ArticlePDF Available

Abstract and Figures

We analyze the long-term relationship between the fine wine market and stock markets, from world level to specific countries, by applying a cointegration approach on a 21-year monthly database. Our database includes the Liv-ex Fine Wine Investables Index, 6 Liv-ex sub-indices, 25 MSCI Indexes and 10 national stock market indexes. Results suggest significant cointegration between emerging markets, especially Asia, and fine wine markets, and also causality from the former to the latter. As a result, the slowdown of economic growth in emerging countries can be a risk to the fine wine market. Furthermore, Mainland China appears to be one of the main drivers of fine wine markets, and more precisely, the leading Bordeaux wines. We suggest that along with single-index analysis, our approach can be taken into consideration in fine wine price forecasting and portfolio diversification strategy.
Content may be subject to copyright.
1
The long-term financial drivers of fine wine prices:
The role of emerging markets
Version 2016
Jean-Marie Cardebat
University of Bordeaux (Larefi) & INSEEC
Bordeaux Wine Economics
jean-marie.cardebat@u-bordeaux.fr
Linda Jiao
University of Bordeaux (Larefi)
and Bordeaux Wine Economics
linda.jiao@u-bordeaux.fr
Abstract
We analyze the long-term relationship between the fine wine market and stock markets, from
world level to specific countries, by applying a cointegration approach on a 21-year monthly
database. Our database includes the Liv-ex Fine Wine Investables Index, 6 Liv-ex sub-indices, 25
MSCI Indexes and 10 national stock market indexes. Results suggest significant cointegration
between emerging markets, especially Asia, and fine wine markets, and also causality from the
former to the latter. As a result, the slowdown of economic growth in emerging countries can be
a risk to the fine wine market. Furthermore, Mainland China appears to be one of the main
drivers of fine wine markets, and more precisely, the leading Bordeaux wines. We suggest that
along with single-index analysis, our approach can be taken into consideration in fine wine price
forecasting and portfolio diversification strategy.
Key words: Wine, cointegration, financial markets, emerging markets
2
1. Introduction
Between 1900 and 2012, the net annualized return on wine investments (net of insurance and
storage costs) has been estimated by Dimson et al. (2015) at 4.1%. In the last decade, this growth
rate has accelerated while the price of fine wines has experienced a dramatic rise. The major fine
wine price indexes
1
tripled over the period 2001-2011. This increase has been largely commented
upon by the press and analyzed by the academic literature which has attempted to explain the
price of wine and to understand what the drivers are.
Due to the dual nature of fine wines, two approaches have been adopted to answer this question.
The first one concentrated on the highly-differentiated product nature of wines. Many papers
then focused on a hedonic analysis to explain the price evolution of fine wines (for a meta-
analysis of hedonic methodology applied to wine, see Oczkowski and Doucouliagos, 2014).
Among other determinants such as objective quality (for oenological comments: see for example
Combris et al., 1997 and 2000; Cardebat and Figuet 2004 and 2009), these papers show the
crucial role of the brand name. Reputation appears as one of the main determinants of wine
prices, especially in China where the name of Bordeaux and more precisely of the first growth
wines of the Bordeaux 1855 classification was a synonym of exorbitant prices until 2011. The
Chinese demand for Bordeaux red wines is obviously a major driver of their prices but it is
difficult to assess with a hedonic approach (Masset et al. 2015).
A second strand of the literature concentrates on fine wines regarded as an alternative financial
asset
2
. Since the seminal papers of Krasker (1979) and Jaeger (1981), many studies have tried to
assess the return on fine wines and to check whether they can be considered as a good tool for
portfolio diversification
3
. A large proportion of these papers, using a CAPM approach, found
higher returns for fine wines compared to stocks or bonds and a low correlation between fine
wine prices and financial markets (see notably Sanning et al. 2008, Fogarty, 2010, Masset and
Weisskopf, 2010).
However, some recent papers reveal some ambiguity as regards the correlation between financial
markets and fine wine markets. Masset et al. (2015) have shown a stronger correlation and
1
The Liv-ex company, created in 1999, publishes several wine price indexes presented as the reference on the fine
wine markets. Their fine wine indexes are built on the basis of transaction prices recorded on their site. See
https://www.liv-ex.com/home.do for a complete description of their methodology, including the wine composition
of each index. These indexes are extensively used in the literature, see for example Masset and Henderson (2010),
Kourtis et al. (2012), Bouri (2013), Cevik and Sedik (2014).
2
Notably because such a price increase has attracted investors seeking portfolio diversification. This trend widely
contributes to a “financialization” of the fine wines market (Cardebat et al. 2016).
3
The market for fine wines is an over-the-counter market. The return trajectories of fine wines and financial assets
would be different and would obey different determinants.
3
indicate that this correlation depends on the frequency of the fine wine price data. A lower
frequency permits the transmission effect from financial markets to the fine wine market and
then exhibits a higher correlation. Faye et al (2015) showed a causality relationship between fine
wine prices and the MSCI world index, while Bouri (2013) presented a MTGARCH analysis and
revealed a transmission of the crude oil price on fine wine prices.
Consequently, if the intrinsic characteristics of wines might explain their price in level
4
, it seems
that the price variation of fine wines might be influenced by macroeconomic determinants
(Chinese demand, etc.) and financial variables (stocks markets, etc.). In fact, several theoretical
channels might justify some links between financial markets and the fine wine market, such as
wealth or cash effect. Common macroeconomic determinants might also help to explain a
positive correlation between wines and financial asset prices, especially during the 2008 crisis.
However, most of these links play in the medium / long term.
Few papers have been devoted to the discovery of the macroeconomic and/or financial
determinants of wine prices
5
. Our paper aims to fill this gap in the literature and to determine
whether stock markets might be considered as a major driver of fine wine prices. A particular
focus will be placed on the geographical aspect of the links in order to verify the specific effect
of China compared to other countries. We then apply a cointegration and causality analysis on a
long-period (1994-2015) monthly database from world level (in line with Faye et al., 2015) to
specific countries.
Our results suggest significant cointegration between emerging markets, especially Asia, and fine
wine markets, and also causality from the former to the latter. China appears as one of the main
drivers of fine wine prices. Due to the illiquid nature of wine (Masset et al. 2015), the link
between the stock market and the fine wine market is not contemporaneous. Wine prices adjust
with a lag. These linkages should be an important aspect to be taken into consideration in
portfolio diversification strategies. This approach can then provide a more accurate anticipation
of wine price fluctuation.
The remainder of the paper is constructed as follows: section 2 provides an overview of existing
literature on wine as a financial asset; section 3 presents the database used in further analysis and
section 4 outlines the methodology and procedures of our approaches; section 5 describes the
results and gives their interpretations and a robustness check; lastly, we conclude in section 6.
4
As established by the abundant hedonic literature applied to wine. Hedonic equations allow for the explanation of
the price differences between different wines but cannot really explain the changes over time in the price of wines.
5
The notable exceptions are Cevik and Sedik (2014), Faye et al (2015) and Bouri (2013).
4
2. Wine as a financial asset
In recent years, fine wines have often appeared in wine economic literature as an alternative
financial asset. By using Liv-ex indexes and wine auction prices (Chicago wine company,
Langton’s auction house, Christie’s, WinePrice.com), several studies estimate the returns, the
performance and the portfolio diversification benefits of wine investment. Various methods are
applied in this literature but repeat sales and hedonic price regression are the most common
approaches to calculate wine returns; the Sharpe ratio, the CAPM model and the Fama-French
Three-Factor Model estimate the performances; the mean-variance efficient frontier is applied to
testify the portfolio diversification. A chronological summary of existing literature is presented in
Annex 1.
Krasker (1979) first estimates the return on storing wines, and concludes that the return on wine
is not higher than the return on riskless assets such as government bonds. Jaeger (1981) reverses
Krasker’s findings by expanding the sample period. She finds that wine impressively
outperformed Treasury bills. When storage, insurance and transaction costs are taken into
account, wine assets may become less attractive. However, several recent studies reveal that the
return on wine is still higher than government bonds and other collectable assets such as art or
stamps, but lower than equities (Burton and Jacobsen, 2001; Fogarty, 2006, 2010; Dimson et al.,
2015). Yet, Masset and Henderson (2010) find that the cumulated return in the wine index is
above the Dow Jones Industrial Average. Lucey and Devine (2015) obtain similar results. Later,
Aytac et al. (2015) indicate that French wines outperform equities for a more recent and shorter
time period (2007-2014). In addition, it is worth noting that outstanding wines perform better
than more basic wines. The best wines achieve higher returns and have a lower volatility than
more basic wines (Fogarty, 2006; Masset and Henderson, 2010). Lucey and Devine (2015) point
out that the performance of wine is attractive, and compared to other collectibles, wine has a
great advantage in trading volume and frequency. But they advise individual investors to invest
indirectly through wine funds or traded indexes since the individual and sub-regional level
returns on wine are highly heterogeneous.
Meanwhile, researchers have found that wine assets can provide portfolio diversification
benefits. Sanning et al. (2008) find that investment-grade wine assets provide excess risk-adjusted
returns and benefits “from low exposure to market risk factors”, thus offering portfolio
diversification benefits. Fogarty (2010) points out that including wine in a portfolio, even an
5
already well-diversified one, maximizes the Sharpe ratio, thus providing a diversification benefit
by achieving the same level of portfolio return with lower risk. Masset and Henderson (2010)
indicate that the addition of wine assets can improve the performance of a portfolio in terms of
skewness, but also increase the kurtosis. They also find a slight correlation between the wine
returns and other assets, but during the financial crisis, the return on wine fluctuated
considerably, and the correlation between wine and other financial assets significantly increased.
However, wine returns declined less and have a lower volatility than other assets. Aytac et al.
(2015) focus on French markets. They analyze the performance of fine wine in French portfolio
diversification. Results show that Bordeaux fine wines are the most efficient for improving the
performance of French portfolios compared to other French fine wines. Besides, they suggest
risk-averse investors include wine in their portfolios. Kourtis et al. (2012) demonstrate the
potential of international diversification benefits of investing in wines from a wide range of
origins such as Italy, Australia and Portugal, since the correlation coefficients with these wines
are relatively low, while the diversification benefits of investing only in French wines are limited.
Fogarty and Jones (2011) and Fogarty (2014) review different approaches in return estimations
and diversification benefit tests. He points out that the conclusion about diversification benefits
offered by wine is sensitive to the return-estimation method, the sample period, whether raw
return or excess return is used, and the type of diversification benefit test. In conclusion, the
repeat-sales approach can overstate the return. As for the diversification benefit test, the mean-
variance efficient frontier approach is preferred. Subsequently, Fogarty (2014) applies a pooled
model to improve the accuracy of return in the estimation. His results indicate that in the case of
Australian fine wines, including wines in a portfolio can provide diversification benefits, but the
benefits are small.
The above studies generally find a negative or limited correlation between fine wines and other
financial asset returns. By contrast, Dimson et al. (2015) find a significant positive correlation
between wine and equities. This contradiction may be explained as following: their estimation is
based on an annualized wine return calculated from wine auction prices between 1900 and 2012,
which differs from other recent studies which use monthly returns with a relatively shorter time
period. This explanation is then supported by Masset et al. (2015). They compare the results
obtained from monthly and quarterly data samples, and find that for a classical CAPM, the beta
go higher when the frequency of data sample is lower.
In addition, the methodology of estimation may also have an impact on the results of correlation
between fine wines and other financial assets. Other methodologies like cointegration and
6
causality analysis can be experimented to explore the long-term relationship and short-term
dynamics between these two markets. Moreover, with the financialization in the fine wine
market, the fine wine market receives the impacts from financial markets via the channels such
as wealth or cash effect, and the transmission of volatilities. This aspect should be taken into
account in portfolio diversification strategies when including wine assets in a portfolio. However,
very little literature is devoted to discovering this linkage between financial markets and the fine
wine market:
- With commodity markets: Bouri (2013) applies the Johansen cointegration test, the
Granger causality test and a multivariate Garch model to examine the price dynamic and
the long-run equilibrium between the fine wines market and the crude oil market. Results
indicate that a mean transmission exists from the crude oil market to the fine wine
market, but there was no long-term equilibrium relationship between the two markets.
Cevik and Sedik (2010) point out that under influences of common macroeconomic
factors, fine wines seem not to behave differently from commodities. Above evidence
shows that the potential of portfolio diversification may be reduced when we include fine
wine and commodities in the same portfolio.
- With stock markets: Faye et al. (2015) use auction prices to analyze the dynamic and the
long-term relationship among fine wines themselves, by applying cointegration and
Granger causality procedures. They also include a global equity index the MSCI World
Index in their analysis. Results indicate short-term and long-term causalities among
wines themselves, and short-term causalities between several fine wines and the MSCI
World Index.
In the following sections, we also propose a cointegration approach with causality tests as well as
a robustness check. But, we offer a richer analysis with a panoramic view of the connections
between the fine wine market and stock markets - from global level to specific countries, so that
we are able to determine the major influencers of the fine wine market coming from stock
markets.
3. Data
We have a database of 21 years in monthly frequency. It covers the period from September 1994
to February 2015 and consists of the Liv-ex Fine Wine Investables Index. As indicated on the
liv-ex website: “The Liv-ex Fine Wine Investables […] aims to mirror the performance of a
typical wine investment portfolio. The Liv-ex Fine Wine Investables Index is a dominant price
7
index of investment-grade wines. It contains Bordeaux red wines, around 200 wines from 24
leading chateaux, from the 1982 vintage and chosen on the basis of their Robert Parker rating
scores
6
. This index is calculated monthly using the method of Liv-ex Mid Price the mid-point
between the current highest bid price and lowest offer price for each component wine, and
weighted according to their coefficients of scarcity
7
. In addition, we include six regional sub-
indices, which will be discussed in our robustness check section.
For financial indexes, we use a set of 25 MSCI indexes. These indexes have the interest of
providing a “fully integrated view across all sources of equity returns in 46 developed and
emerging markets”
8
and are commonly used by professionals and academics. These 25 MSCI
indexes (large and mid-cap) are chosen from three categories All Country (including developed
markets and emerging markets), Developed Markets or Emerging Markets and specific countries
(see MSCI index categories and corresponding components in Annex 2):
- 4 MSCI All Country (AC) indexes: ACWI (All Country World Index), AC Asia Pacific, AC
Americas, and AC Europe
9
;
- 6 MSCI Developed Markets (DM) indexes: Developed Markets (World), G7, Pacific, EAFE
(Europe, Australasia and Far East), DM EU (European Union), and North America;
- 3 MSCI Emerging Markets (EM) indexes: Emerging Markets (World), BRIC (Brazil, Russia,
India and China), and EM Asia;
- 3 MSCI China Markets indexes: Zhong Hua (China and Hong Kong), China, and Hong
Kong
10
;
- 6 MSCI DM Countries indexes: USA, UK, Germany, France, Japan, and Australia;
- 3 MSCI EM Countries indexes: Brazil, Russia, and India.
We have also gathered 10 stock markets indexes that will be used in our robustness analysis to
control for the impact of stocks indexes series. These 10 stock market indexes include the most
influential developed markets and emerging markets: Dow Jones Industrial Average, Nasdaq
6
To qualify for the index, the score must be at least 95/100 points. For the eight first brands of Bordeaux Lafite
Rothschild, Latour, Margaux, Haut-Brion, Mouton-Rothschild, Ausone, Cheval Blanc and Petrus, a score of at least
93/100 points is required.
7
The coefficient of scarcity is applied to the vintages older than 15 years and the wines produced in small quantities
(Ausone and Petrus). Source Liv-ex
8
https://www.msci.com/indexes for further details.
9
MSCI AC Asia Pacific Index contains 5 Developed Markets countries and 8 Emerging Markets countries; AC
America Index contains 2 Developed Markets countries and 5 Emerging Markets countries. AC Europe Index
contains 15 Developed Markets countries and 6 Emerging Markets countries. Source MSCI
10
According to MSCI, Chinese markets include both developed markets (Hong Kong) and emerging markets
(Mainland China). Regarding this specialty, MSCI creates MSCI China Markets that include both.
8
Composite, FTSE100, DAX, CAC40, Nikkei225, Hang Seng, BVSP (Brazil), RTS (Russia) and
BSE SENSEX (India).
Figure 1: Liv-ex Fine Wine Investables Vs. selected MSCI indexes in real Dollar 2010=100
Figure 1 displays the index evolution of the liv-ex Fine Wine Investables and the MSCI for
developed and emerging markets during the studied period. The ten first years do not show clear
links between the curves. The emerging markets have been affected by the Asian crisis in 1997
whereas developed markets drop with the explosion of the Internet bubble in 2000. By contrast,
during the sub-period 2003-2012, the rises and falls (due to the financial crisis) are common to
the three indexes, and the curves exhibit similar movements. This sub-period suggests that wine
prices would not be independent of macroeconomic and financial evolutions. In particular, the
fine wine market has experienced a significant decrease with the financial crisis. To go further in
analyzing the links between stocks indexes and the liv-ex index, we now turn to the cointegration
analysis.
4. Methodology
Unit Root Tests
Before proceeding to further analysis of temporal relations between the fine wine price and stock
market prices, it is relevant to run unit root tests for all price series. Augmented Dickey-Fuller
(ADF) unit root tests are applied to test the stationarity of all variables. The lag length of ADF
tests is selected automatically by information criteria.
0
20
40
60
80
100
120
140
160
180
200
1994M09
1995M03
1995M09
1996M03
1996M09
1997M03
1997M09
1998M03
1998M09
1999M03
1999M09
2000M03
2000M09
2001M03
2001M09
2002M03
2002M09
2003M03
2003M09
2004M03
2004M09
2005M03
2005M09
2006M03
2006M09
2007M03
2007M09
2008M03
2008M09
2009M03
2009M09
2010M03
2010M09
2011M03
2011M09
2012M03
2012M09
2013M03
2013M09
2014M03
2014M09
MSCI Developed Markets MSCI Emerging Markets Liv-ex Fine Wine Investables
Liv-ex Fine Wine Investables Vs. MSCI Developed Markets & MSCI Emerging Markets
9
In addition, all the unit root tests are conducted step by step for all variables, in level and then in
first difference, to see if there are a deterministic trend and a constant in a time series. It will help
us to specify the cointegration test models in the next step. Also, we can determine whether a
time series has a trend through its plot. The procedure of unit root tests is presented as following
(from equation (1) to (3)):
     
 (1)
    
 (2)
  
 (3)
Cointegration Tests
Only non-stationary series and being integrated of the same order could be used in cointegration
tests. For conducting the cointegration analysis, we propose a vector error correction model
(VECM). This model is more flexible and it allows us to make a bilateral analysis of the
relationship between fine wine prices and stock market prices.
In our case, where we test the cointegration relation between the fine wine price and each stock
market index respectively, we consider a two-variable system with one cointegrating equation.
Therefore, the general (the most restricted) VECM is:
        


        


where is the price series in logarithm,  is a vector of deterministic term (trend or constant)
in VAR,    is the cointegration equation (CE) which represents
the long-run relationship between the two variables,  is a deterministic term in CE,


 are the lagged difference variables which represent the short-run dynamic, , are
two uncorrelated white-noise series.
10
The specification of a VECM depends on the specificity of each price series (Johansen, 1990)
11
.
Choosing the lag length is an important step for cointegration tests. So firstly, we construct a
two-variable VAR, and the optimal lag length of the VAR is selected by standard information
criteria. Additionally, we apply an autocorrelation LM test to confirm that the residuals are not
auto-correlated under the selected lagged structure.
Then, we apply Johansen’s cointegration test to determine the number of cointegrating relations.
Johansen’s cointegration tests consist of two tests based on a maximum likelihood ratio trace
test and maximum eigenvalues test:
   

  

where T is the sample size,
estimates the -th largest eigenvalue. Both tests examine the rank
of the cointegrating matrix, in other words, the number of cointegrating vectors. The null
hypothesis of the Trace test is that there are or fewer cointegrating vectors against the
alternative that there are more than r cointegrating vectors, while maximum eigenvalue statistic
tests the null hypothesis of r cointegration vectors against the alternative of +1 cointegration
vectors. In our case, we test if there is one cointegrating relation in our two-variable models ( =
1). The national/regional GDPs are included in the cointegration equations as a control
variable
12
.
Granger Causality
In our case, a two-variable Granger causality model is defined as following:
  


11
We include a linear deterministic trend (constant in CE and constant in VAR) in the VCE models with MSCI EM
World, MSCI EM Asia, MSCI Hong Kong, MSCI Australia, and MSCI India, respectively, because these series have
statistically-significant linear trends in data in level. We do not restrict any linear deterministic trend for the VCE
models with any other price series, as they do not have statistical significant linear trends in data in level. The model
is simplified we only include a constant in CE if it is statistically significant. But in our empirical work, the
different specifications of a VEC model do not change the conclusion of the cointegration tests, anyway.
12
Depends on region or country, the GDP is not always significant in the estimations, and including it or not does
not affect our major conclusions. The results are not reported here but are available on request.
11
  


where , are two uncorrelated white-noise terms. We test the null hypothesis that  does
not Granger cause  (= = … = = 0) and that  does not Granger cause 
(= = = = 0) against the alternative. If not all are zero, we say  cause ,
which means the available information of  helps to predict . Similarly, if not all are
zero,  cause  which means  helps the prediction of . If causation occurs
in both directions, we can say that there is a feedback relationship between  and .
5. Empirical results and interpretation
The logarithms of the real price indexes are used in further econometric estimations
13
. All the
series are integrated of first order I(1), except for the MSCI DM Pacific Index and Dow Jones
Industrial Average, which are stationary in level at 10% significance. As a result, we exclude
MSCI DM Pacific and Dow Jones Industrial Average from further tests. The table of results of
Augmented Dickey-Fuller unit root tests is presented in Annex 3.
Johansen Cointegration Test Results
We aim to analyze the linkage between the fine wine market and stock markets to determine how
stock markets influence the fine wine market, and where the influences come from (from the
whole world or specific regions or specific countries), so that we can tell which are the main
drivers of the fine wine market. Our procedure of the cointegration analysis is constructed as
following: firstly, we perform a cointegration analysis on All Country level. We test the
cointegration between the Liv-ex Fine Wine Investables Index and each MSCI All Country
index, respectively, in a two-variable VEC model; secondly, we distinguish, between developed
markets (DM) and emerging markets (EM), with an emphasis on the Chinese market (CM). We
test the cointegration between the Liv-ex Fine Wine Investables Index and each MSCI DM, EM,
and CM index, respectively, in a two-variable VEC model; lastly, we narrow the scope to specific
countries. Thus, we test the cointegration between the Liv-ex Fine Wine Investables Index and
each MSCI Countries index and representative stock market price indexes (as complementary
tests), respectively, in a two-variable VEC model. As mentioned before, the optimal lag length is
13
We have converted all prices in national currency to US Dollar, and transformed the prices to price indexes on the
base year of 2010, then deflated the nominal price indexes with the US Consumer Price Index (Source OECD, base
year 2010 = 100) to real term.
12
selected by standard information criteria, and the trend assumptions of the Johansen
cointegration test are specified above
14
.
Table 1: Johansen Cointegration Test result - Liv-ex Fine Wine Investables vs. MSCI All Country
(see more results in annex 4)
** denotes rejection of the hypothesis at the 5% significance level.
Among the four MSCI All Country indexes (see Table 1 and others in Annex 4), only AC Asia
Pacific is cointegrated with the fine wine price (at 5% statistical significance). In other words,
only the Asia Pacific region’s stock price generates a long-term relationship with the fine wine
price. The AC Asia Pacific index is composed of 5 developed markets and 8 emerging markets
of the Asia Pacific region, in which Japan, China and Australia are the three highest-weighted
countries constituting 68.04% of the index
15
. Evidence shows that around the world, the Asia
Pacific region would be the most influential area for the fine wine market during the period
considered, and Japan, China, Australia are the three major contributors.
Figure 2: Liv-ex Fine Wine Investables vs. MSCI AC Asia Pacific in real dollar 2010=100
14
On account of the large size of the results table, only cointegrated relations are presented here, the rest can be
found in Annex ??
15
Country weights of MSCI AC Asia Pacific: Japan 43.64%, China 12.29%, Australia 12.11%, South Korea 8.3%,
Taiwan 6.37% and other countries 17.29%. Source MSCI Nov. 2015
Series: Liv-ex Fine Wine Investables vs. MSCI AC Asia Pacific
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None**
0.0698
23.1061**
17.4434**
At most 1
0.0232
5.6627
5.6627
13
Then we run a cointegration analysis on “Markets” level Developed Markets and Emerging
Markets. The results are represented as following:
Table 2: Johansen Cointegration Test result - Liv-ex Fine Wine Investables vs. MSCI Developed
Markets and Emerging Markets (see more results in annex 4)
Series: Liv-ex Fine Wine Investables vs. MSCI Developed Markets
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0535
16.0121
13.2400
At most 1
0.0114
2.7722
2.7722
Series: Liv-ex Fine Wine Investables vs. MSCI Emerging Markets
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None***
0.1014
28.2651***
25.9678***
At most 1
0.0094
2.2973
2.2973
Series: Liv-ex Fine Wine Investables vs. MSCI BRIC
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None***
0.0756
19.5121***
18.8784***
At most 1
0.0026
0.6338
0.6338
Series: Liv-ex Fine Wine Investables vs. MSCI EM Asia
Lags interval: 1 to 2
Hypothesized
Trace
Max-Eigen
0
20
40
60
80
100
120
140
160
180
1994M09
1995M04
1995M11
1996M06
1997M01
1997M08
1998M03
1998M10
1999M05
1999M12
2000M07
2001M02
2001M09
2002M04
2002M11
2003M06
2004M01
2004M08
2005M03
2005M10
2006M05
2006M12
2007M07
2008M02
2008M09
2009M04
2009M11
2010M06
2011M01
2011M08
2012M03
2012M10
2013M05
2013M12
2014M07
2015M02
Liv-ex Fine Wine Investables Vs. MSCI AC Asia Pacific
MSCI AC Asia Pacific Liv-ex Fine Wine Investables
14
No. of CE(s)
Eigenvalue
Statistic
Statistic
None***
0.0835
23.5966***
21.1804***
At most 1
0.0099
2.4162
2.4162
*** denotes rejection of the hypothesis at the 1% significance level.
None of the MSCI Developed Markets indexes are cointegrated with Liv-ex Fine Wine
Investables. These results can be found in Annex 4. By contrast, all the three Emerging Markets
indexes are highly-significantly cointegrated with the fine wine price. Obviously, emerging
markets win a greater impact on fine wine markets against developed markets. Similar evidence is
shown in Cevic and Sedik (2014) that the increasing demand from emerging markets is a key
factor influencing the fine wine market. Through Figure 3, we can observe co-movement trends
between the fine wine price and emerging market stock prices under macroeconomic shocks
since the beginning of 2000. Besides, China is the highest weighted country in all the three
indexes with 24.01% in MSCI Emerging Markets
16
, 34.11% in MSCI EM Asia
17
and 56.88% in
BRIC
18
. Thereafter, we place emphasis on Chinese markets regarding the high weight they hold
in emerging markets. The following table shows the results:
Figure 3: Liv-ex Fine Wine Investables vs. MSCI Emerging Markets in real Dollar 2010=100
16
Country weights of MSCI Emerging Markets Index: China 24.01%, South Korea 16.21%, Taiwan 12.44%, India
8.35%, South Africa 7.68%, and other countries 31.32%. Source MSCI Nov. 2015
17
Country weights of MSCI EM Asia Index: China 34.11%, South Korea 23.03%, Taiwan 17.67%, India 11.87%,
Malaysia 4.63%, and other countries 8.68%. Source MSCI Nov.2015
18
Country weights of MSCI BRIC Index: China 56.88%, India 19.8%, Brazil 14.07%, and Russia 9.25%. Source
MSCI Nov. 2015
15
Table 3: Johansen Cointegration Test result - Liv-ex Fine Wine Investables Vs. MSCI Chinese
Markets
Series: Liv-ex Fine Wine Investables vs. MSCI Zhong Hua (China & HK)
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None**
0.0661
20.2534***
16.6186**
At most 1
0.01485
3.6348
3.6348
Series: Liv-ex Fine Wine Investables vs. MSCI Hong Kong
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None*
0.0629
21.7613**
15.7748*
At most 1
0.0243
5.9865
5.9865
Series: Liv-ex Fine Wine Investables vs. MSCI China
Lags interval: 1 to 5
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None***
0.0720
18.0920***
17.9300***
At most 1
0.0007
0.1621
0.1621
***, **, * denote rejection of the hypothesis at the 1%, 5%, and 10% significance level
It is not surprising to find a long-term equilibrium between the fine wine market and the Chinese
stock market. As we can see in Figure 4, the movement of the fine wine price has performed
similarly to Chinese stock prices since the beginning of the 2000s. According to OIV, wine
0
20
40
60
80
100
120
140
160
180
1994M09
1995M04
1995M11
1996M06
1997M01
1997M08
1998M03
1998M10
1999M05
1999M12
2000M07
2001M02
2001M09
2002M04
2002M11
2003M06
2004M01
2004M08
2005M03
2005M10
2006M05
2006M12
2007M07
2008M02
2008M09
2009M04
2009M11
2010M06
2011M01
2011M08
2012M03
2012M10
2013M05
2013M12
2014M07
2015M02
Liv-ex Fine Wine Investables Vs. MSCI Emerging Markets
MSCI Emerging Markets MSCI EM Asia
MSCI BRIC Liv-ex Fine Wine Investables
16
consumption in China increased 67% from 2000 to 2012, and then came a decrease since 2012: a
drop of 3.8% between 2012 and 2013 and 7% between 2013 and 2014
19
.
Besides, we notice that the statistical significance of the cointegration relation between the fine
wine price and MSCI Hong Kong is rather weak, while the statistic for MSCI China is highly
significant. This evidence indicates that the major role is played by Mainland China. Hong Kong
would not have a significant enough impact on the fine wine market, without the support of the
huge demand from Mainland China. Because more than two-fifths of imported wine in Hong
Kong is re-exported to Mainland China
20
.
Figure 4: Liv-ex Fine Wine Investables vs. MSCI China Markets in real Dollar 2010=100
In country level, based on previous findings, the stock prices of Japan, Australia and the BRIC
countries have the highest probabilities to generate long-term relations with fine wine prices. The
results are presented in Table 4:
Table 4: Johansen Cointegration Test result Liv-ex Fine Wine Investable vs. MSCI Countries
Series: Liv-ex Fine Wine Investables vs. MSCI Japan
Lags interval: 1 to 4
19
Beyond the Chinese stock market evolution, Bordeaux fine wines have been influenced by two major shocks on
the Chinese market. In particular, in 2007, Hong Kong reduced the taxes on imported wine by half and then
removed them in 2008. The importation of wine soared and the prices skyrocketed with the ending of taxation. But
since 2011, fine wine prices have decreased. One reason commonly evoked is the Chinese anti-corruption policy of
Xi (gift-giving crackdown, fine wines are often presented as gifts or consumed during business meals in China).
20
Source: IWSR (International Wine and Spirit Research).
0
50
100
150
200
250
1994M09
1995M03
1995M09
1996M03
1996M09
1997M03
1997M09
1998M03
1998M09
1999M03
1999M09
2000M03
2000M09
2001M03
2001M09
2002M03
2002M09
2003M03
2003M09
2004M03
2004M09
2005M03
2005M09
2006M03
2006M09
2007M03
2007M09
2008M03
2008M09
2009M03
2009M09
2010M03
2010M09
2011M03
2011M09
2012M03
2012M09
2013M03
2013M09
2014M03
2014M09
Liv-ex Fine Wine Investables Vs. MSCI China Markets
MSCI Zhong Hua MSCI China
MSCI Hong Kong Liv-ex Fine Wine Investables
17
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0513
19.8808*
12.6813
At most 1
0.0294
7.1995
7.1995
Series: Liv-ex Fine Wine Investables vs. MSCI Australia
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None***
0.0770
22.3173***
19.4663***
At most 1
0.0117
2.8511
2.8511
Series: Liv-ex Fine Wine Investables vs. MSCI Brazil
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None***
0.0906
26.3261***
23.0771***
At most 1
0.0133
3.2490
3.2490
Series: Liv-ex Fine Wine Investables vs. MSCI India
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None***
0.1063
33.1459***
27.2989***
At most 1
0.0238
5.8470
5.8470
***, * denote rejection of the hypothesis at the 1% and 10% significance level
Figure 5: Liv-ex Fine Wine Investables vs. MSCI Countries in real Dollar 2010=100
18
The results for Australia, Brazil and India are highly significant. Long-run equilibriums between
the fine wine price and stock market prices of the three countries can also be observed through
their plots (see Figure 5). However, Japan produces conflicting results: the Trace statistic rejects
the null hypothesis of non-cointegration at 10% significance level, while the Max-Eigen statistic
cannot reject the hypothesis. Then we tried Nikkei225, and the results show that the null
hypothesis is rejected by the Trace and Max-Eigen statistics at 10% and 5% significance level
respectively (see Annex 5). For other countries, we also provide a cointegration analysis based on
national stock market indexes in the robustness check.
To summarize, the above cointegration test results can be presented in a world map figured
below. Underlined letters refer to the cointegration at 1% significance level, bold letters for 5%
significance level, and italic letters for 10% significance level.
0
50
100
150
200
250
300
1994M09
1995M03
1995M09
1996M03
1996M09
1997M03
1997M09
1998M03
1998M09
1999M03
1999M09
2000M03
2000M09
2001M03
2001M09
2002M03
2002M09
2003M03
2003M09
2004M03
2004M09
2005M03
2005M09
2006M03
2006M09
2007M03
2007M09
2008M03
2008M09
2009M03
2009M09
2010M03
2010M09
2011M03
2011M09
2012M03
2012M09
2013M03
2013M09
2014M03
2014M09
Liv-ex Fine Wine Investables Vs. MSCI Countries
MSCI Japan MSCI Australia MSCI Brazil
MSCI India Liv-ex Fine Wine Investables
19
Figure 6: World map - cointegration test procedure and results (see also Annex 6)
Robustness check
Liv-ex sub-indices analysis
The first robustness check concerns the wine price indexes. To confirm the previous results, one
needs to test the cointegration with different wine indexes and not only the fine wine investables
index which is simply composed of Bordeaux wines. Therefore the cointegration analysis is
extended to other Liv-ex indexes including the fine wines from the five main regions considered
by Liv-ex: Bordeaux, Burgundy, Champagne, Rhone and the Rest of World. In searching further
information, we use six Liv-ex sub-indices to refine the impact of each stock market on each
wine region index. Each sub-index represents the price movement of the ten most physical
vintages for the leading wines of the region, except for Bordeaux Legend 50 that includes only
50 top Bordeaux wines from exceptional older vintages
21
. The data of sub-indexes are only
available since December 2003.
Table 5: Johansen Cointegration Test Results Liv-ex sub-indices vs. MSCI indices
Zhong
Hua
China
Hong
Kong
EU
UK
France
Australia
Bordeaux 500
*
**
Bordeaux Legend 50
**
**
Burgundy 150
**
**
**
***
**
**
21
See https://www.liv-ex.com/ for more details concerning the component wines and vintages.
20
Champagne 50
***
*
**
**
***
***
Rest of World 50
*
**
**
**
Rhone 100
*
***, **, * denote rejection of null hypothesis at 1%, 5% and 10% significance level
Period: Dec. 2003 Feb. 2015; Number of observations: 132 after adjustment.
Table 5 shows the Johansen cointegration results (for countries exhibiting significant results
22
).
These results confirm that a cointegration link does exist between stock markets and fine wine
indexes. However, this table provides another outcome. Stock markets not only influence
Bordeaux fine wines but all the luxury wines from leading regions, because at least some
cointegration links appear for these top wine indexes (with the exception of Rhone 100 where
the cointegration appears very weak). Developed markets therefore seem to influence the prices
of Champagne and Burgundy wines. It might indicate that the major role of Asia on Liv-ex Fine
Wine Investables would be partially due to a compositional effect and the over-representation of
Bordeaux wines in this index.
Cointegration analysis with national stock market indexes
As a robustness check, we analyze the cointegration relationships between Liv-ex Fine Wine
Investables and 10 national stock market indexes: Dow Jones Industrial Average, Nasdaq
Composite, FTSE 100, DAX, CAC40, Nikkei225, Hang Seng, BVSP (Brazil), RTS (Russia) and
BSE SENSEX (India)
23
. The results are in line with our previous findings
24
(see Annex 5): Nikkei
225 is cointegrated with the fine wine price as mentioned above; Hang Seng and BVSP are
respectively cointegrated with the fine wine price at 5% significance level; no cointegration
relationship between the fine wine price and other indexes. Overall, the result is robust to the
choice of stock market index.
Granger Causality Test Results
Granger (1988) indicates that in the case of two cointegrated variables, there must exist causation
in at least one direction. So, to ensure the robustness of our results, we apply Granger causality
tests to see whether the cointegration relations are associated with causations in the correct
22
For readability concerns, the other results are available on request.
23
The data series of BSE SENSEX is from July 1997, shorter than other series (from September 1994).
24
Except for India where there is no cointegration. This result suggests that we should be cautious about drawing
conclusions about India. However our general conclusion is still valid.
21
direction” (i.e. from stock markets to fine wines). In our case, we examine the causation between
the fine wine price and each stock market price that they are cointegrated with, respectively. The
lag lengths of Granger causality tests correspond to the lag lengths of VARs. The results are
presented in Annex 7.
All of the tests reject the null hypothesis of non-causality for the MSCI Indexes to Liv-ex
direction. As a result, these stock market prices impact the fine wine prices. In other words, we
can have a better prediction of fine wine prices with the help of all available information of these
stock market prices. The results confirm the transmission of price fluctuation from financial
markets to the fine wine market in the short-medium term.
Sub-periods analysis
We analyze separately the pre-crisis period and the post-crisis period with May 2008 as a
demarcation point, because May 2008 is the highest point of Liv-ex before the crash. We run
cointegration tests on several representative relationships Liv-ex Fine Wine Investables with
the MCSI of Developed Markets, Emerging Markets, EU, G7, BRIC, Japan, China and Australia,
respectively. MSCI BRIC, China and Australia are cointegrated with the fine wine price on sub-
periods. The index of emerging markets is only cointegrated with the fine wine price on the pre-
crisis period. The indices of developed markets, EU, G7, and Japan
25
do not show a
cointegration relationship with the fine wine price on any of the sub-periods
26
. The financial
crisis has tremendous impacts on all the markets. Results indicate that under the same
macroeconomic shocks, the fine wine market does not respond to the financial crisis differently
to stock markets, especially with the influences of cash effect and transmission of market
volatilities. Therefore, compared to our full-period results, the relationship between the fine wine
market and stock markets on sub-periods does not change significantly.
Comments
Since the beginning of the 2000s, fine wine markets are more closely linked with emerging
markets than with developed markets. This phenomenon may be explained as following:
The consumption of wine in emerging markets has rapidly increased during the 2000s thanks to
economic growth and the increase in households’ disposable income. Studies show that the
existence of a stock market wealth effect explains the link between private consumption growth
and stock returns, and it is also the case for emerging markets (Funke, 2004; Poterba, 2000). The
25
To complement, Nikki225 also shows its ambiguity on sub-period analysis.
26
All the results of sub-period analysis are available on request.
22
boom of stock market-created wealth allows for consumption or reinvestment in other assets,
such as in fine wines. But, since the end of the 2000s, the slowdown of economic growth in
emerging markets has had a negative effect on the wine market. As for the developed markets,
wine consumption has dropped generally except in the USA and the UK
27
, following slower
growth in disposable income.
With the growth in wealth, a well-educated and open-minded population drives and encourages
the pursuit for more fashionable and qualified life styles. For instance, in China, with
government promotions, wine is considered as a healthier alcoholic drink compared to
traditional alcohol made with cereals. As the Veblen effect dominates the prestige-seeking
behavior of Chinese consumers, wealthy people regard the consumption and collection of fine
wines as a reflection of their wealth and social status. Combined with the awareness of country
of origin and top brands, this Veblen phenomenon can explain the old-and-top-Bordeaux fever
in the Chinese fine wine market. However, the Chinese wine market is a market in evolution.
According to the China Portraits report of Wine Intelligence in 2015, the “prestige-seeking
traditionalist” wine consumers are being overtaken by a new generation of wine drinkers who are
developing their wine habit and seeking value in wines. Meanwhile, high-end products are often
used for “gifting” purposes. When paying for themselves, Chinese consumers, especially the
growing middle classes, are willing to pay for higher quality or better taste rather than a higher
social status
28
. Luxury brands, including fine wines, have been suffering since the government’s
“gift-giving” crackdown. Evidence shows that when consumers become more rational and the
market steps towards to maturity, the irrational growth comes to a sudden end while the market
moves to a healthier stage.
Nevertheless, the influence of emerging markets, and especially China, has to be nuanced.
Indeed, the robustness check for the Liv-ex index indicates that Chinese stock markets are the
main drivers of Bordeaux wine prices, but this outcome might be different for wines from other
regions. Overall, luxury wine prices (especially Burgundy and Champagne wines) are influenced
by developed as well as emerging stock markets. However, Bordeaux wines still dominate any
wine price index, including Liv-ex Fine Wine Investables which we used, because of their
notoriety and their liquidity
29
.
27
According to OIV, between 2000 and 2014, major “old world” wine countries consume less wine with a
decrease of 19% in France, 34% in Italy and 29% in Spain; wine consumption increased 45% in the USA and 30%
in the UK.
28
Source : Wine Intelligence and Decanter China
29
Bordeaux wines are traded more also because they are available in higher quantity than Burgundy wines. In fact,
Burgundy can be seen as a very small luxury wine-producing region compared to Bordeaux.
23
6. Conclusion
This article analyzes long-term bilateral relations between the fine wine market and stock
markets, by applying a cointegration approach on a 21-year monthly database.
The analysis is conducted from world level to specific countries. At the world level, the Asia
Pacific region, which includes China, Australia, and Japan, is found to be the most influential
region on the fine wine market, in the long-term as well as in the short term. Comparing
developed markets versus emerging markets, emerging markets, especially the BRIC countries
(except Russia), exhibit a greater impact on fine wine markets, since the increasing demand from
emerging markets is a key factor influencing the fine wine market. As a result, the slowdown of
economic growth in emerging countries might be a risk to the fine wine market. Then we place a
greater emphasis on Chinese markets. Evidence shows that Mainland China plays a major role in
fine wine price fluctuation, more precisely, as regards the leading Bordeaux wines. The
uncertainty of Chinese politics can be another risk to the fine wine market. The robustness
checks seem to confirm these entire results even if for a more balanced wine portfolio (i.e. with
less Bordeaux and more wines from other regions) this outcome (Chinese influence) would have
to be nuanced in the sense that developed markets would also play a role in the price evolution.
We suggest that the above evidence can be taken into consideration in a portfolio diversification
strategy. And along with single-index analysis, our approach can be useful to provide a more
accurate forecast of fine wine price fluctuation.
References
Ashenfelter, O. (2008). Predicting the quality and prices of Bordeaux wine*. The Economic Journal,
118(529), F174-F184.
Aytaç, B., & Mandou, C. (2015). Wine: To drink or invest in? A study of wine as an investment
asset in French portfolios. Research in International Business and Finance.
Bouri, E. I. (2013). Do fine wines blend with crude oil? Seizing the transmission of mean and
volatility between two commodity prices. Journal of Wine Economics, 8(01), 49-68.
Bukenya, J. O., & Schaeffer, P. V. (2007). Do fluctuations in wine stocks affect wine prices?
Commodity modeling and Pricing, 136-166. John Wiley & Sons, Inc..
Burton, B. J., & Jacobsen, J. P. (2001). The rate of return on investment in wine. Economic Inquiry,
39(3), 337-350.
Cevik, S., & Saadi Sedik, T. (2014). A Barrel of Oil or a Bottle of Wine: How Do Global Growth
Dynamics Affect Commodity Prices?. Journal of Wine Economics, 9(01), 34-50.
24
Dimson, E., Rousseau, P. L., & Spaenjers, C. (2015 Forthcoming). The Price of Wine. American
Association of Wine Economists.
Dolado, J. J., Jenkinson, T., & SosvillaRivero, S. (1990). Cointegration and unit roots. Journal of
economic surveys, 4(3), 249-273.
Faye, B., Le Fur, E., & Prat, S. (2015). Dynamics of fine wine and asset prices: evidence from
short-and long-run co-movements. Applied Economics, 47(29), 3059-3077.
Fogarty, J. J. (2006). The return ofAustralian fine wine. European Review of Agricultural Economics,
33, 542-561.
Fogarty, J. J. (2010). Wine investment and portfolio diversification gains. Journal of Wine
Economics, 5(01), 119-131.
Fogarty, J. J., & Jones, C. (2011). Return to wine: a comparison of the hedonic, repeat sales and
hybrid approaches. Australian Economic Papers, 50(4), 147-156.
Fogarty, J. J., & Sadler, R. (2014). To Save or Savor: A Review of Approaches for Measuring
Wine as an Investment. Journal of Wine Economics, 9(03), 225-248.
Funke, N. (2004). Is there a stock market wealth effect in emerging markets?. Economics Letters,
83(3), 417-421.
Granger, C. W. (1969). Investigating causal relations by econometric models and cross-spectral
methods. Econometrica: Journal of the Econometric Society, 424-438.
Granger, C. W. (1988). Some recent development in a concept of causality. Journal of econometrics,
39(1), 199-211.
Greene, W. H. (2003). Econometric analysis. Pearson Education India.
Jaeger, E. (1981). To save or savor: the rate of return to storing wine. The Journal of Political
Economy, 584-592.
Johansen, S. (1988). Statistical analysis of cointegration vectors. Journal of economic dynamics and
control, 12(2), 231-254.
Johansen, S., & Juselius, K. (1990). Maximum likelihood estimation and inference on
cointegration with applications to the demand for money. Oxford Bulletin of Economics and statistics,
52(2), 169-210.
Jones, G. V., & Storchmann, K. H. (2001). Wine market prices and investment under
uncertainty: an econometric model for Bordeaux Crus Classés. Agricultural Economics, 26(2), 115-
133.
Kourtis, A., Markellos, R. N., & Psychoyios, D. (2012). Wine price risk management:
International diversification and derivative instruments. International Review of Financial Analysis,
22, 30-37.
25
Krasker, W. S. (1979). The rate of return to storing wines. The Journal of Political Economy, 1363-
1367.
Lockshin, L., & Corsi, A. M. (2012). Consumer behaviour for wine 2.0: A review since 2003 and
future directions. Wine Economics and Policy, 1(1), 2-23.
Lucey, B. M., & Devine, L. (2015). Was wine a premier cru investment?. Research in International
Business and Finance, 34, 33-51.
Masset, P., & Henderson, C. (2010). Wine as an alternative asset class. Journal of Wine Economics,
5(01), 87-118.
Masset, P., Cardebat, J. M., Faye, B., & Le Fur, E. (Working Paper 2015). Analyzing the risk of
an illiquid asset The case of fine wine.
Masset, P., Weisskopf, J. P., Faye, B., & Le Fur, E. (2015). Red obsession: the ascent of fine wine
in China.
Oczkowski, E., & Doucouliagos, H. (2014). Wine prices and quality ratings: A meta-regression
analysis. American Journal of Agricultural Economics, aau057
Poterba, J. M. (2000). Stock market wealth and consumption. The Journal of Economic Perspectives,
99-118.
Sanning, L. W., Shaffer, S., & Sharratt, J. M. (2008). Bordeaux wine as a financial investment.
Journal of Wine Economics, 3(01), 51-71.
Storchmann, K. (2012). Wine economics. Journal of Wine Economics, 7(01), 1-33.
Tsay, R. S. (2005). Analysis of financial time series (Vol. 543). John Wiley & Sons.
Vigneron, F., & Johnson, L. W. (1999). A review and a conceptual framework of prestige-
seeking consumer behavior. Academy of Marketing Science Review, 1(1), 1-15.
26
Annex 1 - Wine as Financial Asset Literature Summary
Study
Data period
Estimation method
Conclusions
Krasker
(1979)
Annual Heublein Wine Auctions
1973-1977;
Red Bordeaux and California
Cabernet Sauvignon;
Excluded vintages before 1950;
Repeat sales approach;
Weather Expected return to
storing wine exceeds the rate
of return of riskless assets?
The expected return on storing wines is below the rate of return of
riskless assets.
Jaeger (1981)
Annual Heublein Wine Auctions
1969-1977;
Red Bordeaux and California
Cabernet Sauvignon
Exclude vintages before 1950;
Repeat sales approach;
Weather Expected return to
storing wine exceeds the rate
of return of riskless assets?
Reverse Krasker’s finding;
Wine impressively outperforms government bonds.
Burton and
Jacobsen
(2001)
Edgerton
1986-1996 bi-annual data;
Red Bordeaux
Exclude vintages before 1960
Repeat sales approach
Considering the high volatility and transaction costs, the return on wine
is above government bonds but below equities.
Fogarty
(2006)
Langtons’ Auction;
1989Q4-2000Q4;
Premium Australian wine
Exclude vintages before 1965
Hedonic price regression
approach
The return on wine is above government bonds but below equities.
Best wines achieve higher returns and have a lower volatility than
poorer wines.
Premium Australian wine probably achieves higher return than
Bordeaux wine.
Sanning et al.
(2008)
The Chicago Wine Company;
Eight-year series of monthly
auction data;
Red Bordeaux vintages from 1893
to1998
Average return;
Repeat sales approach;
Fama-French Three-Factor
Model;
Capital Asset Pricing Model
Wine provides excess risk-adjusted returns.
s are negative or not significantly different from zero (average is -
0.006).
“Investment grade wine assets benefit from low exposure to market risk
factors”, thus offers portfolio diversification benefits.
Masset and
Henderson
(2010)
The Chicago Wine Company;
Auction data;
January 1996-February 2007
Weighted average of observed
prices;
VECM; GARCH;
Sharpe ratio;
Mean-variance efficient
frontier;
Mean-variance-Skewness-
Kurtosis analysis
Cumulated return on wine is above Dow Jones Industrial Average;
Best wines achieve higher returns and have a lower volatility than
poorer wines. Wines of different categories follow a similar trend in
long-run, but partially independent in short-run.
Including wine in a portfolio provides diversification benefits: Optimal
Sharpe ratio, increase in skewness, but also in kurtosis.
Slight correlation (-0.220.5) between the wine returns and other assets,
while during financial crisis, the correlation increased very significantly.
27
Study
Data period
Estimation method
Conclusions
Fogarty
(2010)
Langton’s auction house;
Quarterly auction data;
1990Q1-2000Q4;
Premium Australian wine
Exclude vintages before 1965
Repeat sales approach;
Sharpe ratio;
Mean-variance efficient
frontier
In the case of Australian wine, the return to and the risk adjusted excess
return are both lower than standard financial assets; the correlation with
equities is 0.14, and -0.11 with bonds; including wine in a portfolio, even
already well diversified one, provides diversification benefits by
achieving the same level of portfolio return with lower risk.
Fogarty and
Jones (2011)
Langton’s auction house;
Quarterly auction data;
1988Q1-2000Q4
Premium Australian wine
Exclude vintages before 1965
Hedonic, repeat sales, and
hybrid approaches;
Mean-variance spanning test;
The return and diversification benefits of wine are dependent on the
estimation method.
Repeat sales approach may overstate the return of wine.
Hybrid approach provides the most efficient estimates.
Kourtis et al.
(2012)
WinePrice.com (2005-2010)
Monthly data of 7 wine
production regions
Liv-ex indexes (2001-2010)
Monthly data
Logarithmic returns of
observed prices;
Skewness- kurtosis analysis;
Continuous time dynamics
models
Returns and diversification benefits vary from different wine production
regions.
Slight correlation (< 0.3) between the wine returns and other assets.
International diversification benefits for investing in wines from Italy,
Australia and Portugal as addition to French wines.
Fogarty and
Sadler (2014)
Langton’s auction house;
Quarterly auction data;
1988Q1-2000Q4
Premium Australian wine
Exclude vintages before 1965
Approach review:
- Return estimations: hedonic,
repeat sales, pooled, hybrid
approaches;
- Diversification benefit tests:
Jobsen approach, Blume and
Elton et al., mean-variance
spanning approach, mean-
variance efficient frontier
The diversification benefits of including wine in a portfolio are sensitive
to the return estimation method, the sample period, whether raw return
or excess return used, and the type of diversification benefit test.
Repeat sales approach overstates the return on wine.
In the case of Australian wine, the diversification benefits of including
wine in a portfolio are small.
Dimson et al.
(2015)
Forthcoming
Auction data from Christie’s;
Dealer data from Berry Bros. &
Rudd;
Historical database;
Since end 1899 from now on
Repeat sales approaches;
The effect of aging on wine
prices through time (life-cycle
price dynamics of low quality
vintage and high quality
vintage)
Between 1900 and 2012, in real GBP terms, wine (net annual return
4.1%) underperforms equities, but outperforms government bonds, art,
and stamps.
A positive correlation between wine and equity returns. is 0.73 for the
full period, 0.57 if excluding 1941-1948.
Lucey and
Devine
(2015)
The Chicago Wine Company;
Auction data 1996-2006
5 wine funds 2003-2010
Repeat sales approach;
Sharpe ratio
Average return on wine is above Dow Jones Industrial Average;
Wine is attractive, but individual and sub-regional level returns to wine
are highly various and volatile. Authors suggest individual investors to
28
Study
Data period
Estimation method
Conclusions
Bordeaux and Rhone fine wines
Invest indirectly through wine funds or traded indexes.
Aytac et al.
(2015)
IdealWine and Liv-ex indexes
January 2007-October 2014
Bordeaux, Bourgogne, Rhone
Wines
Returns of observed prices;
Skewness- kurtosis analysis;
Mean-variance efficient
frontier;
Mean-modified value at risk
Wines outperform equities for the considered period.
Including wine in a portfolio provides diversification benefits
Bordeaux fine wines are the most efficient for improving the
performance of French portfolio compared to other French fine wines.
29
Annex 2 MSCI index categories and components
30
Annex 3 Augmented Dickey - Fuller Unit Root Test Results
Category
Series
t-Stat. in Level
t-Stat. in First
Difference
Result
Wine Prices
Liv-ex Fine Wine
Investables Index
-2.41
trend/constant***
-9.69***
trend/constant*
I(1)
Bordeaux 500
-1.84
trend/constant*
-6.04***
trend/constant
I(1)
Bordeaux Legend 50
-1.77
trend/constant*
-6.61***
trend/constant
I(1)
Burgundy 150
-2.62
trend**/constant***
-6.02***
trend/constant
I(1)
Champagne 50
-1.93
trend/constant**
-8.86***
trend/constant
I(1)
Rest of World 50
-2.61
trend**/constant***
-10.66***
trend/constant
I(1)
Rhone 100
-2.09
trend/constant**
-6.56***
trend/constant
I(1)
MSCI
All Country
(DM &
EM)
ACWI
-2.09
trend/constant**
-13.88***
trend/constant
I(1)
AC Asia Pacific
-2.60
trend/constant**
-13.34***
trend/constant
I(1)
AC America
-2.19
trend/constant**
-14.33***
trend/constant
I(1)
AC Europe
-2.42
trend/constant**
-13.80***
trend/constant
I(1)
MSCI
Developed
Markets
World
-2.06
trend/constant**
-13.90***
trend/constant
I(1)
G7
-1.97
trend/constant**
-13.98***
trend/constant
I(1)
Pacific
-2.61*
trend/constant**
-13.53***
trend/constant
I(0)*
EAFE
-2.46
trend/constant**
-13.69***
trend/constant
I(1)
EU
-2.36
trend/constant**
-13.89***
trend/constant
I(1)
North America
-2.26
trend/constant**
-14.26***
trend/constant
I(1)
MSCI
Emerging
Markets
World
-2.43
trend**/constant**
-13.41***
trend/constant
I(1)
BRIC
0.28
trend/constant
-13.64***
trend/constant
I(1)
Asia
-2.53
trend*/constant**
-12.71***
trend/constant
I(1)
MSCI
China
Markets
Zhong Hua
-2.41
trend*/constant**
-14.08***
trend/constant
I(1)
Hong Kong
-2.28
trend/constant**
-14.31***
trend/constant
I(1)
China
-2.22
trend/constant**
-14.20***
trend/constant
I(1)
USA
-2.19
-14.28***
31
MSCI DM
Countries
trend/constant**
trend/constant
I(1)
UK
-2.16
trend/constant**
-13.94***
trend/constant
I(1)
Germany
-2.38
trend/constant**
-14.88***
trend/constant
I(1)
France
-2.42
trend/constant**
-14.56***
trend/constant
I(1)
Japan
-2.41
trend/constant**
-13.63***
trend/constant
I(1)
Australia
-2.46
trend*/constant**
-15.09***
trend/constant
I(1)
MSCI EM
Countries
Brazil
-0.08
trend/constant
-15.04***
trend/constant
I(1)
Russia
-2.39
trend/constant**
-13.16***
trend/constant
I(1)
India
-2.65
trend***/constant**
-14.41***
trend/constant
I(1)
Stock
Markets
Indexes
Dow Jones Industrial
Average
-2.58*
trend/constant***
-15.22***
trend/constant
I(0)*
Nasdaq Composite
-1.88
trend/constant**
-14.19***
trend/constant
I(1)
FTSE 100
-2.28
trend/constant**
-13.41***
trend/constant
I(1)
DAX
-1.78
trend/constant*
-14.58***
trend/constant
I(1)
CAC 40
-2.27
trend/constant**
-14.12***
trend/constant
I(1)
Nikkei 225
-2.24
trend/constant**
-14.75***
trend/constant
I(1)
Hang Seng
-3.03
trend*/constant***
-14.95***
trend/constant
I(1)
BVSP
0.10
trend/constant
-13.09***
trend/constant
I(1)
RTS
-2.16
trend/constant**
-12.29***
trend/constant
I(1)
BSE SENSEX
-2.30
trend**/constant**
-12.77***
trend/constant
I(1)
***, **, * denotes rejection of null hypothesis (non-stationary for unit root test, non-significance
for trend or constant) at the 1% , 5%, and 10% significance level.
Annex 4 Johansen Cointegration Test Results (non-cointegrated)
Series: Liv-ex Fine Wine Investables vs. MSCI ACWI
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0518
16.2887
12.8271
32
Liv-ex Fine
Wine
Investables
vs.
MSCI
All Country
(AC)
(DM & EM)
At most 1
0.0143
3.4616
3.4616
Series: Liv-ex Fine Wine Investables vs. MSCI AC America
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0562
15.4881
13.9400*
At most 1
0.0064
1.5481
1.5481
Series: Liv-ex Fine Wine Investables vs. MSCI AC Europe
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0420
14.2494
10.4206
At most 1
0.0156
3.8288
3.8288
Liv-ex Fine
Wine
Investables
vs.
MSCI
Developed
Markets
(DM)
Series: Liv-ex Fine Wine Investables vs. MSCI DM World
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0535
16.0121
13.2400
At most 1
0.0114
2.7722
2.7722
Series: Liv-ex Fine Wine Investables vs. MSCI G7
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0547
15.9419
13.5595
At most 1
0.0098
2.3824
2.3824
Series: Liv-ex Fine Wine Investables vs. MSCI EAFE
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0441
18.3981*
10.8734
At most 1
0.0307
7.5248
7.5248
Series: Liv-ex Fine Wine Investables vs. MSCI DM EU
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0428
14.2961
10.6217
At most 1
0.0150
3.6744
3.6744
Series: Liv-ex Fine Wine Investables vs. MSCI North America
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0553
15.0574
13.8195
At most 1
0.0051
1.2380
1.2380
Series: Liv-ex Fine Wine Investables vs. MSCI USA
33
Liv-ex Fine
Wine
Investables
vs.
MSCI
Countries
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0562
15.2573
13.9519*
At most 1
0.0054
1.3054
1.3054
Series: Liv-ex Fine Wine Investables vs. MSCI UK
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0465
15.6497
11.5673
At most 1
0.0167
4.0824
4.0824
Series: Liv-ex Fine Wine Investables vs. MSCI Germany
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0488
16.8569
12.0578
At most 1
0.0197
4.7992
4.7992
Series: Liv-ex Fine Wine Investables vs. MSCI France
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0463
15.0800
11.4269
At most 1
0.0150
3.6532
3.6532
Series: Liv-ex Fine Wine Investables vs. MSCI Russia
Lags interval: 1 to 3
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0332
9.1624
8.0573
At most 1
0.0046
1.1050
1.1050
* denotes rejection of null hypothesis 10% significance level.
Annex 5 - Johansen Cointegration Test Results National stock market indexes
Series: Liv-ex Fine Wine Investables vs. Nasdaq Composite
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0506
14.6099
12.5133
At most 1
0.0087
2.0966
2.0966
Series: Liv-ex Fine Wine Investables vs. FTSE100
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0464
15.1683
11.5345
34
At most 1
0.0015
3.6338
3.6338
Series: Liv-ex Fine Wine Investables vs. DAX
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0400
14.1927
9.9320
At most 1
0.0174
4.2608
4.2608
Series: Liv-ex Fine Wine Investables vs. CAC40
Lags interval: 1 to 4
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0450
14.8542
11.0962
At most 1
0.0155
3.7580
3.7580
Series: Liv-ex Fine Wine Investables vs. Nikkei225
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None*
0.0642
20.1455*
16.1272**
At most 1
0.0164
4.0183
4.0183
Series: Liv-ex Fine Wine Investables vs. Hang Seng
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None**
0.0680
24.5259**
17.1062**
At most 1
0.0301
7.4197
7.4197
Series: Liv-ex Fine Wine Investables vs. BVSP
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None**
0.0778
23.7285**
19.6792**
At most 1
0.0165
4.0493
4.0493
Series: Liv-ex Fine Wine Investables vs. RTS
Lags interval: 1 to 3
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0294
11.9793
6.8697
At most 1
0.0220
5.1096
5.1096
Series: Liv-ex Fine Wine Investables vs. BSE SENSEX (data from 1997)
Lags interval: 1 to 2
Hypothesized
No. of CE(s)
Eigenvalue
Trace
Statistic
Max-Eigen
Statistic
None
0.0451
11.0896
9.6362
At most 1
0.0069
1.4535
1.4535
**, * denote rejection of null hypothesis 5% and 10% significance level.
35
Annex 6 - World map 2 - cointegration test procedure and results
Annex 7: Granger Causality Test Results Liv-ex Fine Wine Investables Vs. MSCI
Indexes
causes
Liv-
ex
AC
Asia
Pacific
EM
World
BR
IC
EM
Asia
Zhong
Hua
Hong
Kong
Chi
na
Japa
n
Aus-
tralia
Brazi
l
Indi
a
Liv-ex
**
AC Asia
Pacific
***
EM
World
***
BRIC
***
EM Asia
***
Zhong
Hua
***
Hong
Kong
**
China
***
36
Japan
**
Australia
***
Brazil
***
India
***
***, ** denotes rejection of the hypothesis at the 1%, 5% significance level
... However, another stream of literature looks at macroeconomic determinants of fine wine prices and models fine wines as a single-asset rather than a combination of assets in a portfolio (Le Fur and Outreville, 2019). Some of the macroeconomic factors that have been found to influence the pricing of fine wines include real exchange rates (Cardebat and Figuet, 2019), growth of fine wine imports in China (Anderson and Wittwer, 2013), changes in GDP, weakening US dollar (Jiao, 2017), the demand from emerging markets (Cardebat and Jiao, 2018;Cevik and Sedik, 2014) the stock markets (Faye et al., 2015;Introvigne et al., 2017;Cardebat and Jiao, 2018), oil markets (Bouri and Azzi, 2013;Bouri, 2014), art, commodities and real estate (Ameur and Le Fur, 2020). Furthermore, fine wine prices have been found to be impacted by bilateral trading agreements and currency fluctuations as a consequence of geopolitical events Economic policy uncertainty (Chavis and Leslie, 2006;Dimson et al., 2015) and more recently, Brexit (Anderson and Wittwer, 2018). ...
... However, another stream of literature looks at macroeconomic determinants of fine wine prices and models fine wines as a single-asset rather than a combination of assets in a portfolio (Le Fur and Outreville, 2019). Some of the macroeconomic factors that have been found to influence the pricing of fine wines include real exchange rates (Cardebat and Figuet, 2019), growth of fine wine imports in China (Anderson and Wittwer, 2013), changes in GDP, weakening US dollar (Jiao, 2017), the demand from emerging markets (Cardebat and Jiao, 2018;Cevik and Sedik, 2014) the stock markets (Faye et al., 2015;Introvigne et al., 2017;Cardebat and Jiao, 2018), oil markets (Bouri and Azzi, 2013;Bouri, 2014), art, commodities and real estate (Ameur and Le Fur, 2020). Furthermore, fine wine prices have been found to be impacted by bilateral trading agreements and currency fluctuations as a consequence of geopolitical events Economic policy uncertainty (Chavis and Leslie, 2006;Dimson et al., 2015) and more recently, Brexit (Anderson and Wittwer, 2018). ...
... We specifically focus on the following MSCI developed economies: the USA, UK, Germany and France as these countries are among the leading wine markets in terms of consumption (Conway, 2019). Additionally, we also consider China in our sample as prior research has indicated that this market is one of the main drivers of the fine wine markets (Cardebat and Jiao, 2018). To measure the EPU of the USA, the BBD measure relies on newspaper articles in ten leading newspapers in the country. ...
Article
Full-text available
Purpose The paper aims to investigate the impact of economic policy uncertainty (EPU) (i.e. uncertainty due to government policies) on fine wine prices. Design/methodology/approach The paper uses the Baker et al. (2016) monthly news-based measure of EPU for the leading wine markets: the USA, the UK, France, Germany and China in conjunction with monthly fine wine pricing data from the London International Vintners Exchange (Liv-ex). The wine sub-indices used are the Liv-ex 500 (Bordeaux), Burgundy 150, Champagne 50, Rhone 100, Italy 100, California 50, Port 50 and Rest of the World 50. The Prais–Winsten and Cochrane–Orcutt regressions are used for our analyses to correct for effects of serial correlation. Time lags are chosen based on the appropriate information criterion. Findings Changes in EPU levels negatively impact changes in the Liv-ex 500 index for all our leading wine markets except France, the Champagne 50 index for the UK and the Burgundy 150 and the Rhone 100 indices for Germany, with the effects being significant for at least up to a quarter before EPU is detected. The authors did not find significant results for the EPU of France. Practical implications The paper aims to provide insights into whether EPU creates opportunities or threats for investors and wineries. Originality/value A forward-looking news-based EPU measure is used to gain insights into how the different Liv-ex sub-indices react to increases in uncertainty centered around government policies across a sample of different countries.
... Their results completely reversed the previous findings. Meanwhile, Cardebat and Jiao (2018) discovered long-term relationships between fine wine and stock markets. Masset and Weisskopf (2018b) indicated that the inconsistency among the different results was partly due to the sensibility of data frequency and to the issue of illiquidity. ...
Article
Full-text available
In this paper, we use copula-GARCH models applied to daily data from March 2010 to March 2018 to test the time-varying dependence of the Liv-ex 50, a secondary market fine wine index comprised of the ten most recent vintages of the five Bordeaux First Growths, with a portfolio composed of the six main stock markets (S&P 500, CAC 40, DAX 30, FTSE 100, and Hang Seng). Our results suggest that the Liv-ex 50 underperforms the six stock indexes, but provides diversification benefits in terms of volatility, asymmetry, and extreme events. (JEL Classifications: G110, G120, Q14)
... The rapidly growing body of literature has examined wine predominantly from a microeconomic angle, with the hedonic pricing model as its most commonly employed work horse. While hedonic analyses indirectly consider demand but do not explicitly take into account macroeconomic variables, various authors have shown that wine prices and the wine trade follow the same macroeconomic determinants, such as exchange rates, incomes, or economic cycles, as do traditional financial assets (e.g., Cevik and Sedik, 2014;Jiao, 2017;Cardebat and Jiao, 2018). ...
Article
Full-text available
This special issue of the Journal of Wine Economics is based on a workshop entitled “Wine, Macroeconomics, and Finance” jointly organized by INSEEC U., INFER, and LAREFI (University of Bordeaux), and sets out to contribute to various macroeconomic aspects of two domains of wine economics, the international wine trade, and the role of wine as a financial asset. (...)
... Furthermore, our study extends the previous literature in two ways. First, we incorporate the Liv-ex 100 index into the ex-n egociant pricing; Liv-ex 100 index serves as a proxy for reflecting consumers' willingness to pay for fine wines (Cardebat and Jiao 2018). Second, we develop unique variable definitions that compare two consecutive vintages in order to account for the 1-year planning phenomenon. ...
Article
Earlier publications indicate that the price of young wines cannot be estimated accurately through weather conditions. Considering that a majority of fine wine is sold in the form of wine futures, even before the wine is bottled, determining realistic prices for these wine futures accurately is one of the most critical decisions. Surveys of leading wine merchants and industry experts exhibit significant departures from the realized market prices. We develop a pricing model for fine wines using weather, market and expert reviews. The financial exchange for fine wines called Liv‐ex has already adopted our estimated prices and tagged them as “realistic prices.” Our approach combines temperature, rainfall, market fluctuations and tasting expert scores and leads to accurate estimations that the wine industry has not seen before. Our study shows that higher temperatures, lower levels of precipitation, appreciation in the Liv‐ex 100 index as a market indicator, and higher barrel scores increase market prices. We conduct a comprehensive set of robustness checks and show that the mean absolute deviation of actual market prices from our estimated prices is substantially smaller than any academic benchmark. Our realistic prices help create transparency in this highly opaque market. When compared with the realized prices, our realistic prices guide buyers (e.g., distributors, restaurateurs, merchants) in their purchase decisions as they can determine whether a wine is underpriced or overpriced.
Chapter
Full-text available
We study the dynamic relation between the global and European stock markets, fine wine market and GBP/ USD exchange rate by using VAR-DCC-GARCH framework and daily closing prices of LIVX50, S&P500 and FTSE100 indices from 2010 to 2018. Three versions of univariate GARCH models, namely standard, exponential (Nelson, 1991) and GJR (Glosten, Jagannathan, Runkle, 1993) were used in order to build best fitted multivariate dynamic conditional correlation model. Results of this study reveal the long-term time-varying links in volatility between the global and European markets. We found evidence of negative correlation between fine wines market and global stock markets in few periods. Most important results provide empirical evidence that fine wines can be hedge against declines of British currency and can help investors minimize risk to build optimal portfolios.
Article
Full-text available
We test the nature of weak form informational efficiency present in the wine market using daily return of LIV-EX 50 index from 1/1/2010 to 12/6/2020. First, we employ a number of statistical tests including variance ratio tests, tests for linear and non-linear dependence and Hurst coefficient. The tests are applied on the full dataset and on four non overlapping sub-samples of equal length. The variance ratio tests provide a mixed regarding informational efficiency. Evidence of non-linear dependence in the return series was found. The Hurst coefficient values confirm the presence of long run persistence in the wine market. Based on the mixed evidence, we test the possibility of adaptive nature of the wine market. We employ the newly proposed Adaptive Index (AI) to quantify the degree of information inefficiency in the wine market at any instance. Our results confirm that wine market is adaptive and periodically shifts between states of efficiency and inefficiency. The wine market is found to be relatively free from the Covid-19 induced shock and the safe haven property of wine is thus confirmed. Finally, impact of various macroeconomic and financial events on wine market efficiency is identified by using AI.
Article
Full-text available
This paper explores price effects in the “passion investments” market after days with abnormal returns. To do this, daily prices for stamps and diamonds over the periods 1999–2021 and 1989–2021 are analyzed. The following hypothesis is tested: One-day abnormal returns create stable patterns in price behavior on the next day. Statistic tests (t-test, ANOVA, Mann–Whitney U test, modified cumulative abnormal returns approach, regression analysis with dummy variables) confirm the presence of price patterns related to extreme returns: price fluctuations on the day after extreme returns are higher than returns on “normal” days. On the days after positive abnormal returns, the momentum effect is detected. Contrarian effect is typical for the days after negative abnormal returns. A trading strategy based on detected price effects showed the presence of exploitable profit opportunities. Results of this paper provide additional pieces of evidence in favor of inconsistencies between the efficient market hypothesis and practice and can be used by traders to generate extra profits in the “passion investments” market. Acknowledgment The authors gratefully acknowledge financial support from the Ministry of Education and Science of Ukraine (0121U100473).
Article
This paper models optimal release prices of an experience good recurrently issued on markets. Using a sample of Bordeaux wines, we find that using a minimal number of intrinsic and extrinsic attributes is sufficient to explain a large proportion of release prices. We further observe a significant relationship between primary market prices and secondary market prices and general economic conditions. Release prices can deviate from secondary market prices in the short run but remain aligned over the long run. Finally, an out‐of‐sample analysis indicates that short‐run mispricing directly affects the purchase behavior of customers.
Although economic theory clearly provides reasons for a hedonic price function to change over time, this is not how it is traditionally approached empirically. For wine, which is one of the most widely used implications of hedonic price analysis, such time invariance can be traced down to the issues of data availability and price rigidity in the food and beverage sector. The development of e-commerce changes both these premises, providing reasons for more frequent price adjustments and making price data available at each point of time. Is willingness to pay for particular product attributes indeed time-invariant? In this paper, daily price quotes from the largest online grocery market in Russia are used to address this question using sparkling wines as an example. The results indicate that time invariance might be a rather strong empirical assumption, suggesting that the time dimension should be included into hedonic price studies whenever the data allows for it.
Article
Full-text available
We use a unique and very deep database to examine the performance of wine investments over the period 2003–2014. Our results reveal that the returns stemming from those investments are important but can largely be explained by their exposure to common risk factors. It appears essential to properly account for the lack of liquidity on the wine market and its exposure to global currency risk. Controlling for these wine market specificities and contradicting prior evidence, fine wines do not seem to offer abnormal returns.
Article
Full-text available
This article uses hammer prices from five global auction houses to analyse the price premium Bordeaux fine wine yielded at Hong Kong wine auctions. We find that fine wine was on average sold at a 19% premium in Hong Kong. We further observe that the Hong Kong premium is not uniform and most pronounced for wines with perfect Parker scores and the most powerful brands. The premium has declined throughout the sample period from 60% in 2008 to a level of 15% since 2012. This can be attributed to the increase in knowledge on fine wine by Chinese customers.
Article
Full-text available
IntroductionBackground Inventory and Price BehaviorTesting for TrendsCointegration Between Stocks and PricesVector Autoregression and Impulse ResultsConclusions
Article
Full-text available
This paper investigates the causes of extreme fluctuations in commodity prices from 1990 to 2010. Analyzing two very distinct commodities-crude oil and fine wine, we find that macroeconomic factors are the main determinants of commodity prices. Although supply constraints have the expected effect, aggregate demand growth is the key factor. The empirical results show that while advanced economies account for more than half of global consumption, emerging economies make up the bulk of the incremental change in demand, thereby having a greater weight in commodity price formation. The results also show that the shift in the composition of aggregate commodity demand is a recent phenomenon.
Article
This article examines the empirical support for the hypothesized hedonic theoretical relation between the price of wine and its quality. The examination considers over 180 hedonic wine price models developed over 20 years, covering many countries. The research identifies that the relation between the price of wine and its sensory quality rating is a moderate partial correlation of +0.30. This correlation exists despite the lack of information held by consumers about a wine's quality and the inconsistency of expert tasters when evaluating wines. The results identify a moderate price-quality correlation, which suggests the existence of strategic buying opportunities for better informed consumers. Strategic price setting possibilities may also exist for wine producers given the incomplete quality information held by consumers. The results from the meta-regression analysis point to the absence of any publication bias, and attribute the observed asymmetry in estimates to study heterogeneity. The analysis suggests the observed heterogeneity is explained by the importance of a wine's reputation, the use of the 100-point quality rating scale, the analysis of a single wine variety/style, and the employed functional form. The most important implication from the analysis is the relative importance of a wine's reputation over its sensory quality, inferring that producers need to sustain the sensory quality of a wine over time to extract appropriate returns. The reputation of the wine producer is found not to influence the strength of the price-quality relationship. This finding does not contradict the importance of wine producer reputation in directly influencing prices.
Article
In the literature, there is no standard approach for estimating the return to wine or testing for a portfolio risk diversification benefit from holding wine. Using auction data for Australian wine, we show that the estimation method has a material impact on the estimated wine return distribution and that the type of diversification benefit test used influences whether or not wine is found to provide a portfolio risk diversification benefit. Our results indicated that a simple modification to the hedonic model, which we call a pooled model, is an appropriate method for estimating the return to infrequently traded heterogeneous assets such as wine. Across the various approaches to testing for a risk diversification benefit, we find direct estimation of the efficient frontier with bootstrapped confidence intervals to be the most transparent and comprehensive method of illustrating wine's potential for lowering portfolio risk.
Article
Using a dataset that spans the period 1996 to 2007 and contains transaction prices for all reported auctions at the Chicago Wine Company, we analyze how the prices of high-end wines have evolved during this time period. The best wines according to characteristics like vintage, rating and ranking earn higher returns and tend to have a lower variance than poorer wines. Nevertheless, the different categories of wines seem to follow a rather similar trend over the long run. Wine returns are only slightly correlated with other assets and can consequently be used to reduce the risk of an equity portfolio. Wine looks even more attractive when the investor also has concerns about the skewness of his portfolio. However, the part to be invested in wine is reduced once the kurtosis is included into the analysis. Finally, it seems advisable to diversify across different wine categories as their short-run movements are partially independent of each other. First growths and wines rated as extraordinary by Robert Parker deliver the best tradeoff in terms of portfolio expected returns, variance, skewness and kurtosis for most investor preference settings under consideration.