|
|
|
|
The Contentious World of Agricultural Trade |
|
Jul
8th 2005, C.P. Chandrasekhar and Jayati Ghosh |
|
Agriculture,
including activities allied to it, has a rather divergent
role in economies across the world. To start with, there
is a great diversity in the share of agriculture as
a percentage of GDP, with the figure varying from 52
percent in Laos, for example, to just 6 percent in Korea.
Similar differences exist with regard to the share of
agricultural employment in total employment. Generally,
the lower the GDP, the higher is the significance of
agriculture in the economy. This also means that engaging
in world agricultural trade despite its distortions
would have the largest impact on the poorest countries
in the world.
However, the role of trade should not be exaggerated.
Though the global market for agricultural commodities
is estimated at $600 billion, the share of that market
serviced through cross-border trade is small. As Einarsson
has argued, there is a basic difference here between
staples and commercial plantation crops.[1] Much of
the world’s population obtains staples from domestic
production in their countries of domicile. Of the staple
foods, it is only in wheat that global trade is consistently
above 10 percent of total world production. And only
in a few of the typical plantation crops does global
trade represent more than 50 per cent of world production
(Table 1).
Table
1 >>
Overall, agricultural trade takes place largely between
developed countries, which account for about 70 percent
of both world exports and imports. Yet, there are three
reasons why the level and pattern of global agricultural
trade has important implications for developing countries.
To start with, exports of plantation crops and tropical
products of various kinds are crucial sources of foreign
exchange earnings for many developing countries. This
is the group that Tim Groser, the Chairman of the WTO’s
Committee on Agriculture describes as the developing
countries with ''offensive'' or ''export'' interests.
They ''either already have substantial economic interests,
relative to their economies, in world agriculture markets
and can build on this''; or, ''they can see a future
for themselves, once the massive trade distortions are
either eliminated or substantially reduced.''
The second reason why trade is important for some developing
countries is that they either depend on imports of staples
to meet domestic consumption or their currently prevalent
food security is threatened by agricultural trade liberalisation,
inasmuch as cheap imports can result in a sharp decline
in domestic production of staples or the promise of
quick profits may encourage rich farmers and agribusiness
interests to ensure or encourage production of more
profitable commercial crops.
Globally, not only is a small share of production of
food staples traded, but exports of food staples are
dominated by a very small group of countries which have
been described as "natural exporters", such
as Argentina, Australia, Brazil, Canada, New Zealand,
Uruguay, and the USA. These are countries where favourable
geographical conditions, sparse populations and a very
specific experience with colonisation, have encouraged
a large-scale and extensive agriculture that delivers
substantial surpluses of these staples. Production costs
here are far lower than elsewhere making them ''natural
exporters''. The only exception to these conditions
among global exporters of staples is Europe, where,
as widely recognised, state support to farmers has been
responsible for ensuring the availability of exportable
surpluses.
What is noteworthy is that only a few developing countries
that figure among the group of natural exporters are
significant exporters of grains or animal products.
They are Thailand (rice and poultry), Vietnam (rice),
Argentina (wheat, feed grains, soybeans, beef and milk
powder), Brazil (soybeans, beef and poultry) and Uruguay
(beef). The net result of this phenomenon is a peculiar
distribution of exporters and importers of food products
between the developed and the developing countries.
Over 80 percent of globally traded rice and wheat is
imported by developing countries, as is a sizeable portion
of feed grain and soybean exports. Few developing countries
import animal products. The one exception is powdered
milk, a low-value surplus product, of which 85 percent
of world trade goes to the South. And all the high-value
animal products such as beef, pork, poultry and cheese
are traded either between developed countries or from
South to North.
Thus in the realm of trade in staples developing countries
can be divided into three rather distinct groups: (i)
a small group of exporters, consisting of "natural
exporters" that can compete with developed countries
on the global markets for wheat, feedstuffs and animal
products and a few with higher population densities
and more traditional agricultural structures that are
also consistent net exporters, such as Thailand and
Vietnam; (ii) a large majority of developing countries
which belong to a group that thus far has been more
or less self-sufficient in food-though many of these
countries may more often buy food from global markets
rather than sell food in those markets, they are not
chronically dependent on food imports and most often
are in a position to finance those imports with foreign
exchange revenues from the export of other agricultural
products, typically tropical plantation crops; and (iii)
a significant number of net importers that are chronically
dependent on the world market for basic food supply-many
of which are also among the world’s poorest countries
(LDCs).
It is indeed true that over time developing countries
have been adjusting their export profiles depending
on trends in global trade. In response to declines in
the prices of tropical products, the middle-income developing
countries have reduced the share of tropical beverages
and raw materials-coffee, tea, cocoa, sugar, cotton
and tobacco-in their agricultural exports. That share
has fallen from 55 per cent in the 1960s to 30 per cent
by 1999-2001. They have turned to the export of more
high value food products-vegetables, fish, meat, nuts
and spices. These now account for more than 50 per cent
of their agricultural exports. However, the financial
and technological requirement for switching over to
higher value food crops are clearly beyond the resources
of farmers in the LDCs. Hence, the LDCs as a whole saw
their reliance on raw materials and tropical beverages
increase-from 59 per cent in the 1960s to 72 per cent
by 1999-2001. In the event, the pattern of trade dependence
of the kind delineated above has only got accentuated.
These features of agricultural trade imply that changes
in agricultural trade rules relating to staples can
affect developing countries in three different ways.
They can increase dependence on imports of staples of
developing countries, since countries that currently
do not import but access adequate supplies through domestic
production backed by protection and support measures
would find domestic production displaced by imports.
They could affect the access of the few developing country
exporters to global markets, by allowing for so-called
non-trade distorting support for developed country farmers
in the EU and US, which restricts imports into those
regions. They could raise the prices at which those
already dependent on imports of staples can access those
commodities. Thus there is reason to believe that only
the few developing countries that are competitive exporters
of staples would wholeheartedly support liberalisation
of trade in staples.
The other group of supporters of a more liberal trade
regime would be the exporters of tropical and non-food
products in the export of which they have an ''offensive''
interest. These countries could believe that better
market access and reduced domestic support in the developed
countries could helps increase the volume of their exports
as well as push up prices.
However, most developing countries would be wary about
agricultural trade liberalisation because of the threat
to livelihoods and food security that the process could
imply.
The Asian Experience
All these features of agricultural trade are illustrated
by the trade involved of countries in the Asia-Pacific
region. Asian countries are by no means dominant participants
in the world trade in agricultural products (Table 2).
Only one Asian country (China) features among the top
ten global exporters of agricultural products, at rank
9. In fact, Brazil and China are the only two developing
countries among the top ten. If we consider the set
of countries which each account for 1 percent of world
exports of agricultural products, they include the following
from the Asia-Pacific: China, Australia, Thailand, Malaysia,
Indonesia, New Zealand and India. Vietnam, Japan, Hong
Kong, Rep. of Korea, Chinese Taipei and Singapore have
world market shares between 0.5 percent and less than
1 percent.
Table
2 >>
The implication of this should be clear: Asia-Pacific
countries are by no means significant influences on
global trade and therefore their strength in global
negotiations is by no means substantial. But the reverse
is also true: few Asia-Pacific countries are dependent
on agricultural exports for their export revenues. Chart
1 plots countries according to their ranks as merchandise
exporters and agricultural exporters. If a country falls
along the diagonal (45 degree line) its rank as a merchandise
exporter tallies with its rank as an agricultural exporter.
Interestingly, this is not true of countries that hold
the high ranks in either category, excepting for China.
In the case of countries such as Japan, Korea, Hong
Kong and Singapore, they rank high as merchandise exporters
but not as agricultural exporters. That is, the truly
successful Asian exporters have not earned that success
based on agricultural exports, excepting for countries
like China. On the other hand, important agricultural
exporters such as Australia, Thailand, Indonesia and
New Zealand are by no means among the top merchandise
exporters in Asia.
Chart
1 >>
This could result in a substantial degree of difference
in the approach that countries in the Asia-Pacific region
adopt with regard to the trade negotiations-while exporters
like Australia, New Zealand, Thailand and Vietnam would
want substantial liberalisation of agricultural trade
involving all three pillars-market access, domestic
support and export competition-many others may be more
interested in limiting agricultural trade liberalisation
in order to protect domestic livelihoods and ensure
food security.
These features of the Asia situation are reflective
of differences among developing countries as a group
in the global negotiations, since the structures of
agricultural production and trade at the global level
are the same. Thus if there is still some degree of
solidarity in the developing country camp, it comes
from three sources: first, the desire of the developed
countries to protect their agriculture, while demanding
greater agricultural and non-agricultural market access
in the developing countries; second, the belief among
developing countries that there must be adequate safeguard
measures especially for ''sensitive'' products in order
to protect livelihoods and ensure food security; and,
third, the conviction that there must be a degree of
differential treatment for developing countries, especially
the poorest amongst them.
On the question of developed country protectionism,
the evidence is indeed overwhelming. The OECD report
on Agricultural Policies in OECD Countries: Monitoring
and Evaluation released in June 2005 states: ''There
has been little change in the level of producer support
since the late 1990s for the OECD as a whole. It has
fallen from 37 percent of farm receipts in 1986-88 to
30 percent in 2002-04, but this level of support was
first reached seven years ago in 1995-97.'' That, we
must recall was almost at the beginning of the implementation
of the Uruguay Round. This constancy in support has
been ensured in large part by ''box shifting'' or by
the substitution of support measures that are considered
trade distorting, by those that are supposedly not.
In 2004, the value of support to producers in the OECD
as a whole is estimated at USD 279 billion or EUR 226
billion. As measured by the percentage PSE, support
accounted for 30 percent of farm receipts, the same
level as in 2003. Including support for general services
to agriculture such as research, infrastructure, inspection,
and marketing and promotion, total support to the agricultural
sector was equivalent to 1.2 percent of OECD GDP in
2004.
This protectionism in the North has helped ensure a
degree of solidarity in the developing country camp.
Not surprisingly, G-33 Ministers who met in Jakarta
on 11 and 12 June 2005 to assess the progress of the
agriculture negotiations declared that: ''the problem
of food and livelihood security as well as rural development
constitute a concrete expression of developing countries’
right to development and therefore require a comprehensive
solution in all three pillars of the agriculture negotiations.''
They also ''reiterated that the concepts of Special
Products (SP) and Special Safeguard Mechanism (SSM)
as provided for in the July 2004 framework are fundamental
to any meaningful operationalization of Special and
Differential Treatment, and crucial for addressing food
and livelihood security as well as rural development
needs of developing countries. They emphasized that
SP and SSM are key policy instruments for securing the
survival of the vast number of small farmers and the
rural poor. Therefore modalities on this matter shall
be finalized by the Hong Kong Ministerial.''
It appears that Tim Groser, the Chairman of the WTO’s
Committee on Agriculture, is sensitive to these demands
of the developing countries. He states in his status
report on the agricultural negotiations released on
27 June 2005: ''Many developing countries, and particularly
LDCs, have deeply vulnerable people dependent on agriculture.
Integrating these parts of their agriculture sectors
into any emerging reform framework is deeply sensitive.
Such sensitivities have to be accommodated as the reform
process takes shape.''
However, in what is called the ''first approximation''
or the structure of agreement that has to be arrived
at on all three pillars by July 31, 2005 as a basis
for the ''political phase'' of discussions from September
to December, he feels that all features of the instruments
designed specifically to take account of the realities
of much developing country agriculture, such as SSM,
cannot be addressed.
Past experience suggests that this may not be just a
postponement, but a tactic to delay discussions on issues
on which the developed countries will not give in and
then demand that developing countries should be reasonable
and not derail agreement at the Ministerial on account
of such issues. If so, the limited solidarity within
the developing country camp must be used to stall progress
along lines that reproduces the inequities inherent
in the Uruguay framework.
[1]
Refer Einarsson, P., 2001 ''The Disagreement on Agriculture''.
Seedling 18.
|
|
|
|
|
|
|