant advancements in inflation measurement, regulatory changes in REITs markets or crucial financial turbulences. Third, the asset classes should not be analysed in an isolated context
linked to only inflation. We require a system that models interactions between the different asset
classes and also other variables that might influence returns (e.g. macroeconomic variables). Of
210
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
course such a system has to allow for long-term equilibria and short-term deviations thereof. Fourth,
especially for real estate, data flaws have to be ruled out (e.g. by concentrating on REITs only).
Furthermore, potential endogenity biases related to the interaction of housing and goods prices have
to be addressed. Fifth, to obtain not only country-specific evidence, we have to consider a representative cross-section of markets. Finally, the analysis should cover several investment horizons in order
to distinguish short-term from long-term inflation protection.
Considering all these aspects allows giving a profound answer to the question of when, where and
to what extend investments in stocks, gold, fixed income securities and real estate can shield investors
from inflation risk. Given today’s situation of close to zero interest rates, we might expect that bills
cannot protect against inflation and arising bubbles in stock and real estate markets (especially in
Europe) that are the result of money shifts from bills to stocks and real estate have a crucial effect on
the historically documented hedging ability of these assets.
Acknowledgement
We thank an anonymous reviewer for valuable comments and suggestions.
References
Adrangi, B., Chatrath, A., & Raffiee, K. (2003). Economic activity, inflation, and hedging: The case of gold and silver investments.
Journal of Wealth Management, 6(2), 60–77.
Adrangi, B., Chatrath, A., & Raffiee, K. (2004). REIT investments and hedging against inflation. Journal of Real Estate Portfolio
Management, 10(2), 97–112.
Ahmed, S., & Cardinale, M. (2005). Does inflation matter for equity returns? Journal of Asset Management, 6(4), 259–273.
Alagidede, P., & Panagiotidis, T. (2010). Can common stocks provide a hedge against inflation? Evidence from African countries.
Review of Financial Economics, 19(3), 91–100.
Amenc, N., Martellini, L., & Ziemann, V. (2009). Inflation-hedging properties of real assets and implications for asset-liability
management decisions. Journal of Portfolio Management, 35(4), 94–110.
Anari, A., & Kolari, J. (2002). House prices and inflation. Real Estate Economics, 30(1), 67–84.
Atkeson, A., & Ohanian, L. E. (2001). Are Phillips curves useful for forecasting inflation? Federal Reserve Bank of Minneapolis
Quarterly Review, 25(1), 2–11.
Bai, J., & Perron, P. (2003). Computation and analysis of multiple structural change models. Journal of Applied Econometrics, 18(1),
1–22.
Barkham, R. J., Ward, C. W., & Henry, O. T. (1996). The inflation-hedging characteristics of UK property. Journal of Property Finance,
7(1), 62–76.
Batten, J. A., Ciner, C., & Lucey, B. M. (2014). On the economic determinants of the gold–inflation relation. Resources Policy, 41,
101–108.
Baur, D. G. (2011). Explanatory mining for gold: Contrasting evidence from simple and multiple regressions. Resources Policy,
36(3), 265–275.
Baur, D. G., Beckmann, J., & Czudaj, R. (2014). Gold price forecasts in a dynamic model averaging framework: Have the determinants
changed over time? Ruhr Economic Papers No. 506.
Baur, D. G., & Lucey, B. M. (2010). Is gold a hedge or a safe haven? An analysis of stocks, bonds and gold. Financial Review, 45(2),
217–229.
Baur, D. G., & McDermott, T. K. (2010). Is gold a safe haven? International evidence. Journal of Banking & Finance, 34(8), 1886–1898.
Beckmann, J., Berger, T., & Czudaj, R. (2015). Does gold act as a hedge or a safe haven for stocks? A smooth transition approach.
Economic Modelling, 48, 16–24.
Beckmann, J., & Czudaj, R. (2013). Gold as an inflation hedge in a time-varying coefficient framework. North American Journal of
Economics and Finance, 24, 208–222.
Bekaert, G., & Engstrom, E. (2010). Inflation and the stock market: Understanding the Fed model. Journal of Monetary Economics,
57(3), 278–294.
Bekaert, G., & Wang, X. (2010). Inflation risk and the inflation risk premium. Economic Policy, 25(64), 755–806.
Białkowski, J. P., Bohl, M. T., Stephan, P. M., & Wisniewski, T. P. (2014). The gold price in times of crisis. International Review of
Financial Analysis (in press).
Blose, L. E. (2010). Gold prices, cost of carry, and expected inflation. Journal of Economics and Business, 62(1), 35–47.
Bodie, Z. (1976). Common stocks as a hedge against inflation. Journal of Finance, 31(2), 459–470.
Boudoukh, J., & Richardson, M. (1993). Stock returns and inflation: A long-horizon perspective. American Economic Review, 83(5),
1346–1355.
Brière, M., & Signori, O. (2012). Inflation-hedging portfolios: Economic regimes matter. Journal of Portfolio Management, 38(4),
43–58.
Brown, K. C., & Howe, J. S. (1987). On the use of gold as a fixed income security. Financial Analysts Journal, 43(4), 73–76.
Campbell, J. Y., & Shiller, R. J. (1996). A scorecard for indexed government debt. NBER Macroeconomics Annual, 11, 155–208.
Caporale, T., & Grier, K. B. (2000). Political regime change and the real interest rate. Journal of Money, Credit and Banking, 32(3),
320–334.
Carmichael, J., & Stebbing, P. W. (1983). Fisher’s paradox and the theory of interest. American Economic Review, 73(4), 619–630.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
211
Charemza, W. W., Hristova, D., & Burridge, P. (2005). Is inflation stationary? Applied Economics, 37(8), 901–903.
Charles, A., Darné, O., & Kim, J. H. (2015). Will precious metals shine? A market efficiency perspective. International Review of
Financial Analysis (in press).
Chatrath, A., & Liang, Y. (1998). REITs and inflation: A long-run perspective. Journal of Real Estate Research, 16(3), 311–326.
Chen, K., & Tzang, D. (1988). Interest-rate sensitivity of real estate investment trusts. Journal of Real Estate Research, 3(3), 13–22.
Christopoulos, D. K., & León-Ledesma, M. A. (2007). A long-run non-linear approach to the Fisher effect. Journal of Money, Credit
and Banking, 39(2–3), 543–559.
Chua, J., & Woodward, R. S. (1982). Gold as an inflation hedge: A comparative study of six major industrial countries. Journal of
Business Finance & Accounting, 9(2), 191–197.
Ciner, C., Gurdgiev, C., & Lucey, B. M. (2013). Hedges and safe havens: An examination of stocks, bonds, gold, oil and exchange
rates. International Review of Financial Analysis, 29, 202–211.
Clemente, J., Montañés, A., & Reyes, M. (1998). Testing for a unit root in variables with a double change in the mean. Economics
Letters, 59(2), 175–182.
Cochran, S. J., & Defina, R. H. (1993). Inflation’s negative effects on real stock prices: New evidence and a test of the proxy effect
hypothesis. Applied Economics, 25(2), 263–274.
Coudert, V., & Raymond-Feingold, H. (2011). Gold and financial assets: Are there any safe havens in bear markets? Economics
Bulletin, 31(2), 1613–1622.
Crowder, W. J., & Hoffman, D. L. (1996). The long-run relationship between nominal interest rates and inflation: The Fisher
equation revisited. Journal of Money, Credit and Banking, 28(1), 102–118.
Culver, S. E., & Papell, D. H. (1997). Is there a unit root in the inflation rate? Evidence from sequential break and panel data
models. Journal of Applied Econometrics, 12, 435–444.
Darby, M. R. (1975). The financial and tax effects of monetary policy on interest rates. Economic Inquiry, 13(2), 266–276.
DeJong, D. N., Nankervis, J. C., Savin, N. E., & Whiteman, C. H. (1992). The power problems of unit root test in time series with
autoregressive errors. Journal of Econometrics, 53(1), 323–343.
Ely, D. P., & Robinson, K. J. (1997). Are stocks a hedge against inflation? International evidence using a long-run approach. Journal
of International Money and Finance, 16(1), 141–167.
Enders, W., & Siklos, P. L. (2001). Cointegration and threshold adjustment. Journal of Business & Economic Statistics, 19(2),
166–176.
Engle, R. F., & Granger, C. W. (1987). Co-integration and error correction: Representation, estimation, and testing. Econometrica,
55(2), 251–276.
Engsted, T., & Tanggaard, C. (2002). The relation between asset returns and inflation at short and long horizons. Journal of
International Financial Markets, Institutions and Money, 12(2), 101–118.
Erb, C. B., & Harvey, C. R. (2013). The golden dilemma. Financial Analysts Journal, 69(4), 10–42.
Evans, M. D., & Lewis, K. K. (1995). Do expected shifts in inflation affect estimates of the long-run Fisher relation? Journal of
Finance, 50(1), 225–253.
Fama, E. F. (1975). Short-term interest rates as predictors of inflation. American Economic Review, 65(3), 269–282.
Fama, E. F. (1981). Stock returns, real activity, inflation, and money. American Economic Review, 71(4), 545–565.
Fama, E. F., & Schwert, G. W. (1977). Asset returns and inflation. Journal of Financial Economics, 5(2), 115–146.
Fisher, I. (1930). The theory of interest. New York: Macmillan Company.
Froot, K. A. (1995). Hedging portfolios with real assets. Journal of Portfolio Management, 21(4), 60–77.
Ganesan, S., & Chiang, Y. (1998). The inflation-hedging characteristics of real and financial assets in Hong Kong. Journal of Real
Estate Portfolio Management, 4(1), 55–67.
Garbade, K., & Wachtel, P. (1978). Time variation in the relationship between inflation and interest rates. Journal of Monetary
Economics, 4(4), 755–765.
Garcia, J. A., & Van Rixtel, A. A. (2007). Inflation-linked bonds from a central bank perspective. ECB Occasional Paper No. 62.
Garcia, R., & Perron, P. (1996). An analysis of the real interest rate under regime shifts. Review of Economics and Statistics, 78(1),
111–125.
Geltner, D., MacGregor, B. D., & Schwann, G. M. (2003). Appraisal smoothing and price discovery in real estate markets. Urban
Studies, 40(5-6), 1047–1064.
Geske, R., & Roll, R. (1983). The fiscal and monetary linkage between stock returns and inflation. Journal of Finance, 38(1), 1–33.
Ghosh, D., Levin, E. J., MacMillan, P., & Wright, R. E. (2004). Gold as an inflation hedge? Studies in Economics and Finance, 22(1),
1–25.
Glascock, J. L., Lu, C., & So, R. W. (2000). Further evidence on the integration of REIT, bond, and stock returns. Journal of Real
Estate Finance and Economics, 20(2), 177–194.
Glascock, J. L., Lu, C., & So, R. W. (2002). REIT returns and inflation: Perverse or reverse causality effects? Journal of Real Estate
Finance and Economics, 24(3), 301–317.
Greene, W. H. (2008). Econometric analysis (6th ed.). Upper Saddle River: Person Education.
Granger, C. W. (1983). Co-integrated variables and error correction models. University of California Discussion Paper No. 83–113.
Granger, C. W., Hyung, N., & Jeon, Y. (2001). Spurious regressions with stationary series. Applied Economics, 33(7), 899–904.
Gultekin, N. B. (1983a). Stock market returns and inflation forecasts. Journal of Finance, 38(3), 663–673.
Gultekin, N. B. (1983b). Stock market returns and inflation: Evidence from other countries. Journal of Finance, 38(1), 49–65.
Hardin, W. G., III, Jiang, X., & Wu, Z. (2012). REIT stock prices with inflation hedging and illusion. Journal of Real Estate Finance
and Economics, 45(1), 262–287.
Hartzell, D., Hekman, J. S., & Miles, M. E. (1987). Real estate returns and inflation. Real Estate Economics, 15(1), 617–637.
Haug, A. A., Beyer, A., & Dewald, W. (2011). Structural breaks and the Fisher effect. B.E. Journal of Macroeconomics, 11(1), 1–31.
Haug, A. A. (2014). On real interest rate persistence: The role of breaks. Applied Economics, 46(10), 1058–1066.
Hautcoeur, P., & Jastram, R. W. (2010). The golden constant. Journal of Economics, 100(2), 189–190.
Hendry, D. F. (1986). Econometric modelling with cointegrated variables: An overview. Oxford Bulletin of Economics and Statistics,
48(3), 201–212.
212
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Hoesli, M., Lizieri, C., & MacGregor, B. (2008). The inflation hedging characteristics of US and UK investments: A multi-factor
error correction approach. Journal of Real Estate Finance and Economics, 36(2), 183–206.
Hoevenaars, R. P., Molenaar, R. D., Schotman, P. C., & Steenkamp, T. B. (2008). Strategic asset allocation with liabilities: Beyond
stocks and bonds. Journal of Economic Dynamics and Control, 32(9), 2939–2970.
Huang, H., & Hudson-Wilson, S. (2007). Private commercial real estate equity returns and inflation. Journal of Portfolio Management, 33(5), 63–73.
Jaffe, J. F. (1989). Gold and gold stocks as investments for institutional portfolios. Financial Analysts Journal, 45(2), 53–59.
Jaffe, J. F., & Mandelker, G. (1976). The Fisher effect for risky assets: An empirical investigation. Journal of Finance, 31(2), 447–458.
Jastram, R. W. (2009). The golden constant: The English and American experience 1560–2007. Cheltenham: Edward Elgar Publishing.
Kaul, G. (1987). Stock returns and inflation: The role of the monetary sector. Journal of Financial Economics, 18(2), 253–276.
Kim, J. H., & Ryoo, H. H. (2011). Common stocks as a hedge against inflation: Evidence from century-long US data. Economics
Letters, 113(2), 168–171.
Knif, J., Kolari, J., & Pynnönen, S. (2008). Stock market reaction to good and bad inflation news. Journal of Financial Research,
31(2), 141–166.
Kolari, J. W., & Anari, A. (2001). Stock prices and inflation. Journal of Financial Research, 24(4), 587–602.
Kothari, S., & Shanken, J. (2004). Asset allocation with inflation-protected bonds. Financial Analysts Journal, 60(1), 54–70.
Lai, K. S. (1997). Long-term persistence in the real interest rate: Some evidence of a fractional unit root. International Journal of
Finance and Economics, 2(3), 225–235.
Lai, K. S. (2004). On structural shifts and stationarity of the ex ante real interest rate. International Review of Economics and
Finance, 13(2), 217–228.
Lai, K. S. (2008). The puzzling unit root in the real interest rate and its inconsistency with intertemporal consumption behavior.
Journal of International Money and Finance, 27(1), 140–155.
Lee, M., & Lee, M. (2012). Long-run inflation-hedging properties of United State equity real estate investment trusts (REITs):
Before and after the structural break in the 1990. African Journal of Business Management, 6(6), 2162–2168.
Le Moigne, C., & Viveiros, É. (2008). Private real estate as an inflation hedge: An updated look with a global perspective. Journal
of Real Estate Portfolio Management, 14(4), 263–286.
Levin, E. J., Montagnoli, A., & Wright, R. E. (2006). Short-run and long-run determinants of the price of gold. World Gold Council
Research Study No. 32.
Lintner, J. (1975). Inflation and security returns. Journal of Finance, 30(2), 259–280.
Lothian, J. R., & McCarthy, C. H. (2001). Equity returns and inflation: The puzzlingly long lags. Research in Finance and Banking,
2, 149–166.
Lothian, J. R., & Simaan, Y. (1998). International financial relations under the current float: Evidence from panel data. Open
Economies Review, 9(4), 293–313.
Lucey, B. M. (2011). What do academics think they know about gold. Alchemist, 62, 12–14.
Lucey, B. M., & Li, S. (2015). What precious metals act as safe havens, and when? Some US evidence. Applied Economics Letters,
22(1), 35–45.
Luintel, K. B., & Paudyal, K. (2006). Are common stocks a hedge against inflation? Journal of Financial Research, 29(1), 1–19.
Lütkepohl, H., & Saikkonen, P. (1999). Order selection in testing for the cointegrating rank of a VAR process. In R. F. Engle, &
H. White (Eds.), Cointegration, causality, and forecasting. A Festschrift in Honor of Clive W.J. Granger (pp. 168–199). Oxford:
Oxford University Press.
Madsen, J. B. (2007). Pitfalls in estimates of the relationship between stock returns and inflation. Empirical Economics, 33(1),
1–21.
Mahdavi, S., & Zhou, S. (1997). Gold and commodity prices as leading indicators of inflation: Tests of long-run relationship and
predictive performance. Journal of Economics and Business, 49(5), 475–489.
Maki, D. (2003). Nonparametric cointegration analysis of the nominal interest rate and expected inflation rate. Economics Letters,
81(3), 349–354.
Matysiak, G., Hoesli, M., MacGregor, B., & Nanthakumaran, N. (1996). The long-term inflation-hedging characteristics of UK
commercial property. Journal of Property Finance, 7(1), 50–61.
McCown, J. R., & Zimmerman, J. R. (2006). Is gold a zero-beta asset? Analysis of the investment potential of precious metals. Oklahoma
City University Working Paper.
McCown, J. R., & Zimmerman, J. R. (2007). Analysis of the investment potential and inflation-hedging ability of precious metals.
Oklahoma City University Working Paper.
Million, N. (2004). Central bank’s interventions and the Fisher hypothesis: A threshold cointegration investigation. Economic
Modelling, 21(6), 1051–1064.
Mishkin, F. S. (1984). The real interest rate: A multi-country empirical study. Canadian Journal of Economics, 17(2), 283–311.
Mishkin, F. S. (1992). Is the Fisher effect for real? A reexamination of the relationship between inflation and interest rates.
Journal of Monetary Economics, 30(2), 195–215.
Modigliani, F., & Cohn, R. A. (1979). Inflation, rational valuation and the market. Financial Analysts Journal, 35(2), 24–44.
Mundell, R. (1963). Inflation and real interest. Journal of Political Economy, 71(3), 280–283.
Neely, C. J., & Rapach, D. E. (2008). Real interest rate persistence: Evidence and implications. Federal Reserve Bank of St. Louis
Review, 90(6), 609–641.
Nelson, C. R., & Schwert, G. W. (1977). Short-term interest rates as predictors of inflation: On testing the hypothesis that the
real rate of interest is constant. American Economic Review, 67(3), 478–486.
Nelson, C. R. (1976). Inflation and rates of return on common stocks. Journal of Finance, 31(2), 471–483.
O’Connor, F. A., Lucey, B. M., Batten, J. A., & Baur, D. G. (2015). The financial economics of gold – A survey. International Review
of Financial Analysis (in press).
Obereiner, D., & Kurzrock, B. (2012). Inflation-hedging properties of indirect real estate investments in Germany. Journal of
Property Investment and Finance, 30(3), 218–240.
Park, Y. W., & Bang, D. W. (2012). Direct commercial real estate as an inflation hedge: Korean evidence. Journal of Real Estate
Portfolio Management, 18(2), 187–203.
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
213
Parkin, M. (2008). Inflation. In S. N. Durlauf, & L. E. Blume (Eds.), The New Palgrave Dictionary of Economics. Basingstoke: Palgrave
Macmillan.
Perron, P., & Ng, S. (1996). Useful modifications to some unit root tests with dependent errors and their local asymptotic
properties. Review of Economic Studies, 63(3), 435–463.
Pullen, T., Benson, K., & Faff, R. (2014). A comparative analysis of the investment characteristics of alternative gold assets. Abacus,
50(1), 76–92.
Rapach, D. E., & Wohar, M. E. (2005). Regime changes in international real interest rates: Are they a monetary phenomenon?
Journal of Money, Credit and Banking, 37(5), 887–906.
Roll, R. (1972). Interest rates on monetary assets and commodity price index changes. Journal of Finance, 27(2), 251–277.
Rose, A. K. (1988). Is the real interest rate stable? Journal of Finance, 43(5), 1095–1112.
Rubbaniy, G., Lee, K. T., & Verschoor, W. F. (2011). Metal investments: Distrust killer or inflation hedging? 24th Australasian Finance
and Banking Conference Paper.
Rödel, M. G. (2012). Inflation hedging. An empirical analysis on inflation nonlinearities, infrastructure, and international equities
(PhD Thesis). Technical University of Munich.
Rödel, M. G. (2014). Inflation hedging with international equities. Journal of Portfolio Management, 40(2), 41–53.
Rödel, M. G., & Rothballer, C. (2012). Infrastructure as hedge against inflation – Fact or fantasy? Journal of Alternative Investments,
15, 110–123.
Saikkonen, P., & Lütkepohl, H. (2000a). Testing for the cointegrating rank of a VAR process with an intercept. Econometric Theory,
16(3), 373–406.
Saikkonen, P., & Lütkepohl, H. (2000b). Testing for the cointegrating rank of a VAR process with structural shifts. Journal of
Business and Economic Statistics, 18(4), 451–464.
Saikkonen, P., & Lütkepohl, H. (2000c). Trend adjustment prior to testing for the cointegrating rank of a vector autoregressive
process. Journal of Time Series Analysis, 21, 435–456.
Schotman, P. C., & Schweitzer, M. (2000). Horizon sensitivity of the inflation hedge of stocks. Journal of Empirical Finance, 7(3),
301–315.
Schwert, G. W. (1987). Effects of model specification on tests for unit roots in macroeconomic data. Journal of Monetary Economics,
20(1), 73–103.
Schwert, G. W. (1989). Tests for unit roots: A Monte Carlo investigation. Journal of Business & Economic Statistics, 7(2), 147–159.
Shafiee, S., & Topal, E. (2010). An overview of global gold market and gold price forecasting. Resources Policy, 35(3), 178–189.
Simpson, M. W., Ramchander, S., & Webb, J. R. (2007). The asymmetric response of equity REIT returns to inflation. Journal of
Real Estate Finance and Economics, 34(4), 513–529.
Smirlock, M. (1986). Inflation announcements and financial market reaction: Evidence from the long-term bond market. Review
of Economics and Statistics, 68(2), 329–333.
Solnik, B., & Solnik, V. (1997). A multi-country test of the Fisher model for stock returns. Journal of International Financial Markets,
Institutions and Money, 7(4), 289–301.
Spierdijk, L., & Umar, Z. (2014). Are commodity futures a good hedge against inflation? Journal of Investment Strategies, 3(2),
35–57.
Spyrou, S. I. (2004). Are stocks a good hedge against inflation? Evidence from emerging markets. Applied Economics, 36(1), 41–48.
Stevenson, S. (1999). The performance and inflation hedging ability of regional housing markets. Journal of Property Investment
and Finance, 17(3), 239–260.
Stevenson, S., & Murray, L. (1999). An examination of the inflation hedging ability of Irish real estate. Journal of Real Estate
Portfolio Management, 5(1), 59–69.
Stock, J. H., & Watson, M. W. (1999). Forecasting inflation. Journal of Monetary Economics, 44(2), 293–335.
Stock, J. H., & Watson, M. W. (2007). Why has US inflation become harder to forecast? Journal of Money, Credit and Banking,
39(1), 3–33.
Summers, L. H. (1983). In J. Tobin (Ed.), The non-adjustment of nominal interest rates: A study of the Fisher effect (2nd ed., pp.
201–241). Washington, DC: Symposium in honor of Arthur Okun, Brookings Institution.
Sun, Y., & Phillips, P. C. (2004). Understanding the Fisher equation. Journal of Applied Econometrics, 19(7), 869–886.
Swinkels, L. (2012). Emerging market inflation-linked bonds. Financial Analysts Journal, 68(5), 38–56.
Tarbert, H. (1996). Is commercial property a hedge against inflation? A cointegration approach. Journal of Property Finance, 7(1),
77–98.
Taylor, J. B. (1993). Discretion versus policy rules in practice. Carnegie-Rochester Conference Series on Public Policy, 39, 195–214.
Taylor, N. J. (1998). Precious metals and inflation. Applied Financial Economics, 8(2), 201–210.
Tobin, J. (1965). Money and economic growth. Econometrica, 33(4), 671–684.
Tsay, R. S. (2005). Analysis of financial time series. Hoboken: John Wiley & Sons.
Tsong, C., & Lee, C. (2013). Quantile cointegration analysis of the Fisher hypothesis. Journal of Macroeconomics, 35, 186–198.
Tully, E., & Lucey, B. M. (2007). A power GARCH examination of the gold market. Research in International Business and Finance,
21(2), 316–325.
Urich, T., & Wachtel, P. (1984). The effects of inflation and money supply announcements on interest rates. Journal of Finance,
39(4), 1177–1188.
Wang, K., Lee, Y., & Thi, T. N. (2011). Time and place where gold acts as an inflation hedge: An application of long-run and
short-run threshold model. Economic Modelling, 28(3), 806–819.
Wilcox, S. E. (2012). Equity valuation and inflation: A review. Research Foundation Literature Reviews, 7(1), 1–23.
Wilson, P. J., & Zurbruegg, R. (2003). International diversification of real estate assets: Is it worth it? Evidence from the literature.
Journal of Real Estate Literature, 11(3), 257–278.
Worthington, A. C., & Pahlavani, M. (2007). Gold investment as an inflationary hedge: Cointegration evidence with allowance
for endogenous structural breaks. Applied Financial Economics Letters, 3(4), 259–262.
Yobaccio, E., Rubens, J. H., & Ketcham, D. C. (1995). The inflation-hedging properties of risk assets: The case of REITs. Journal of
Real Estate Research, 10(3), 279–296.
Yoon, G. (2010). Does nonlinearity help resolve the Fisher effect puzzle? Applied Economics Letters, 17(8), 823–828.
214
S. Arnold, B.R. Auer / North American Journal of Economics and Finance 34 (2015) 187–214
Zhou, X., & Clements, S. (2010). The inflation hedging ability of real estate in China. Journal of Real Estate Portfolio Management,
16(3), 267–277.
Zhou, Z. (2014). Commodity futures as a hedge against inflation: An application of Markov-switching cointegration and errorcorrection modeling. Bocconi University Working Paper.
Zivot, E., & Andrews, D. W. K. (2002). Further evidence on the great crash, the oil-price shock, and the unit-root hypothesis.
Journal of Business & Economic Statistics, 20(1), 25–44.