ANALYZING TRENDS AND CYCLES IN INTERNATIONAL WINE TRADE - PARADIS.pdf

(127 KB) Pobierz
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
Colloque Œnométrie XI
VDQS
Dijon 21/22 mai 2004
Walter C. Labys and Bruce C. Cohen
Walter C. Labys
Professor of Resource Economics and Faculty Research Associate
Regional Research Institute, West Virginia University
Morgantown, WV, USA, 26506-6108
Tel.: (1-304)293-4832; Fax: (1-304)293-3752
E-mail: wlabys@wvu.edu
Bruce C. Cohen
Economist (ret), United Nations Economic Commission for Europe
United Nations – Palais des Nations
Geneva, Switzerland, CH 1202
Keywords: wine trade shifts, new world wine producers, old world wine producers,
market shares, wine trends, wine cycles, international wine trade
JEL codes: Q13, Q16, O56
Abstract: The global wine market has witnessed major changes in recent years. Some of
these changes are structural in nature, while others are cyclical. The structural
adjustments reflect changing trade shares due to long term forces. Shifts in wine
production capacity, mainly through increasing vineyard area and replacing table wine
grapes with higher quality grape varieties, take time. New producing entrants have
appeared in Latin America, the United States, Australia and New Zealand. International
trade patterns and shares have thus shifted with the new market entrants increasing their
exports not only to traditional European markets but to other importing regions as well.
The cyclical adjustments in wine markets are medium or shorter term in nature. These
vary between the impacts of weather fluctuations on grape production to the influences of
business cycles on wine trade, consumption and prices. This paper employs econometric
methods to analyze the recent major shifts in world wine market shares and to explain
whether these are more of a secular trend-setting nature or of a temporary cyclical nature.
Paper prepared for the OENOMETRICS XI conference of the VDQS-AEA, Dijon France,
May 20-22,2004. Draft May 1, 2004.
www.vdqs.net
Colloque Œnométrie XI
VDQS
Dijon 21/22 mai 2004
ANALYZING TRENDS AND CYCLES IN
INTERNATIONAL WINE TRADE
The global wine market has witnessed major changes in recent years. Some of
these changes are structural in nature, while others are cyclical. The structural
adjustments reflect changing trade shares due to long term forces. Shifts in wine
production capacity, mainly through increasing vineyard area and replacing table wine
grapes with higher quality grape varieties, take time. New producing entrants have
appeared in Latin America, the United States, Australia and New Zealand. International
trade patterns and shares have thus shifted with the new market entrants increasing their
exports not only to traditional European markets but to other importing regions as well.
The cyclical adjustments in wine markets are medium or shorter term in nature. These
vary between the impacts of weather fluctuations on grape yields to the influences of
business cycles on wine trade, consumption and prices. The purpose of this paper is to
analyze the global significance of these changes. This paper consists of the following
parts: The Setting, The Facts, Trends or Cycles, and Prospects.
The Setting
We occupy the center of a wine revolution that is without parallel. A form of
globalization has truly struck wine markets. Quality wines are being produced on six
continents, and developed as well as developing nations are active in international wine
trade. This has occurred because high incomes, finance, politics and gourmet culture
gradually emerged in former politically and culturally wine-restrictive nations.
Historically wine production and consumption have been mainly in western Europe;
France, Italy, Portugal and Spain constitute the core of old world wine producers. Greece
and, in eastern Europe, Bulgaria, Hungary, Moldava and Romania are added to the above
western European grouping to complete the list of old world wine producers. But now
we have large and growing wine production and consumption in the new world:
Argentina, Australia, Chile, New Zealand, South Africa, Uruguay and the United States.
The new world expansion has changed the way is which wine is appreciated
according to flavor, variety and national origin as well as to wine market structure and the
way that wine is traded internationally. The traditional concept of wine, designated
according to vineyard of origin or appellation, is being replaced by that of varietal or
grape variety. This transition began with an emerging preference for cabernet, merlot
and chardonnay varietal wines. As wine producers, franchisers and distributors responded
to this demand, the planting of these grapes and consequent wine production exploded
globally. These three varietals provide a flavor and aroma that is agreeable and easily
identifiable. Once this stage was passed, the production of varietals expanded to include
many of the other grapes used in traditional wines: riesling, sauvignon blanc, gamay,
gewurtztraminer, pinot noir, shiraz, just to name a few. The reaction of the old world
producers to this change has been to bottle some of their wines as varietals and to begin
to blend some new varietals into their traditional wines. Examples include planting
viognier, supplementing sangiovese in the making of Chianti or varying pinot noir clones
in the making of Bourgogne Rouge.
www.vdqs.net
Colloque Œnométrie XI
VDQS
Dijon 21/22 mai 2004
According to Kramer (2003), improvements in technology have made it possible
to adjust the flavor and aroma of wines in almost any direction. The new wine drinkers,
who represent the largest demand group, have reshaped modern taste. Not having had the
benefit of tasting mature wines nor of having been tutored by traditional experts, the new
drinkers seem to want ripe-tasting, rich, full-flavored wines with some oakiness and at
reasonable prices. Thus, the definition of wine quality is no longer the domain of the
wine producer. One of the first innovations that enabled the improvements in wines
among regions and nations was the temperature-controlled fermentation tank. Then the
field of sensory science made it possible to understand and to alter the sensory perception
of food and beverages through olfactory profiling. Most noticeable are the Australian-led
innovations. Instead of using traditional oak barrels to impart flavor, oak chips and
sawdust can be added to the fermenting wine, then filtered. Wine yields are increased
rather than reduced through selection. The excessively juice-laden grapes then pass
through vacuum concentrators (an ultra-fine filtration process) that remove water from
the diluted must made from these grapes. Also by over-ripening the grapes, wine flavor is
enhanced but the sugar rises and concomitantly the alcohol of the wine. This can be
lowered by using a spinning cone that can control the alcohol content fairly precisely.
Finally acid levels can be modified by adding tartaric or citric acid; and powdered tannin
can be added to increase the body or structure of red wines.
The existing very large wine and alcohol conglomerates have tried to address this
new consumer interest mainly through franchising and marketing. Franchising here refers
to a large societal impulse toward predictability, reliability and security in wine
purchasing. Today consumers demand more flavorful, more aromatic, identifiable and,
accessible wines, regardless of the grape variety or the traditional taste of the wine. There
seems to be a preference for dry rather than sweet wines. But sparkling wines too are on
the rise. Wine makers have responded by employing the new technologies to manipulate
the structure of their products so that this goal is met, enlarging their efforts on marketing
branded wines. Whereas European and even American wine producers identify quality
with localized districts and small regions, other new world producers control
fermentation and blend wines to create these uniform or branded wines that characterize
their entire production.
The Facts
The statistical facts accompanying this setting are well known. Wine exports and
market shares have declined among the major old world producers, while the exports and
shares of the new world producers have increased. Figures 1 and 2 illustrate these
changes in market shares in volume terms for the major countries. Old world shares
including the core have declined. Export shares show sharp gains for the new world
producers. This transition has been described in greater depth and in considerable detail
by Anderson and Norman (2003). They report that while European producers account for
three-fourths of world wine production and almost all of wine exports, the new world
producers have begun to challenge this dominance. Between 1990 and 2001, the latter
group’s share grew from 4 to 18 percent in value terms (AN, Table 41). Excluding intra-
EU trade, the old world group’s share fell from 88 to 64 percent. The new world’s
increase is noteable, because it has occurred at the same time as a decline in world
production and consumption. During the 1990’s, global wine production fell 0.5 per cent
per year, while global wine trade increased by 5.2 percent per year in volume terms and
www.vdqs.net
Colloque Œnométrie XI
VDQS
Dijon 21/22 mai 2004
7.0 per cent in nominal $US terms (AN, Tables 11,15,37). Another aspect of these
changes is the growing importance of wine in total merchandise exports. For example
since 1990, this indicator has gone from 0.6 to 3.7 per cent for Chile and from 0.3 to 1.4
per cent for Australia (AN, Table 46).
Market concentration is fairly high with the top-ten wine exporting countries
accounting for more than 90 percent of world wine trade. This concentration is divided
about half-way between the old and new world countries. In descending order, they are
in value terms for 2001: France (41.5 per cent), Italy (17.4), Spain (9.6), Australia (6.4),
Chile (4.7), U.S. (3.7), Portugal (3.4), Germany (2.6), South Africa (1.6) and Argentina
(1.1), (AN, Table 124). The share of France is 10 percentage points lower than 1990,
which with smaller declines for Portugal and Germany have ensured that the shares of
Australia and other new world suppliers correspondingly have increased substantially.
Australia has gone from 6 to 23 per cent; the combined increase for Argentina, Chile,
New Zealand, South Africa and the United States is from 6 to 23 percent. Wine imports
are also highly concentrated. By 2001, one-half of all imports continued to be purchased
by three nations: UK (19 per cent), US (16) and Germany (14) (AN, Table 125).
Anderson and Norman also interpret these market changes employing two other
indicators. The wine trade volume (value) specialization index is defined as the ratio of
net bilateral exports (exports-imports) to the sum of bilateral exports plus imports, so that
the index measures between -1 and +1. Between 1990-2001, the value version of this
index has hovered about 0.84 for the core old world producers (AN, Table 90). For
Australia, the index rose from 0.71 to 0.88 during that period, for New Zealand from
-0.29 to 0.16, for Argentina from 0.40 to 0.83, and for South Africa from 0.86 to 0.93.
The index of comparative advantage in wine explains the share of a country’s exports in
total merchandise exports divided by the share of world wine exports in total world
merchandise exports. Between 1990–2001, this index increased only from 6.09 to 6.35
for the core old world wine producers (AN, Table 48). But for some of the new world
producers, the change has been astonishing. For Australia, the index rose from 1.29 to
6.26, for New Zealand from 0.48 to 0.67, for Chile from 2.56 to 16.46, and for the United
States from 0.11 to 0.31.
Trends or Cycles
Thus far all of this evidence supports the received view of recent market changes:
declining market shares for old world producers and increasing market shares for new
world producers. However, a more careful examination of the underlying export data
(AN, 2003) would suggest that we accept this view with caution. Further analysis occurs
with the regression results. Let us begin with aggregate measures. Figures 1 and 2 above
suggest that for both old and old core wine producers, market shares peaked somewhere
between 1987 and 1990 and declined afterwards. To appropriately test for any trends or
cycles in these shares, we thus examine the data beginning then until now. To simplify
our analysis, we also concentrate only on the four core producers and the four major new
world producers.
The levels of market shares and their percentage changes (growth rates) are
presented in Figures 3 and 4. Though downward trends appear, the shares are dominated
by cycles. This is true for the old world as well as the new world producers. When the
shares are transformed into fluctuations or percentage changes, in fact, only cycles are
witnessed with little evidence of trends. Such cyclical dominance is not surprising, given
www.vdqs.net
Colloque Œnométrie XI
VDQS
Dijon 21/22 mai 2004
the annual fluctuations that occur in grape and wine production. The sources of these
fluctuations, of course, are annual fluctuations in the underlying conditions, including
rainfall, temperature and sunlight at different times of year. Recall that the frequency and
amplitude of these fluctuations and their impacts on wine prices and producer earnings
have been sufficiently severe that the OIV has organized a special investigative
committee, Analyse Economiques et Conjoncturelle .
Another source of the market share fluctuations have been changes in demand and
prices generated by business cycle fluctuations (Labys, 2001). This is not surprising
given the increased globalization of the world wine market (Anderson, 2001). Earlier
Lindsey (1987) examined the impacts of exchange rates (and trade barriers) on the U.S.
wine industry. A study of French wine industry impacts was made by Mathis et. al.
(1997). Later Phares (2000) and Auzias (2001) examined the impacts of a wide range of
business cycles indicators on the wine industries of Australia, France, Germany, Italy,
Spain, the United Kingdom and the United States. At the same time the studies of
Wittwer and Anderson (2001), Anderson and Berger (1999), and Wittwer et.al. (2001)
employed much more precise (disaggregated) macro and industry variables to this effect.
An important business cycle indicator for wine consumption is per capita income,
as partially determined by such leading indicators as productivity, hours worked, and
wages. Income, along with expectations, interest rates, and other economic variables, are
key factors in global wine production and consumption. Some important questions
surround these variables: (1) Are national product, incomes and earnings likely to
influence domestic wine demands? Similar variables defined for wine importing
countries or global incomes might be useful for explaining the wine exports of a country.
(2) Do exchange rates (including devaluation effects) explain changes in wine exports or
wine imports of a country? And (3) do changes in interest rates cause increased liquidity
to finance new industry investments changing the capital stock or to store wines in
inventory?
An attempt has been made to decipher the relative importance of the various
trends and cycles in the market share data employing econometric regression analysis.
The market share variables of the respective countries and regions have been regressed
linearly against time to determine the significance of trends in the share data.
Interpreting the t-statistics on the time or trend variables indicates their positive or
negative significance. The regression errors reflecting cyclical movements about the
trend are further tested to evaluate the presence of first or second order lags. The
significance of second order lags provides some evidence though weak of the presence of
a cyclical component (Mills, 2003).
The results of the regression analyses have been illustrated initially in Figures 5
and 6, again divided between the core old world and new world producers. These figures
present the statistical test results for each of the equations and depict the fitted trends,
actuals and errors or residuals. Table 1 further summarizes these results as well as those
of the second order autoregressive tests.
Consider first the declining trends in market shares among the core producers.
Clearly France and Portugal display a negative trend coefficient that is statistically
significant according to the t-test. The shares of Italy appear to have a declining trend, but
no confirmed significance. In contrast the shares of Spain are increasing and significant.
www.vdqs.net

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika:

Zgłoś jeśli naruszono regulamin