market access and preferential trading schemes: evidence

FAO COMMODITY AND TRADE POLICY RESEARCH WORKING PAPER

No. 20

MMaarrkkeett aacccceessss aanndd pprreeffeerreennttiiaall ttrraaddiinngg sscchheemmeess:: eevviiddeennccee ffrroomm sseelleecctteedd ddeevveellooppeedd

aanndd ddeevveellooppiinngg ccoouunnttrriieess

PPiieerroo CCoonnffoorrttii** aanndd LLuuccaa SSaallvvaattiiccii****

JJuunnee 22000066

* Food and Agriculture Organization of the United Nations, Commodities and Trade Division [email protected] ** Università degli Studi del Molise, Via De Sanctis, 86100 Campobasso, Italy [email protected], consultant to FAO.

FAO Commodity and Trade Policy Research Working Papers are published by the Commodities and

Trade Division of the Food and Agriculture Organization of the United Nations (FAO). They are working documents and do not reflect the opinion of FAO or its member governments.

Also available at http://www.fao.org/es/ESC/

Additional copies of this working paper can be obtained from [email protected]

The designations employed and the presentation of material in this information product do not imply the expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the United Nations concerning the legal or development status of any country, territory, city or area or of

its authorities, or concerning the delimitation of its frontiers or boundaries.

All rights reserved. Reproduction and dissemination of material in this information product for educational or other non-commercial purposes are authorized without any prior written permission

from the copyright holders provided the source is fully acknowledged. Reproduction of material in this information product for resale or other commercial purposes is prohibited without the written

permission of the copyright holders. Applications for such permission should be addressed to the Chief, Publishing Management Service, Information Division, FAO, Viale delle Terme di Caracalla, 00100

Rome, Italy or by e-mail to [email protected].

© FAO 2006

i

ABSTRACT This report is aimed at analyzing the degree of protection faced by exporters in the EU, Japan, the United States, China, India and Brazil, and at identifying the contribution of product groups to the observed degree of market access. Data on the level of applied trade barriers provided by the MAcMap database is employed to compute Mercantilistic Trade Restrictiveness Indexes (MTRIs) on the basis of bilateral trade flows generated within a general equilibrium model framework. Results indicate that notwithstanding the rhetoric on trade preferences, developing country exporters appear to be still substantially restricted in their trade with some of the major developed country markets, such as the EU, Japan and the United States. Also, the three developing countries involved however impose significant restrictions on the access to their markets.

RÉSUMÉ Ce rapport a pour but d’analyser le degré de protection auquel sont confrontés les exportateurs dans l’UE, aux États-Unis, en Chine, en Inde et au Brésil et de déterminer la part des groupes de produits dans le degré d’accès au marché en question. Des données relatives au niveau des barrières commerciales appliquées provenant de la base de données MAcMap ont été utilisées pour calculer des indicateurs de restriction des échanges (Mercantilistic Trade Restrictiveness Indexes (MTRI)) sur la base des courants commerciaux bilatéraux créés dans un contexte de modèle d’équilibre général. Les résultats indiquent qu’en dépit des discours sur les préférences commerciales, les exportateurs des pays en développement continuent de rencontrer d’importantes restrictions dans leur commerce sur les principaux marchés de pays développés tels que l’UE, le Japon et les États-Unis. Cependant, il apparaît également que les trois pays en développement imposent d'importantes restrictions à l'accès à leurs marchés.

RESUMEN Este informe está orientado a analizar el grado de protección que enfrentan los exportadores en la UE, Japón, EE.UU., China, India y Brasil y a identificar la contribución de grupos de productos al grado observado de acceso al mercado. Los datos sobre el nivel de barreras que se aplican al comercio que proporciona la base de datos MAcMap se emplean para calcular los Índices Mercantilistas de Restricciones Comerciales (MTRI, por sus siglas en inglés) sobre la base de corrientes comerciales bilaterales generadas dentro de una estructura de modelos de equilibrio general. Los resultados indican que, sin perjuicio de la retórica relativa a las preferencias comerciales, los exportadores de los países en vías de desarrollo parecen seguir experimentando importantes restricciones comerciales en sus negocios con algunos de los principales mercados de los países desarrollados, como la Unión Europea, Japón y Estados Unidos. Además los tres países en vías de desarrollo mencionados, empero, imponen restricciones significativas al acceso a sus mercados.

ii

CONTENTS

ABSTRACT/RESUME/RESUMEN....................................................................................................... i 1 INTRODUCTION.......................................................................................................................... 1 2 A BRIEF REVIEW OF MAJOR TRADE POLICIES................................................................... 2 3 MEASURING THE EFFECTIVENESS OF PREFERENTIAL ARRANGEMENTS: THE MERCANTILISTIC TRADE RESTRICTIVENESS INDEX (MTRI) ................................................. 5 4 RESULTS....................................................................................................................................... 8 5 CONCLUDING REMARKS ....................................................................................................... 13 REFERENCES..................................................................................................................................... 14 TABLES .......................................................................................................................................... 16 FIGURES .......................................................................................................................................... 25 APPENDIX 1 ....................................................................................................................................... 35

Model and database ................................................................................................................. 35 APPENDIX 2 ....................................................................................................................................... 37

The 2004 baseline.................................................................................................................... 37

.

1

1 INTRODUCTION Despite the existing long tradition of studies which have dealt with the impact of tariffs, measuring accurately the magnitude of border protection has remained extremely difficult. The Uruguay Round Agreement has left in place a number of trade measures, such as the specific tariffs and tariff rate quotas still widespread in the agriculture of Japan and the European Union (EU), which maintain a strong disparity in tariff levels, and several tariff peaks. At the same time, the number of preferential trade agreements dramatically increased during the 1990s.

Preferential agreements are discriminatory policies entailing trade liberalization with respect to a subset of trading partners, which have been frequently utilized as an instrument for integrating the developing countries into the world trading system, on the basis of the assumption that this would promote their development. The world trading system is characterized by a wide variety of such agreements, which can be broadly categorized into two major types: reciprocal, entailing symmetric trade liberalization, and nonreciprocal, entailing asymmetric trade liberalization aimed at providing support to a country, which gains improved market access without being required to open up its own domestic market.

Several developed countries provide developing countries (DCs) with preferential access to their market, through a variety of agreements. Preferences granted on a non-reciprocal basis include the Generalized System of Preferences (GSP); the EU "Cotonou" agreement with African, Caribbean and Pacific (ACP) countries; and several regional United States of America (US) schemes such as the Africa Growth Opportunity Act (AGOA).

The GSP and the other Non-Reciprocal Preferential Regimes (NRPRs) rely on the concept of “Trade for Aid”. Since the 1970s, Trade for Aid has been thought to be a more effective way to promote development than the funding of projects, which generated rent seeking and whose impact often fell short of the expected benefits (Easterly, 2001). However, the effectiveness of preferences from a development standpoint has long been questioned, and evidence concerning the developmental effectiveness of preferences is mixed (Ozden and Reinhardt, 2003).

The purpose of this report is to shed light on the effects of the trade policies of a selected group of countries, including both selected developed and developing countries, with the aim of comparing the protection faced by each exporter, and to identify the products that contribute most to the observed differences in market access.

Two features of this study represent improvements over other attempts in the literature. Firstly, we use the detailed and comprehensive information on the level of applied trade barriers provided by the MAcMap database (Bouët et al., 2005a). Secondly, rather than relying on purely statistical measure of protection such as a (however weighted) average tariff rate, we assess the differential incidence of tariffs on the basis of a theoretically consistent measure of market access, the Mercantilistic Trade Restrictiveness Index (MTRI) (Anderson and Neary, 2003). Bilateral MTRIs are estimated on the basis of bilateral trade flows generated within a general equilibrium model framework.

The report includes five more sections. The next section summarizes some major trade initiatives recently undertaken by the three developed countries (the EU, the US and Japan) and the three developing countries (Brazil, China and India) considered here. Section 3 explains the approach followed to develop the MTRI index which is employed to compare the protection faced by different exporter in each market. Section 4 presents the results, and Section 5 concludes. Two brief appendixes are devoted to the description of the model and database employed and to the construction of the baseline.

2

2 A BRIEF REVIEW OF MAJOR TRADE POLICIES European Union

The EU is engaged in a web of preferential trade relations with other countries or regional groupings, which can be conceived as being made up of “concentric circles”, each one involving a different intensity of preferences, reciprocity and cooperation instruments. These range from the core integration among the 25 members, to the most distant circle where most-favoured-nation (MFN) treatment is applied according to the WTO rules. Between these two extremes there are the trade regimes applied to the European Economic Area (Norway, Iceland and Liechtenstein) and Switzerland; the Mediterranean partners (the Euro-Med agreements); the Africa, Caribbean and the Pacific (ACP – formerly the Lomé Convention, now the Cotonou Agreement) regime; the “Everything But Arms” (EBA) preferences for least-developed countries (LDCs); the bilateral free trade areas with Mexico, South Africa (SA) and Chile, plus the ongoing negotiations with Mercosur; and the Generalized System of Preferences (GSP) for developing countries which are not included in the previous categories.

The unilateral preferences granted by the EU for developing countries exports are regulated by two main trade arrangements:

a) The GSP scheme recently extended till 2008.1 The new EU GSP scheme, which was decided in April 2005, includes three categories of benefits: the General Scheme for all developing countries (with 40 percent of products receiving duty-free access, but with ceilings and graduation criteria that eliminate the largest exporters); the EBA initiative granting to the LDCs duty-free access on all products with the exception of arms and munitions; the “GSP plus”, providing duty-free access to all products from “countries with special development needs” which implement international conventions on the environment, and on human rights and labour standards. The EBA considerably improved the preferential market access granted to LDCs, though a significant limitation may be found in the absence of improvement in the field of rules of origin.

b) The Cotonou Partnership Agreement includes preferences and linkages between trade and financial assistance for the over 70 ACP countries, which are mostly former colonies of the EU member States. The agreements constitute the follow-up of a series of Yaoundé and Lomé Conventions which provided non-reciprocal trade benefits in 99 percent of industrial goods and some agricultural products. The Lomé preferences will last until 31 December 2007 (except for LDCs), after which reciprocity will be gradually introduced through new Economic Partnership Agreements (EPAs). While the GSP is conceived as a unilateral, unbound grant by industrialized countries, the Lomé/Cotonou preferences are an integral part of a broader international treaty which is legally binding upon the two parties and by which the EU has committed itself on a contractual basis to ensure non-reciprocal preferential market access conditions for ACP products.

Concerning the bilateral Agreements, those with Mexico, Chile, and South Africa (SA) provide for progressive mutual liberalization of goods and services, although free trade in agriculture and fisheries is not fully reciprocal and is limited to lists of products. In the case of SA, for example, EU is bound to offer duty-free access to 95 percent (only 62 percent in the case of agricultural products) of SA products by 2010.

Finally, the Euro-Mediterranean Agreements apply to 10 Mediterranean partners as agreed at the 1995 Barcelona Conference which launched the Euro-Mediterranean partnership with the goal of establishing a Free Trade Area by 2010. The Bilateral Euro-Mediterranean Association Agreements are a first step in this direction; some of these provide for non-reciprocal free access for non-sensitive products into the EU market and progressive liberalization for other products. 1 The GSP Scheme is an exception to the MFN principle and was introduced into the GATT in 1971: it allows GATT (and later WTO) members to grant unilateral preferences to products originating from developing countries.

3

United States of America

In the case of the US, a number of nonreciprocal preferential agreements are granted mainly to Latin American and African countries, either for developmental purposes, or with the aim of tackling specific economic and/or social problems of the regions. Moreover, a number of free-trade agreements are undertaken with Canada and Mexico – under the North American Free Trade Agreement (NAFTA) – and also with Australia, Chile, Israel, Jordan, and Singapore.

In general, the US follows the criterion of granting non reciprocal or unilateral preferential tariff treatment mostly to countries which qualify for the GSP. The US GSP programme provides for duty-free entry to all products covered by the scheme from designated beneficiaries. The scheme has been in operation since 1976, and since then it has always been renewed.2

A significant improvement in the US scheme was recorded in 1997, when 1 783 new products originating in LDCs were granted duty-free treatment. However, certain articles are excluded from the list of eligible products, and any article determined to be “import-sensitive” cannot be made eligible. Furthermore, the US scheme provides for ceilings for each product and country (“competitive need limits”), as well as for a “graduation” mechanism (Inama, 2006).

Other examples of non reciprocal developmental-based initiatives are the Caribbean Basin Trade Partnership Act (CBTPA), which was signed in 2000, with the aim of promoting the development of trade relations and the diversification of the small economies of the region. A similar initiative is the Andean Trade Preference Act, which was started in 1991 and expanded in 2002 under the Andean Trade Promotion and Drug Eradication Act, with the aim of combating drug production and trafficking in Bolivia, Colombia, Ecuador and Peru.

Another important initiative is the African Growth and Opportunity Act (AGOA), which was first signed in 2000, and subsequently extended in July 2004 to a time horizon up to year 2015. The initiatives involves 37 African countries, to which the initiative offers duty free access for most agricultural commodities, some subject to tariff rate quotas and quota free access, as well as a certain number of textile and oil products. Rules of origin require that products be grown, produced or manufactured in a beneficiary sub-Saharan African country, subject to a number of additional conditions related to national security, trade liberalization, and respect of human rights, which are reviewed on an annual basis. Altogether, LDCs are eligible for duty-free access with respect to the vast majority – 83 percent according to the US authorities - of the products listed in the US tariff schedule.

Concerning the free trade agreements, the implementation of the NAFTA started in 1994, and it has brought about a progressive elimination of trade barriers among the three countries involved. The agreement with Australia has been implemented since January 2005, and involves a progressive elimination of tariffs between the two countries over a period of 18 years as a maximum. Most tariffs on manufactured goods have been eliminated from the beginning, while exceptions include sugar and some dairy products on the US side, while Australia has agreed to maintain its quarantine system. With Chile, the agreement came into force in 2004, andinvolves the establishment of a free trade area over a 12-year period. The vast majority of industrial goods have been covered by the agreement since its inception – up to 85 percent - while duties on other products are due to be gradually phased out. Finally, a free trade agreement was negotiated in 2004 with Morocco, providing for the implementation since 1 January 2006, as a first step in a wider strategy involving other countries in the Middle Eastern region. Most tariffs have to be eliminated immediately, while 5 percent of the lines will be gradually phased out over a period of nine years.

Japan

Also for Japan, the preferential trade policy includes mainly the GSP system on the one hand, and a number of reciprocal regional agreements on the other. The Japanese scheme of generalized preferences was recently reviewed and extended for one more decade, until 31 March 2014. It

2 The latest renewal occurred in 2002 and officially reauthorized the scheme through December 2006.

4

provides selected preferences to 140 developing countries with reduction in the MFN duties for agricultural products according to a positive list approach. Most manufactured products are covered, and granted duty-free status subject to ceilings and maximum country amounts. On average, the degree of preference is higher for manufactured products than for agricultural ones, and a number of specific goods are excluded from the agreements such as dairy products, footwear and textiles and clothing. Preferential margins are relatively low also for certain manufactured goods, such as leather, rubber and footwear, as well as travel services.

LDCs enjoy a special treatment (duty-free entry, exemption from ceiling restrictions) for an extended list of products, which is deemed to account for 80 to 90 percent of the total import value from these countries. Similar benefits have been granted since 2003 also to Singapore, under the Japan-Singapore Economic Agreement for a New Age Partnership (JSEPA), which covers a number of agricultural and fishery products.

Concerning reciprocal agreements, the main initiative in which Japan is involved is the Asia-Pacific Economic Cooperation (APEC), which was started in 1989, and sees the participation of 21 member economies, including Asian countries but also the US and the Russian Federation. The APEC aims at establishing a free trade area and an open investment space in the Asia-Pacific region; the more industrialized economies are due to reach such target by year 2010, for while the developing economies which are participating in the initiative are due to achieve the target by year 2020. Since 2003, Japan has also initiated a structured dialogue with the Association of South-East Asian Nations (ASEAN), within the “Framework for a Japan–ASEAN Comprehensive Economic Partnership”, which involves regular consultations on the liberalization and facilitation of trade in goods, with the aim of negotiating a free-trade area by year 2012.

India

This country participates in a number of regional agreements, including the South Asian Association for Regional Cooperation (SAARC), whose major achievement has been the establishment of a South Asian Preferential Trading Arrangement (SAPTA) Treaty, signed in Islamabad in 2004. This arrangement includes provision for the progressive reduction of customs duties down to the level of 5 percent by year 2013. The countries involved are Sri Lanka, Pakistan, Bangladesh, Bhutan, Maldives, and Nepal. The Treaty also grants special treatment to the last four in the list, which are LDCs.

India signed a free trade agreement with Sri Lanka in 2000, involving a reciprocal progressive phasing out of tariffs. Bilateral tariffs are to be removed in three years in the case of India, and in eight years in the case of Sri Lanka. Commonwealth preferences are granted by India also to Mauritius, Tonga, and the Seychelles.

Together with Bangladesh, the Republic of Korea, the Lao People's Democratic Republic, Papua New Guinea, Sri Lanka, and China, in 2001 India signed the Bangkok Agreement, aimed at negotiating reductions in tariff and non-tariff measures within a number of structured rounds. The Indian Ocean Rim Association for Regional Cooperation, another initiative in which India has been involved since 1997, is also aimed at promoting economic co-operation, including trade facilitation, promotion and liberalization. The same themes are addressed also in another group in which India has been participating since 1997 – the BIMST-EC (Bangladesh, India, Myanmar, Sri Lanka, Thailand Economic Cooperation) which is aimed at dialoguing with the EU.

Brazil

The most important preferential trade initiative in which Brazil participates is Mercosur, a customs union including Argentina, Brazil, Uruguay and Paraguay. Through Mercosur, Brazil is participating also in a number of preferential trade relations with other countries of the region, as those involved in the Latin-America Integration Association (LAIA), i.e., Bolivia, Chile, Peru, Colombia, Ecuador, and Venezuela. Brazil also signed a bilateral agreement with Mexico in July 2002.

Outside the South American region, Brazil is undertaking negotiations with the United States on the Free Trade Agreement of the Americas (FTAA), and with the EU on a regional Association Agreement. Moreover, a framework agreement with South Africa was signed by Mercosur in

5

December 2000, with the goal of negotiating a free-trade agreement in which the other members of the South African Customs Union (SACU) have been participating since 2003.

China

China acceded to the WTO on 11 December 2001. As a consequence of accession talks, China has made efforts to improve market access by reducing barriers to trade in goods and services. While participation in the multilateral trading system forms a major part of China's trade policy, it has also stepped up efforts to participate in bilateral and regional trade negotiations. China is involved in a number of recent preferential trade initiatives, and is currently negotiating an increasing number bilateral concessions with a number of countries.

The Asia-Pacific region is the most important region for China’s economic and trading activities. China became a member of the Asia-Pacific Economic Cooperation (APEC) forum in November 1991 and supports its policy of "open regionalism".

China, along with Japan and the Republic of Korea, holds regular meetings with ASEAN under the ASEAN+3 framework for cooperation. A Framework Agreement on Comprehensive Economic Cooperation between China and ASEAN was signed on 4 November 2002, and entered into force on 1 July 2003. Under the agreement, both parties agreed to negotiate the establishment of an ASEAN–China Free Trade Area (ACFTA) within ten years.

The Bangkok Agreement entered into force in 1976 as a preferential trading arrangement between developing countries in the Asia-Pacific region. China acceded to the agreement on 12 April 2001, and started implementing concessions on 1 January 2002.

China is also undertaking negotiations on bilateral free-trade agreements. Its current agreements are very diverse, ranging from the Closer Economic Partnership Arrangements with Hong Kong and Macao, which are concrete, to the agreements with Australia and New Zealand that are mainly statements of intent.

Free trade agreements have also been signed with Chile and Pakistan. The agreement with Chile was signed on 18 November 2005: most tariffs are to be eliminated within five or ten years, with 97 percent of both countries' tariffs to be eliminated by 1 January 2015. On 5 Apri 2005, China and Pakistan signed an FTA-Early Harvest Agreement under which bilateral tariffs on certain products are to be eliminated gradually between 1 January 2006 and 1 January 2008.

Finally, it is worth recalling that unilateral special preferential tariffs are offered to LDCs for some products: the number of countries receiving such unilateral preferences increased to 39 in September 2005.

3 MEASURING THE EFFECTIVENESS OF PREFERENTIAL ARRANGEMENTS: THE MERCANTILISTIC TRADE

RESTRICTIVENESS INDEX (MTRI) The effectiveness of preferential trade schemes depends upon the way in which trade reacts to a complex set of rules, arising not only from the different existing regimes, but also from the heterogeneous tariffs implemented for different commodity specifications, and from a number of exceptions. In the case of bilateral protection indices, trade restrictiveness is the product of the structure of protection and the trade flows product specialization. Even if the importer country applied MFN bound tariffs to all exporters, the impact would be differentiated: trade would be more restricted in the case of countries exporting products facing the highest tariffs. As a consequence, the determinants of the extent to which specific preferences are effective can be summarized as follows: (i) composition of exports of the beneficiary country (e.g. primary versus processed); (ii) elements that are common to other trade schemes; and (iii) exceptions for specific commodities.

To measure the protection granted by a country’s trade policy regime, one needs to overcome two important aggregation hurdles: aggregation of different forms of trade policies and aggregation across goods with very different economic importance. Regarding the first aggregation problem, one needs to

6

bring all types of trade policy instruments into a common metric, in most cases an ad valorem equivalent.

The second aggregation problem is usually solved by calculating trade-weighted averages. As it is known, however, this would underestimate the protection effect, because of the endogeneity bias: actual trade is much lower with high tariffs than it would be with lower tariffs. On the other hand, using an “equivalence-based” index with a behavioural underpinning, the weights depend on import volumes evaluated at world prices (Anderson and Neary, 2005).

In order to follow a theoretically sound aggregation procedure, it is necessary to specify the type of information we want to maintain, so that the final number is equivalent to the original multiple data in the dimension we are interested in. According to Anderson and Neary (1996), a general definition of a policy index is as follows: depending on a pre-determined reference concept, any aggregate measure is a function mapping from a vector of independent variables – defined according to the policy coverage – into a scalar aggregate. The greatest advantage of this approach is that it is theoretically consistent, since the equivalence is determined according to a fundamental economic structure. Secondly, it provides unequivocal interpretation of the results, since the definition and properties of these “equivalence-based” indicators are predetermined.

The MTRI relies on the idea that trade policy can be evaluated using trade volume as the reference standard, by considering the extent to which trade distortions limit imports from the rest of the world. The policy aggregation procedure answers the following question: what is the equivalent uniform tariff that, if imposed to a country’s imports, would leave its aggregate imports unchanged?

Therefore a uniform tariff τµ, is defined, that yields the same volume (at world prices) of tariff-restricted imports as the initial vector of (non-uniform) tariffs. This can be expressed through the import demand functions M, holding the balance of trade function at level B0 constant, according to:

(1) ( )[ ] ( )0*000* ,,,1: BppMBpM =+ µµ ττ , with ( )µµ τ+≡ 1*pp .

where *p denotes a vector of international prices ( *kp ) of N goods k = (1,…,N), M0 is the value of

aggregate imports (at world prices) in the reference period, and p0 is the initially distorted price vector. Define the scalar import demand as

(2) ( ) ( )∑∑= =

≡r

c

N

k

mkckc BpIpBppM

1 1,

*, ,*,,

where mkcI , denotes the uncompensated (Marshallian) import demand function of good k from

country c. Accordingly, the MTRI uniform tariff τµ, would lead to the same volume of imports (at world prices) as the one resulting from the uneven tariff structure, denoted by the N-r bilateral tariffs matrix T whose elements are tc,k:

(3) [ ] [ ]∑∑∑∑= == =

=r

c

N

k

mkckc

r

c

N

k

mkckc BpIpBpIp

1 1

00,

*,

1 1

0,

*, ,,µ

The previous definition focuses on the overall distortion imposed by a country’s trade policies on its import bundle. However, we focus on the calculation of a bilateral version of the MTRI uniform tariff, in order to obtain the level of trade restrictiveness that the EU imposes on exports of country c. Accordingly, in equation (2) we only sum over k, rather than over k and c, in order to obtain a bilateral

uniform tariff MTRI ( µτ c ), defined as follows:

7

(4) ( )[ ] 00* ,1: cccc MBpM =+ µµ ττ ,

where

( ) ( )[ ]∑=

+≡N

kkckc

mkckcc BtpIpBppM

1

0,

*,,

*,

000 ,1*,,

is the value of aggregate imports (at world prices) from country c in the reference period.

In this bilateral protection index, trade restrictiveness is the product of the structure of protection and the trade flows product specialization. Even if the importing country were to apply MFN bound tariffs to all exporters, the impact would be differentiated: trade would be more restricted for countries whose exports face the highest tariffs. In order to show the structure of protection we compute a separate MTRI uniform tariff for each of the three main sectors in the economy: agriculture, industry and services. In order to compute the i-th sectoral MTRI uniform tariffs (

icµτ ), equation 4 is modified as

follows:

(5) ( )[ ] ( )[ ]∑∑==

+≡+Ni

kkckc

mkckc

Ni

k

ickc

mkckc

ic BtpIpBpIp

1

0,

*,,

*,

1

0*,,

*, ,1,1: µµ ττ ,

where Ni is the number of products included in each sector (Table 1).

Finally, in the standard definition prices are assumed to be fixed in world markets. Anderson and Neary (2003), argue (footnote 8) that “there is a rationale for a ceteris paribus trade restrictiveness index that fixes world prices even when these prices are in fact endogenous”. Such a rationale may be represented by the fact that, by keeping world prices constant, one focuses on the component of protection explained by national policies, rather that by the degree of market power of the country.

In our case, though, we need to recast the definition of the MTRI to make it consistent with the model used for its computation, since the GTAP model is global and calculates world prices endogenously. Therefore, in order to compute the MTRI taking into account the terms of trade impact, we need to redefine the uniform tariff equivalent relaxing the small country assumption. The vector of world prices p*, then, is a function of the tariffs T. To accommodate this, the definition of the MTRI [see equation (4)] is modified as follows:

(6) ( ) ( )[ ] 00* ,1: cwcc

wc MBTpM =+ττ ,

where ( wcτ ) is the bilateral MTRI uniform tariff with endogenous world prices.

In the case of τµ,, totally differentiating (1) to derive the effects of tariff changes, holding p* and B0 fixed, gives:

(7) µµµ

µ

ττ

pMdpMd

p

p00

1=

+.

With endogenous world prices ( ** tdpdtpdp += ), totally differentiating equation (1) we get

(8) ( )www

p

pww

p

pw

w

tpM

dpMpMdtpMd τ

ττ

Φ−+=+

0*00*0

1,

8

where *0

*

dpMdpM

p

wp≡Φ and φ is a correction factor which is needed because the import volume

function is evaluated at two different points (denoted by superscripts): the initial tariff-distorted price

vector p0 and the uniform-tariff-equivalent price vector ( )ww pp τ+≡ 1* . Comparing (7) and (8) it

appears thatτw could be either larger or smaller than τµ.

The computation of MTRI is performed through the definition of either a new variable, tr(r,s) which represents the product-generic tariff levied on imports from region r into region s (EU25, US, Japan, India, Brazil), or three variables corresponding to the sectoral indexes. Then we run the model, starting from our counterfactual baseline, assuming that all trade policies (i.e. tariffs and export subsidies) with respect to a specific region s are removed. In the closure, tr(r,s) is set as endogenous, while aggregate imports at world prices from region s into the EU are exogenized. Alternatively, we fix the three aggregate sectoral imports, endogenizing the respective uniform tariffs. In conclusion, we ask the model to compute the uniform tariff(s) that would eliminate all incentives to increase or decrease the volume of imports from the region/country under consideration.

4 RESULTS The results obtained for the MTRI bilateral uniform tariffs are presented in Table 1.

Total MTRI tariff values for the three developed countries, reported in the columns under the heading “tr”, show that the overall border protection is generally higher in Japan, and lower in the case of the US. For the other three countries, the most protected appears to be India, while China is the relatively more open. The low levels of Chinese protection can be related to the commitments that the country has accepted in the framework of its accession in the WTO in 2001.

Looking at the coefficient of variation of the MTRI uniform tariffs across partners, developed countries appear to be discriminating than developing countries, with Japan standing out as the most discriminating importer. Developing countries, on the other hand, show rather “flatter” protection structures, with India presenting the lowest value of the coefficient.

Table 2 summarizes the degree of reciprocity in market access between the six countries considered, while Table 3 shows the correlation between the bilateral MTRI uniform tariffs imposed and faced.

Figures 1 to 6 provide a graphical representation of the bilateral indexes imposed and faced by each country in different sectors. Since there are three sectors and five trade partners, each graph includes 15 points.

On the basis of the data reported in the six scatter diagrams, as we regress the bilateral uniform tariffs faced by a country on the tariff that such country imposes on its trading partners, in the case of the US, Brazil, and China we obtain a regression line (through the origin) which is steeper than the 45° line. This indicates that these countries face higher bilateral duties on their exports compared to those they apply on imports originating in the other five countries considered.

In the case of the US, at the aggregate level foreign trade barriers are always higher than domestic ones. At the sectoral level, there are only two cases where US exports face a lower protection than imports: services with the EU, and manufactures with Japan (Figure 1).

Brazil faces a higher protection on its exports compared to the protection it imposes on imports in its trade relations with the EU and India. Duties levied on Brazilian agricultural exports are often higher than the corresponding imports into Brazil; and the same applies in the case of services exported to India and Japan, and in the case of the Brazilian exports of manufactured goods to India (Figure 6).

In aggregate terms, the bilateral trade barriers that China imposes on imports turn out to be higher than the protection faced by exports only in the case of the EU and the US. Chinese exports face higher duties than the corresponding imports for all products, except for manufactured goods exported to the EU, the US and Japan, and for services exported to the EU and the US (Figure 5).

9

On the other hand, for the EU, India and Japan, the regression line through the origin always lies below the 45° line, indicating that these countries face lower bilateral duties on their exports compared to the ones they apply on imports from the other five countries considered.

Regarding the EU, barriers on imports are higher than those faced by exports in the Brazilian and US markets. In both cases, the overall result is driven by the differences in the agricultural protection levels (Figure 2).

Japan always imposes a higher overall level of protection compared to its trading partners, with the exception of Brazil. Considering individual products, however, it is worth emphasizing that exports of manufactured goods generally face higher trade barriers than those imposed on imports into Japan (Figure 3); the country shows a significant correlation between the duties applied on imports and those levied on exports.

India always imposes a higher overall level of protection compared to its trading partners, with the exception of Japan. Considering individual products, there is only one case – trade in services with the US – in which the tariffs levied on exports are higher than those applied on imports (Figure 4). India also presents a significant correlation between protection faced in foreign markets and that imposed on foreign imports (Table 3).

Comparing the bilateral protection indexes faced by the exporting countries or regions in the EU, Japan and the US markets, we notice that for only one group (“Rest of Asia”) is the American market the most protected; in two cases (“ASEAN” and Brazil) this happens with reference to the European market; all remaining exporters face the highest protection in the Japanese market.

There are apparent differences in the geographical structure of protection of the six countries. Table 4 presents the correlation and rank correlation (Spearman’s rho) between the results in Table 1.

As far as developed countries are concerned, the geographical pattern followed by US trade policy appears to be substantially different from the one followed by both the EU and Japan. On the other hand, the overall bilateral protection indexes of these two countries are significantly correlated: five countries or regions – India, “Rest of America”, Argentina, Australia and New Zealand, “ACP countries” – face high barriers both in the European and the Japanese merkets; while the exporters facing the highest barriers in the US market are “Rest of Asia”, “EU candidates” – including Bulgaria, Croatia and Romania, Turkey - and “Euromed countries”.

Also in the case of India, Brazil and China, a pairwise comparison shows statistically significant correlation indexes only between the last two, while the geographical distribution of India’s trade barriers appears to be rather different.

As expected, there is no overall similarity in the geographical distribution of protection between the three developed countries and the other three. Therefore, in Table 4 we report only a comparison between China and the US. The bilateral protection indexes of these two countries, as a matter of fact, appear to be somewhat related, despite the rankings of the exporters who face the highest tariffs in the two markets is quite different.

Although a number of asymmetric preferential trade initiatives have been implemented (see Section 2), our results show that they still face significant barriers to their exports. Figures 7 to 18 clearly show that for the six countries considered here there is no clear relationship between the economic size of the exporter – in terms of total or per capita GDP – and the overall level of protection imposed.

The overall bilateral MTRI uniform tariffs are negatively correlated both with the exporters’ GDP and their per capita GDP (Table 5): for the EU, the US and Japan, the higher the economic size of the trading partner, the lower the trade barriers imposed on its exports. In most cases, the absolute value of the correlation is quite low, and it is only significant for the US GDP per capita; however, an opposite relationship would have been expected as far as developed countries policies are concerned. Despite the fact that this finding is consistent with other recent results (Kee, Nicita and Olarreaga, 2005), it is certainly at odds with the rhetoric on preferential access for developing countries.

In this respect, it should be noted that a direct relationship between the level of the duties and the total and per capita GDP of the exporting country only appears in the case of the developing countries.

10

Even if it is not statistically significant, the trade policies of China, India and Brazil at least do not appear to be biased against other developing countries.

A common explanation for the lack of evidence about preferential treatment by the developed countries refers to the under-utilization of preferences, arising from the constraints attached in terms of rules of origin requirements, administrative costs as well as uncertainty on eventual eligibility. However, the most recent evidence points out that the rate of utilization is quite high, since only a very small proportion of the imports eligible for trade preferences is exported outside a preferential regime (Bureau, Chakir, Gallezot, 2006).

The flow of imports, nevertheless, is very limited especially from the LDCs, which leads the overall impact of the preferential agreements to be questioned. The main explanations seem to be beyond the scope of the preferences themselves, and a relevant role is certainly played by the ability of developing countries to comply with standards - sanitary, phytosanitary and technical - and the certification and traceability requirements of developed importers, in particular those imposed by the private sector. In this respect, there is evidence that the incidence of nontariff measures in a number of OECD countries tends to be disproportionately high on products that developing countries export – especially agricultural products (World Bank, 2005). An implication is that nontariff measures matter, as they tend to reduce the effective value of the preferential access granted through tariff exemptions.

Even if the MAcMap database only includes a few of all existing nontariff barriers (section 3), we do want to check the hypothesis that the higher MTRIs against low-income countries may reflect the product composition of imports – i.e., developing countries happen to export the goods most affected by protection (both tariff and nontariff) measures. To this end, we computed the MTRI uniform tariffs for individual products: results are reported under the headings “tr1”, “tr2”, “tr3” in Table 1.

As expected, protection levels for services (tr3), which were introduced into the model through the estimate of the ad valorem equivalents at a rather aggregate level, are quite homogeneous across exporters. For goods, agriculture (tr1) and manufactures (tr2), protection levels are much more uneven, as is shown by the values of the coefficient of variation (Table 1). Developed countries consistently show the highest levels of protection in agriculture. The same ranking of protection does not always apply in the case of the three developing countries and, in any case, the difference is not as large as in the developed countries.

It is also interesting to check the extent to which national trade policies are consistent in terms of preferences across sectors, by looking at the correlation and the rank correlations between the results (Table 6).

The geographical distribution of EU agricultural barriers appears not to be correlated with the access granted in the case of manufactures and services, either in terms of intensity or in terms of ranking. On the contrary, the protection granted to these two sectors seems to follow a common geographical pattern: bilateral values are directly related, and the countries that are granted preferential access for their manufactured products are the same that benefit from a better access also for services in several cases.

In the case of Brazil (Table 6), the two product groups showing a relatively more similar geographical distribution are agricultural products and services. Concerning India, the geographical distribution of barriers to trade in the services sector is significantly correlated to that of agriculture and to that of manufactures, while there does not seem to be any relation between the distributions of these last two product groups.

On the contrary, the results in terms of correlation and rank correlation for the US, Japan and China are not significantly different from zero (Table 6). This implies that the levels of protection faced by each exporter are independent. Since these countries’ trade policies determine a different geographical pattern of protection in each sector, the countries facing the lowest barriers in one sector are not necessarily the same benefiting from a preferential treatment in another.

In fact, the existence of preferential policies does not show up in the overall bilateral indexes of protection; this is mostly due to the distribution of import tariffs and export subsidies across activities. Most trade among OECD countries is in manufactures, while developed countries’ tariff profiles are

11

heavily biased against agricultural imports. As a consequence, developing countries still face an overall level of protection higher than developed countries, especially in the case of middle income countries, which do not benefit from preferences granted to the poorest countries.

Considering the agricultural MTRIs, the most restricted countries in their exports to the EU are developing countries such as Brazil and India, but also the ACP countries, a group enjoying a long-standing preferential access into the EU market within a quota system. The LDCs benefit from the implementation of the EBA initiative, while it is remarkable that the countries still waiting to be accepted as WTO members, such as Russia, also appear to enjoy quite a favourable treatment.

In the case of the US market, the higher barriers to agricultural exports are faced by non-EU countries, as well as by Latin-American countries – such as Brazil and Argentina - while Canada and Mexico clearly benefit from the implementation of the NAFTA. Finally, Japanese agricultural protection is relatively high vis-à-vis India, Oceania and China, while Middle-Eastern and North African countries appear to enjoy more favourable access conditions.

It is worth emphasizing three somehow unexpected results. Firstly, the fact that LDCs appear to enjoy a substantive preferential access only in the EU market. On this matter, however, it should be recalled that the elimination of all restrictions on LDCs’ exports following the implementation of the EBA initiative was explicitly introduced in the baseline.

Secondly, the countries which are not members of the WTO do not seem to be worse off in terms of the trade barriers they face; this finding may cast doubts on the (real) reasons for joining such a club. Also in this case, the result is indeed confirmed by other evidence (Rose, 2004), showing that countries belonging to the GATT/WTO do not show very different trade patterns compared to outsiders.

Thirdly, the ACP countries which are not included in the LDCs group face significant protection levels in all exporting markets considered. This is a particularly surprising result in the case of the EU, since, as already recalled, these countries benefit from one of the most generous trade preference scheme. However, despite a number of beneficiaries increasing over time, the share of EU imports from the ACP in total EU imports decreased from 6.7 percent in 1976 to 3.11 percent in 2002 (Manchin, 2005), and the European Commission itself expressed serious doubts on the benefits of the ACP preferential regime during the design of the Cotonou agreement (Bureau and Matthews, 2005a).

In order to gain more insights on the difficulties faced by some countries in accessing the agricultural markets of three of the most important developed countries, it is worth recalling that a significant share of imports are MFN duty free in the QUAD countries (Canada, EU, Japan, US). On the other hand, all of the QUAD countries deny preferential access on some imports subject to positive MFN duties (Low et al., 2006).3 In the following, we show the contributions of the most relevant agricultural products to the total agricultural uniform tariff for each importing country.

Table 7 confirms what are the “usual suspects” for the European protection. In the case of Brazil, the main difficulties are with the meat sector; while the dairy sector is responsible for most of the protection imposed on Australia and New Zealand agricultural exports, and “vegetable oils & fats” (such as olive oil) raises the largest difficulties for the Middle East and North Africa region.

As far as developing countries are concerned, the most problematic product is certainly sugar, which accounts for almost two thirds of the EU protection towards the ACP countries. These countries are subject to tariff rate quotas granting a preferential access up to a certain volume of imports. However, if the quota is binding, the level of protection reported in the database employed equals the out-of-quota tariff rate (Section 3). Accordingly, the high level of protection denoted by the MTRI indicates

3 For example, in the case of the EU, agro-food imports under a non zero MFN duty from developing countries represent 18 percent of total agro-food imports (Bureau and Matthews, 2005b).

12

that sugar regime “at the margin” is quite constraining and the exporters may benefit from a liberalization in this sector4.

Table 8 shows that Japanese agricultural protection is much more concentrated in terms of products. Processed rice is by far the most protected product, with protection spread over several potential exporters. Wheat from Australia and New Zealand, and from Canada, meat products from Mexico and Chile, and sugar from the ACP countries and South America are the other most protected goods.

As far as US agricultural protection is concerned (Table 9), dairy products stand out as the more protected ones across several exporters. ACP and Latin American countries face the most significant protection in the case of sugar; while the processed food sector is the most protected in the case of Asian countries, such as Japan or China.

Evidence for India shows that the most restricted agricultural exporters in this market are some of the Asian countries (Table 10) – particularly the ASEAN group and Japan – as well as South America and the EU. The most protected agricultural products are by far vegetable oils and fats, followed by the group of “vegetables, fruit & nuts”.

In the case of China, the highest agricultural duties are levied against the EU, ASEAN and Chile, followed by India and the US (Table 11). Considering individual products, the protection structure appears to focus on processed food, especially for meats.

Finally, Brazilian agricultural protection is higher towards Asian countries (“Rest of Asia”), followed by Oceania (“Australia & New Zealand”), but also Latin American countries, such as Mexico and Chile, rank quite high (Table 12). Also in this case, processed foods stand out as the most protected products, followed by “other crops” and “vegetable oils & fats”.

To conclude this long array of results, it is useful to compare the values of the MTRI with a more traditional indicator of protection, the import-weighted average tariff; we run this comparison for agricultural goods because these are reported in details in our database aggregation, and because these are the most relevant goods in terms of protection. The correlation and rank correlations between the 20 bilateral results obtained for each of the six importers considered is reported in Table 13.

The MTRI uniform tariffs are positively correlated with the trade-weighted averages, but the two indexes apper to yield similar results only when protection is rather low, as is the case for the US, China and Brazil. However, correlation is weaker in the case of the EU, and it is not statistically significant in the case of Japan. For India, a high and significant correlation arises between the two indexes, even for high levels of protection.

In general terms, these results are in line with the findings of Anderson and Neary (2003) and Bach and Martin (2001), showing that the trade-weighted average tariff is a linear approximation to the tariff aggregator based on the expenditure function. Anderson and Neary (2003) also prove that the MTRI uniform tariff is more likely to be higher than the trade-weighted average, the more elastic is the demand for the tariff-constrained imports.

In order to explore how trade-weighted averages are associated with MTRI tariffs we run a simple regression of MTRI uniform tariffs on weighted averages in our sample of 60 bilateral observations regarding agriculture (Figure 7). Our results are consistent with the finding of Anderson and Neary: the trade-weighted average significantly underpredicts the MTRI uniform tariff, with large differences in the case of Japan and the EU. Although the MTRI uniform tariff and the trade-weighted index tend to move together on average, it should be emphasized that a purely statistical average does not provide a reliable approximation, especially when there are very high tariffs.

4 This result, which may be credible for the aggregated country group included in this analysis, needs to be detailed further by (i) considering individual ACP countries, whose situation varies considerably depending on the production costs and production scale, (ii) by considering the overlapping between the ACPs and the LDCs which are going to be involved in the EBA initiative, and (iii) by modelling more explicitly the Tariff Rate Quotas available to each country.

13

5 CONCLUDING REMARKS The objective of this paper is to provide estimates of trade restrictiveness indices for six countries, including Brazil, China and India among developing countries, the EU, Japan and the US among the developed ones. In order to assess the overall level of protection we choose the MTRI, which is the uniform tariff equivalent of the different trade policy instruments observed for a country that would generate the prevailing level of trade. The calculated index captures both tariff and AVEs of NTBs at the tariff line level included in the MAcMap-HS6 (version 1) database. The computation was carried out using a modified version of the GTAP static model.

The MTRI uniform tariffs are computed bilaterally, with the objective of measuring the restrictiveness of trade policy regimes vis-à-vis developing and developed countries. They are also estimated for the broad disaggregates of agriculture, manufacturing and services products.

The main conclusion of this study can be summarized as follows:

• Notwithstanding the rhetoric on trade preferences, main developing countries exporters appear to be still substantially restricted in their trade with the EU, Japan and the US.

• Even considering a high level of aggregation - higher than in the recent literature on the relationship between per capita incomes of trading partners and tariff rates (Clark and Bruce, 2006) - our results indicate that tariff rates are on average lower for the poorest and richest countries, and higher for countries located in the middle of the income distribution, such as Brazil, India or Argentina.

• The structure of protection across sectors appears to be fairly homogenous among the EU, Japan, and the US. In particular, agricultural products originating in the largest exporters among the developing countries appear to face an overall higher degree of protection compared to developed country exporters, such as Australia and New Zealand.

The results also confirm that both developed and developing countries tend to discriminate significantly across products and sectors, as a consequence of the various existing preferential schemes, as well as of tariff peaks. Several different reasons can be given for the fact that trade policies of high-income countries result in higher MTRIs on the imports from low income countries; this evidence reflects both the product composition of developing countries’ exports, and the preferential regimes that they benefit from. Many developing countries, as a matter of fact, are specialized in highly protected product worldwide, in particular among agricultural products, and in textile and clothing. It should also be recalled that developing countries exhibit low export unit values, and this leads to higher ad valorem equivalents.

Recent studies have concluded that protection levels are positively related with the exporters’ GDP (Kee, Nicita and Olarreaga, 2005). Our results do confirm these findings, and call for more careful considerations of the effective value of preferential trade schemes for the promotion of growth and development. When protection is assessed using a theoretically sound methodology, the actual tariff structures may still represent a serious limit to growth prospects of several developing countries.

Estimates provided in this study only pertain to the preference schemes included in the database we employ, the MAcMap, with the addition of a few schemes that are going to be implemented in the next future, such as the EBA initiative. When all of the preferential trade initiatives mentioned in Section 2, will have been implemented, the actual protection faced by developing countries, or at least some of them, may be significantly lower.

14

REFERENCES Anderson, J. & Neary, J. (2003). The Mercantilistic Index of Trade Policy, International Economic

Review. 44(2). pp. 627-649.

Anderson, J. E. & Neary, P.J. (2005). Measuring the restrictiveness of international trade policy, The MIT Press, Cambridge MA-USA, 2005.

Armington, P.A. (1969). A theory of demand for products distinguished by place of production. International Monetary Fund Staff Papers 16.

Bach, C.F., Frandsen, S.E. & Jensen, H.G. (2000). Agricultural and economy-wide effects of European enlargement: modelling the Common Agricultural Policy. Journal of Agricultural Economics, 51 (2). , pp. 162-180.

Bach, C.F. & Martin, W. (2001) Would the right tariff aggregator please stand up? Journal of Policy Modelling. 23(6). 621-635.

Balassa, B. (1965). “Tariff Protection in Industrial Countries: An Evaluation”, Journal of Political Economy, 73, 6.

Balassa, B. (1967). The impact of the industrial countries tariff structure on their imports of manufactures from less developed areas. Economica, 34.

Brockmeier, M., Herok, C.A. & Salamon, P.B. (2001). Technical and institutional changes in an Enlarged EU: welfare effects for old and new members with a focus on the agri-food chain. Paper presented at the IV Annual Conference on Global Economic Analysis, Purdue University, West Lafayette, Indiana, USA, 2001.

Bouët, A., Decreux, Y., Fontagné, L., Jean, S. & Laborde, D. (2005a). Tariff Data, Documentation GTAP Database version 6, draft.

Bouët, A, .Bureau, J.-C., Decreux, Y. & Jean S. (2005b). Multilateral agricultural trade liberalization: the contrasting fortunes of developing countries in the Doha Round. The World Economy, Blackwell Publishing, vol. 28(9).

Bouët, A., Fontagné, L. & Jean, S. (2005c). Is the erosion of preferences a serious concern? Working paper, CEPII, Paris.

Bureau, J.-C. & Matthews, A. (2005a). The consequences of agricultural trade liberalization for developing countries: distinguishing between genuine benefits and false hopes. Discussion Paper, 73, IIIS.

Bureau, J.-C. & Matthews, A., (2005a). EU Agricultural policy: what developing countries need to know. Discussion Paper, 91, IIIS.

Bureau, J.-C. & Salvatici, L. (2004). WTO negotiations on market access: what we know, what we don't and what we should. In Agricultural policy reform and the WTO: where are we heading? G. Anania, M.E. Bohman, C.A. Carter and A. F. McCalla (eds.). Cheltenham, UK & Northampton MA, USA: Edward Elgar Publishing.

Corden W.M. (1971). The theory of protection, Clarendon Press, Oxford. Chapters 1,2,3,4 and 6.

Candau F. & S. Jean. (2005). What are EU trade preferences worth for sub-Saharan Africa and other developed countries? TRADEAG Working Paper, 05/09.

Clark, D.P. & Bruce, D. (2006). On the incidence of US tariffs. The World Economy, 29, 2, 2006.

Eastwood, R., Lipton, M. & Newell, A. (2004). Farm size. Paper prepared for Volume III of the Handbook of Agricultural Economics, University of Sussex.

Francois J. (1998). Scale economies and imperfect competition in the GTAP model. GTAP Technical Paper No. 14, 1998.

15

Hertel, T. W. (1997). Global trade analysis. Modeling and Applications, Cambridge University Press.

Hoekman, B. & Ozden, C. (2005). Trade Preferences and differential treatment of developing countries: a selective survey. Policy Research Paper, 3566, World Bank.

FAO, (2005). Review of basic food policies 2003-2004, Rome, Commodities and Trade Division.

Inama, S. (2006). Erosion of trade preferences and Aid for Trade initiative: towards a new paradigm? Briefing Papers, Friedrich Ebert Stiftung, January.

Kee, H. L., Nicita, A. & Olarreaga ,M. (2005). Estimating Trade Restrictiveness Indices, mimeo.

Low, P., Piermartini, R. & Richtering, J. (2006). Non-reciprocal preference erosion arising from mfn liberalization in agriculture: what are the risks? Staff Working Paper ERSD-2006-02, WTO Economic Research and Statistics Division.

Manchin M. (2005). Preference utilization and tariff reduction in European Union imports from Africa, Caribbean, and Pacific countries. World Bank Policy Research Working Paper, 3688.

OECD, (2003). Regionalism and the Multilateral Trading System, Paris.

Park, S.-H. Measuring tariff equivalents in cross border trade in services. KIEP Working paper, 02-15, 2002.

Rose, A.K. (2004). Do we really know that the WTO increases trade? American Economic Review, 94(1).

van Meijl, H. & van Tongeren, F. (2002). The Agenda 2000 Cap reform, world prices and Gatt–Wto export constraints., European Review of Agricultural Economics, 29, 2002.

World Bank, (2005). Globa1 Monitoring Report, Washington D.C., USA.

16

TABLES Table 1 - Uniform tariff (MTRI)* of United States, with respect to selected countries/areas

United States Japan EU 25

tr1 tr2 tr3 tr tr1 tr2 tr3 tr tr1 tr2 tr3 tr

Rest of Europe 9.8 0.8 22.9 4.0 13.3 0.8 28.3 8.2 9.3 0.0 21.6 1.8

EU candidates 11.8 3.6 20.1 11.3 9.1 2.4 23.1 16.7 10.4 -0.7 19.2 1.8

LDC 1.8 3.4 24.1 3.9 62.9 0.2 25.5 15.9 0.0 0.0 20.1 0.8

ACP 9.4 4.6 21.1 6.1 105.5 0.3 23.8 25.3 45.2 0.2 18.7 11.4

Australia & New Zealand 5.5 1.2 22.4 6.1 121.7 0.2 24.1 26.3 15.7 0.6 19.6 9.5

Euromed countries 2.4 2.4 21.3 9.0 7.9 3.9 22.6 15.0 39.1 0.0 18.1 5.9

China 2.6 3.7 38.5 5.5 92.5 3.2 32.4 10.7 25.9 3.5 25.5 7.5

ASEAN 2.8 2.6 25.3 3.4 63.1 0.7 30.4 6.5 15.1 2.9 27.8 7.3

Rest of Asia 0.7 12.3 26.4 12.5 6.8 4.7 27.7 9.3 5.7 7.8 22.0 9.0

Rest of Latin America 3.0 4.4 17.8 4.4 129.4 0.4 22.4 65.9 35.8 0.0 16.7 15.9

Japan 3.5 1.6 24.8 2.0 - - - - 10.2 3.3 26.1 6.1

India 1.2 3.9 25.5 4.9 151.8 1.4 35.2 72.7 48.8 4.1 26.7 17.6

Canada 1.2 0.0 19.2 0.0 81.9 0.5 27.7 36.2 7.5 1.0 21.7 8.8

United States of America - - - - 67.7 0.2 25.4 16.5 12.1 1.6 21.4 7.9

Mexico 0.3 0.0 22.7 0.0 54.0 1.7 23.2 34.6 8.4 0.1 19.6 5.1

Argentina 9.3 1.5 15.5 4.6 57.1 0.3 21.3 34.7 13.1 1.4 16.9 10.8

Brazil 9.1 2.2 23.5 3.7 13.1 0.4 32.9 6.8 52.2 0.8 25.6 28.9

EU 25 4.0 1.3 23.7 5.1 69.0 1.8 27.8 18.3 - - - -

No WTO 2.4 1.1 23.9 2.0 19.6 0.2 27.1 2.1 5.7 0.7 21.6 2.1

Turkey 7.8 6.3 25.7 9.8 4.5 2.0 22.3 15.9 23.1 0.2 19.8 8.1

Chile 1.8 1.2 21.5 2.2 35.5 0.1 23.4 15.6 9.4 0.1 18.7 3.3

Coefficient of variation 0.80 0.95 0.20 0.68 0.77 1.08 0.15 0.82 0.81 1.50 0.16 0.77

Source: GTAP simulation

*tr: overall MTRI; tr1: MTRI agricultural sector; tr2: MTRI industrial sector; tr3: MTRI services sector

17

Table 1 (continued)- Uniform tariff (MTRI)* of United States, with respect to selected countries/areas

India China Brazil

tr1 tr2 tr3 tr tr1 tr2 tr3 tr tr1 tr2 tr3 tr

Rest of Europe 24.2 27.7 36.8 29.2 2.5 5.8 27.1 8.7 6.2 9.9 27.8 12.6

EU candidates 26.1 29.0 36.8 30.2 2.3 1.4 24.4 8.6 2.7 6.7 26.0 14.4

LDC 26.9 24.2 37.2 26.1 2.7 0.8 24.7 1.7 10.3 0.2 26.5 1.2

ACP 21.8 33.1 37.3 33.0 4.7 1.0 24.7 3.7 11.1 7.3 26.0 9.8

Australia&New Zealand 20.3 25.9 37.9 27.6 4.7 2.5 24.4 3.8 16.2 2.0 28.0 5.5

Euromed countries 6.9 17.4 37.1 17.5 2.9 0.5 24.4 7.0 7.6 1.4 25.4 3.9

China 21.6 27.7 38.7 28.1 - - - - 8.6 13.9 29.5 16.6

ASEAN 77.5 18.9 36.3 27.0 8.4 3.1 24.8 3.8 11.3 13.6 30.1 15.3

Rest of Asia 21.0 26.2 37.6 25.7 2.5 6.4 27.2 6.0 25.7 14.4 29.0 17.3

Rest of Latin America 36.1 15.01 36.01 15.60 4.9 1.18 24.6 4.87 1.6 3.72 21.9 4.92

Japan 55.6 29.2 37.0 30.2 3.4 4.4 22.3 4.6 8.1 13.5 29.1 14.7

India - - - - 5.8 4.0 25.0 5.6 12.1 7.5 31.3 10.8

Canada 45.0 19.6 36.5 29.1 4.8 3.6 27.1 6.0 9.1 6.3 28.3 11.2

United States of America 29.9 24.2 37.1 27.1 5.6 4.4 25.7 6.9 11.5 9.2 25.9 10.6

Mexico 36.9 15.2 36.4 15.3 1.5 0.7 26.4 4.3 15.5 15.9 29.3 16.8

Argentina 55.2 15.2 35.5 51.7 2.7 1.1 24.2 2.9 0.0 6.4 20.8 5.7

Brazil 43.0 27.6 37.0 38.4 4.4 1.0 27.4 2.9 - - - -

EU 25 46.8 28.0 35.8 29.8 8.7 5.0 25.9 8.2 11.9 10.7 27.4 13.5

No WTO 31.4 22.0 36.8 22.7 3.9 3.7 25.5 4.2 9.0 5.1 28.5 7.1

Turkey 28.7 27.3 36.6 29.9 1.6 2.1 23.2 8.4 11.2 14.8 25.7 17.9

Chile 35.2 5.4 35.8 6.7 8.4 0.8 24.7 2.5 12.4 5.9 28.1 6.8

Coefficient of variation 0.46 0.29 0.02 0.34 0.51 0.70 0.05 0.41 0.55 0.57 0.09 0.46

Source: GTAP simulation

*tr: overall MTRI; tr1: MTRI agricultural sector; tr2: MTRI industrial sector; tr3: MTRI services sector

18

Table 3 - Correlation coefficient between protection faced and imposedCountries r

EU25 0.301Japan 0.492**USA 0.285Brazil 0.408India 0.527*China 0.244

Source: GTAP simulation* significant at 0.05 ** significant at 0.10

r rhoDEVELOPED COUNTRIES

EU-Japan 0.409** 0.451**EU-US -0.034 0.183

Japan-US -0.189 0.018DEVELOPING COUNTRIES

Brazil-India 0.096 0.227Brazil-China 0516* 0.445**India-China 0.084 0.192

DEVELOPED-DEVELOPING

US-China 0.438** 0.346Source: GTAP simulation * significant at 0.05 ** significant at 0.10

Table 4 - Correlation coefficient and rank correlation coefficient among MTRI uniform tariffs

Table 2 - Reciprocal market access between USA, EU and Japan

Exporter/importer EU25 Japan USA Brazil India China

EU25 - 18.3 5.1 13.5 29.8 8.2

Japan 6.1 - 2.0 14.7 30.2 4.6

USA 7.9 16.5 - 10.6 27.1 6.9

Brazil 28.9 6.8 3.7 - 38.4 2.9

India 17.6 72.7 4.9 10.8 - 5.6

China 7.5 10.7 5.5 16.6 28.1 -Source: GTAP simulation

19

Total GDP Per capita GDP

EU -0.052 -0.100

US -0.178 -0.396*

Japan -0.118 -0.018

Brazil 0.142 0.086

India 0.071 0.147

China 0.299 0.224Source: GTAP simulation * significant at 10%

Table 5 - Correlation coefficient between MTRI uniform tariffs and GDP values

Table 6 - Correlation coefficient and rank correlation coefficient between EU bilateral MTRI uniform tariffs in different sectors

r rho r rho r rho

tr1-tr2 -0.035 0.076 tr1-tr2 -0.346 -0.343 tr1-tr2 -0.045 0.108

tr1-tr3 0.141 -0.066 tr1-tr3 0.211 0.243 tr1-tr3 -0.285 -0.297

tr2-tr3 0.0563* 0.676** tr2-tr3 0.080 0.097 tr2-tr3 0.271 344.000

r rho r rho r rho

tr1-tr2 0.142 0.179 tr1-tr2 0.346 0.302 tr1-tr2 -0.215 -0.215

tr1-tr3 0.047 0.142 tr1-tr3 0.605** 0.475** tr1-tr3 -0.563** -0.731**

tr2-tr3 0.324 0.307 tr2-tr3 0.412 0.431 tr2-tr3 0.570** 0.483*

*tr1: MTRI agricuoltural sector; tr2: MTRI industrial sector; tr3: MTRI services sector** significant at 0.01 Source: GTAP simulation

EU Japan US

China Brazil India

20

Table 7 - Decomposition of EU agricultural MTRIs by sector

Uniform tariffs ACP Argentina ASEAN CHINA India Brazil Japan LDCEuromed countries

NoWTO USARest of

EUEU

candidatesAustralia&Ne

w ZeelandRest of

AsiaCanada Mexico Turkey Chile

Rest of Latin

AmericaWeighted average 15.1 11.3 11.7 15.0 7.8 20.4 7.3 0.0 10.4 4.3 8.8 6.3 11.0 9.2 4.3 6.1 4.7 12.4 9.0 20.8

MTRI 45.2 13.1 15.1 25.9 48.8 52.2 10.2 0.0 39.1 5.7 12.1 9.3 10.4 15.7 5.7 7.5 8.4 23.1 9.4 35.8paddy rice 0.0 0.0 0.8 0.1 2.5 0.0 0.2 0.0 0.0 0.2 0.6 0.0 0.0 0.2 1.0 0.0 0.0 0.0 0.0 0.1wheat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.0 0.2 0.0 0.1 0.0 0.0cereal grains 0.0 0.7 0.0 0.1 0.0 0.2 0.0 0.0 0.0 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1oilseeds 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0vegetables 2.3 1.4 0.1 7.5 0.1 0.2 0.1 0.0 2.3 0.5 0.4 0.0 0.5 0.6 0.2 0.1 1.0 0.4 4.6 14.3sugar cane 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0sugar 26.3 0.1 0.1 0.1 1.1 1.4 0.1 0.0 0.1 0.3 0.1 0.1 0.1 0.0 0.1 0.0 4.2 0.6 0.0 14.5milk 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0oils&fats 0.0 0.0 0.8 0.0 0.1 0.0 0.2 0.0 15.6 0.2 0.1 0.0 0.3 0.0 0.0 0.0 0.0 4.3 0.0 0.0cattle 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0fibers 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0crops 0.1 0.2 0.5 0.2 0.4 0.5 0.5 0.0 0.1 0.4 1.6 0.1 0.3 0.0 2.5 0.4 0.3 0.0 0.0 0.0animal products 0.0 0.0 0.1 0.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.1 0.0 0.0 0.0wool 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0forestry 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0fishing 0.1 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0meat 15.2 5.0 0.1 0.0 37.4 44.4 0.5 0.0 20.1 0.7 0.8 2.2 1.7 0.9 0.0 0.7 0.0 10.7 0.0 6.2meat products 0.1 0.7 5.7 0.6 0.2 2.5 1.2 0.0 0.0 0.4 1.7 0.2 1.0 0.2 0.0 0.6 0.9 4.2 1.4 0.0beverages&tobacco 0.4 0.1 0.2 0.3 0.1 0.0 0.3 0.0 0.1 0.1 0.6 0.7 1.0 0.4 0.1 0.1 0.2 0.0 1.3 0.0food 0.8 4.9 3.1 2.6 1.1 3.2 3.3 0.0 0.3 1.5 3.6 2.0 1.5 0.4 0.8 2.8 1.7 0.2 1.8 0.3dairy 0.1 0.1 0.0 0.0 0.0 0.0 0.6 0.0 0.1 1.4 0.7 4.7 0.7 12.8 0.0 2.5 0.1 2.5 0.1 0.0processd rice 0.0 0.0 3.8 14.2 5.9 0.0 3.3 0.0 0.9 0.3 1.8 0.0 0.0 0.1 1.1 0.0 0.0 0.1 0.0 0.5Source: GTAP simulation

21

Table 8 - Decomposition of Japan agricultural MTRIs by sector

Uniform tariffs ACP Argentina ASEAN Brazil India USA China LDC Euromed countries NoWTO EU25 Rest of

Europe EU candidates Australia&New Zealand

Rest of Asia Canada Mexico Turkey Chile

Rest of Latin


MTRI 105.5 57.1 63.1 13.1 151.8 67.7 92.5 62.9 7.9 19.6 69.0 13.3 9.1 121.7 6.8 81.9 54.0 4.5 35.5 129.4paddy rice 0.0 0.0 15.8 0.0 0.0 32.4 0.1 0.0 0.0 10.9 22.9 0.0 0.0 15.2 0.0 5.8 0.0 0.0 0.0 0.0wheat 61.8 0.0 0.0 0.0 25.4 3.5 0.0 0.0 0.0 4.7 0.7 0.0 0.0 25.4 0.0 36.2 0.0 0.0 0.0 0.0cereal grains 2.4 5.1 0.0 1.3 0.0 0.2 0.0 0.0 0.0 1.0 0.0 0.0 0.0 1.9 0.0 1.3 0.0 0.0 0.0 0.0oilseeds 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0vegetables 0.3 2.0 0.6 0.4 0.1 2.0 1.9 0.1 0.3 0.5 0.2 0.0 0.0 0.3 0.4 2.2 0.4 0.4 0.4 0.7sugar 39.4 35.8 5.4 4.5 6.0 5.3 0.0 1.9 1.0 0.7 4.2 0.2 0.0 12.7 0.3 0.0 0.0 0.0 0.4 1.3oils&fats 0.0 0.2 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0cattle 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.0crops 0.0 0.0 0.1 0.0 0.2 0.3 0.1 0.0 0.1 0.0 0.1 0.0 0.0 0.0 1.5 0.0 0.0 0.0 0.0 0.0animal products 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0wool 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0fishing 0.2 0.0 0.1 0.0 0.0 0.1 0.0 0.1 0.1 0.0 0.1 0.4 0.1 0.1 0.4 0.0 0.0 0.2 0.0 0.0meat 0.2 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 9.4 0.7 0.1 0.0 13.8 0.0 3.0 1.6 0.0 5.6 2.0meat products 0.0 0.2 1.2 3.0 0.0 3.6 0.0 0.0 2.1 19.0 28.9 5.1 3.5 1.2 0.0 32.2 51.6 0.2 16.0 0.1beverages&tobacco 0.3 1.8 0.4 0.1 0.1 0.2 0.1 0.1 0.1 0.6 1.8 0.1 1.6 0.2 0.1 0.2 0.1 0.2 0.5 0.1food 0.9 1.5 3.7 2.8 1.2 4.1 0.9 5.8 2.7 2.0 2.7 4.6 2.2 2.1 4.2 0.8 0.3 1.3 2.2 1.1dairy 0.0 10.5 5.9 0.4 0.3 0.3 0.0 0.0 13.2 1.5 5.9 2.8 1.7 4.6 0.0 0.2 0.0 2.2 9.5 0.1processd rice 0.0 0.0 29.9 0.0 118.4 40.6 59.8 0.0 0.0 17.5 1.0 0.0 0.0 44.2 0.0 0.0 0.0 0.0 0.0 124.0Source: GTAP simulation Table 9 - Decomposition of US agricultural MTRIs by sector

Uniform tariffs ACP Argentina ASEAN Brazil China India Japan LDC Euromed countries NoWTO EU25 Rest of


Rest of Asia Canada Mexico Turkey Chile Rest of Latin

America

Weighted average 5.7 7.0 2.4 6.9 2.6 1.1 3.2 2.3 2.3 2.0 3.3 6.7 7.0 4.8 0.7 1.1 0.3 6.7 1.5 2.3MTRI 9.4 9.3 2.8 9.1 2.6 1.2 3.5 1.8 2.4 2.4 4.0 9.8 11.8 5.5 0.7 1.2 0.3 7.8 1.8 3.0

paddy rice 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0wheat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0oilseeds 0.2 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0vegetables 0.0 0.1 0.0 0.1 0.3 0.2 0.0 0.0 0.4 0.1 0.1 0.0 0.1 0.0 0.1 0.0 0.1 0.2 0.5 0.1sugar 7.4 1.2 0.7 3.9 0.1 0.2 0.4 0.8 0.2 0.7 0.1 0.1 0.1 1.0 0.0 0.1 0.0 0.1 0.0 2.5oils&fats 0.0 0.3 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0crops 0.8 1.3 0.8 3.1 0.4 0.2 0.1 1.7 0.0 0.4 0.3 0.7 1.4 0.0 0.2 0.0 0.0 6.0 0.1 0.3meat 0.0 1.2 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 3.0 0.0 0.0 0.0 0.0 0.0 0.2meat products 0.0 0.0 0.1 0.1 0.2 0.0 0.1 0.0 0.0 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0beverages&tobacco 0.1 0.1 0.0 0.0 0.1 0.0 0.2 0.0 0.1 0.1 0.3 0.0 0.3 0.3 0.0 0.0 0.0 0.1 0.3 0.0food 0.7 1.4 1.0 1.2 1.5 0.4 2.5 0.1 0.6 0.9 1.1 1.5 0.6 0.2 0.3 0.9 0.2 0.1 0.5 0.2dairy 0.2 3.1 0.1 0.1 0.0 0.0 0.1 0.0 0.9 0.1 2.2 7.6 9.1 1.0 0.0 0.3 0.0 1.3 0.5 0.4processd rice 0.0 0.0 0.1 0.0 0.3 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0Source: GTAP simulation

22

Table 10 - Decomposition of India agricultural MTRIs by sector

Uniform tariffs ACP Argentina ASEAN China Brazil Japan LDCEuromed countries

NoWTO USA EU25Rest of Europe

EU candidat

es

Australia&New

Zealand

Rest of Asia

Canada Mexico Turkey ChileRest of Latin



23

Table 11 - Decomposition of China agricultural MTRIs by sector

Uniform tariffs ACP Argentina ASEAN Brazil India Japan LDC Euromed countries NoWTO USA EU25 Rest of

Europe EU candidatesAustralia&

New Zealand


Rest of Latin



24

Table 12 - Decomposition of Brazil agricultural MTRIs by sector

Uniform tariffs ACP Argentina ASEAN China India Japan LDC Euromed countries NoWTO USA EU25 Rest of



Rest of Latin



COUNTRIES r rhoEU 0.723** 0.833**

India 0.934** 0.880**USA 0.973** 0.962**Brazil 0.988** 0.961**China 0.827** 0.717**Japan 0.433 0.388

Source: GTAP simulation ** significant at 0.01

Table 13 - Correlation coefficient and rank correlation coefficient among bilateral MTRI uniform tariffs and trade-weighted averages in the agricultural sectors

25

FIGURES Figure 1

USA: Bilateral uniform tariffs on imports and exports

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0

Imposed

Face

d

Regression: y = βx

R2 0.438 β 1.059* * Significant at 0.01

Figure 2

EU: Bilateral uniform tariffs on imports and exports

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0

Imposed

Face

d

Regression: y = βx R2 0.513 β 0.822* * Significant at 0.01

26

Figure 3

Japan: Bilateral uniform tariffs on imports and exports

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0

Imposed

Face

d

Regression: y = βx R2 0.564 β 0.320*

Figure 4

India: Bilateral uniform tariffs on imports and exports

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0

Imposed

Face

d


27

Figure 5

China: Bilateral uniform tariffs on imports and exports

0.010.020.030.040.050.0

60.070.080.090.0

100.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0

Imposed

Face

d


Figure 6

Brazil: Bilateral tariffs on imports and exports

0.0

10.0

20.0

30.0

40.0

50.0

60.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0

Imposed

Face

d


28

Figure 7

India trade restrictiveness and exportes total GDP (billions US$)

0

2000

4000

6000

8000

10000

12000

0.0 10.0 20.0 30.0 40.0 50.0 60.0

MTRI uniform tariff

GDP

Figure 8

India trade restrictiveness and exporters per capita GDP (thousands US$)

05

101520

2530354045

0.0 10.0 20.0 30.0 40.0 50.0 60.0

MTRI uniform tariff

GDP

Regression: y =α + βx R2 0.022 β 0.184

29

Figure 9

Japan trade restrictiveness and exporters total GDP (billions US$)

0

2000

4000

6000

8000

10000

12000

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0

MTRI uniform tariff

GDP

Regression: y =α + βx R2 0.014 β -18.168 Figure 10

Japan trade restrictiveness and exporters per capita GDP (thousands US$)

05

101520

2530354045

0 10 20 30 40 50 60 70 80

MTRI uniform tariff

GDP

Regression: y =α + βx R2 0.000 β -0.010

30

Figure 11

US trade restrictiveness and exporters total GDP (billions US$)

0100020003000400050006000700080009000

10000

-2.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0

MTRI uniform tariff

GD

P


US trade restrictiveness and exporters per capita GDP (thousands US$)

0

5

10

15

20

25

30

35

40

-2.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0

MTRI uniform tariff

GDP

Regression: y =α + βx R2 0.154 β -1.092* * Significant at 0.10

31

Figure 13

EU trade restrictiveness and exporters total GDP (billions US$)

0

2000

4000

6000

8000

10000

12000

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0

MTRI uniform tariff

GDP


EU trade restrictiveness and exporters per capita GDP (thousands US$)

0

5

10

15

20

25

30

35

40

45

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0

MTRI uniform tariff

GDP

Regression: y =α + βx R2 0.010 β -0.177

32

Figure 15

China trade restrictiveness and exporters total GDP

0

2000

4000

6000

8000

10000

12000

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0

MTRI uniform tariff

GDP

Regression: y =α + βx R2 0.089 β 407.885 Figure 16

China trade restrictiveness and exporters per capita GDP

0

5

10

15

20

25

30

35

40

45

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0

MTRI uniform tariff

GDP


33

Figure 17

Brazil trade restrictiveness and exporters total GDP (billions US$)

0

2000

4000

6000

8000

10000

12000

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0

MTRI uniform tariff

GDP


Figure 18

Brazil trade restrictiveness and exporters per capita GDP (thousands US$)

05

101520

2530354045

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0

MTRI uniform tariff

GDP


34

Figure 19

Comparison between the trade weighted and the MTRI uniform tariffs

-20.0

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0

TW tariffs

MTR

I tar

iffs


35

APPENDIX 1 Model and database

We use the GTAP model of global trade (version 6.2), that is a static, multi-region, general equilibrium model based on a representative household taking decisions about consumption, production and public expenditure. The model includes international trade and transport margins, and bilateral international trade flows are handled assuming that products are exogenously differentiated by origin (Armington, 1969). As a standard closure, global investment adjusts to global saving, so that national balances of payments are endogenous: a “global bank” allocates world savings and investment across the countries or regions. As from Hertel (1997) and the GTAP web site (www.gtap.org), the model includes: demand for goods for final consumption (based on a Constant Difference of Elasticity functional form), intermediate use and government consumption, demands for factor inputs (based on a Constant Elasticity of Substitution functional form), supplies of factors and goods, and international trade in goods and services.

The GTAP database is based on a set of Social Accounting Matrices for individual countries; the latest version of GTAP database, version 6.2, provides a baseline referred to year 2001 for up to a maximum of 87 regions and 57 sectors. Trade policy is set at the tariff line level, but this implies a level of detail that is not consistent with the GTAP (or any other existing) model: the EU tariff schedule, for example, includes more than 10 000 tariff lines. To reach consistency between trade distortions and model aggregation, atheoretic trade weighted average tariffs are used.

It should be noted that the quality of the trade distortion data included in the version 6 of the GTAP database is much better than in the previous release due to the use of the MAcMAP-HS6 (version 1), a database at the HS-6 level intended to provide a set of consistent and exhaustive ad valorem equivalents (AVEs) of applied border protection across the world. 1 This resulted in considering applied/preferential tariffs rather than bound ones, and in a more accurate computation of the AVE for each trade instrument (Bouët et al., 2005a).

Specific tariffs are converted into AVEs terms by taking the ratio of each import duty and a unit value, whose choice is a rather sensitive issue both from a theoretical and a political point of view. In MAcMAP, AVE calculations are based on the median unit value of worldwide exports originating from a reference group to which the exporter belongs.2

For mixed tariffs, i.e. tariffs involving a choice (a maximum or a minimum operator) between different terms, the MAcMap approach is the following:

• when the tariff is defined as an ad valorem base tariff, with in addition a cap and a floor (which are defined in specific terms), the base tariff is retained. If the base tariff is in specific terms and the cap and the floor are ad valorem, a simple average of the two bounds is retained. This prevents additional noise being added through AVE calculation;

• when the tariff involves a choice between two terms, priority is given to the one defined in ad valorem terms.

Concerning tariff rate quotas, three market regimes are considered, depending on the extent to which the quota is filled:

• if less than 90 percent of the quota is filled, it is considered not to be binding, hence the in-quota tariff rate is chosen as the applied rate;

• in the 90 percent to 99 percent range the quota is assumed to be binding, hence a simple arithmetic mean is considered as the applied rate;

• if more than 99 percent of the quota is filled, this is considered to be binding, and therefore out-quota tariff rate is chosen as the applied rate.

1 MAcMAP-HS6 is regularly improved and updated, and the corresponding information is available on the CEPII's website (www.cepii.fr). 2 These groups are defined on the basis of a hierarchical clustering analysis based on GDP per capita (in terms of PPP) and trade openness.

36

Finally, for prohibitive tariffs – whose presence is problematic when calculating AVEs - an upper limit is established starting at the HS6 level, which involves setting 1 000 percent as a maximum for the sum of all measures.

Table A1. Countries, regions, products and endowmentsCountry/region Products Endowments Australia & New Zealand Agriculture Land China Paddy rice Skilled labour Japan Wheat Unskilled labour ASEAN Cereal grains Capital Rest of Asia Vegetables, fruit, nuts Natural resources ACP countries Oil seedsLDC countries Sugar cane, sugar beetEU-25 Plant-based fibersEU candidades Crops necRest of european countries Bovine cattle, sheep and goats, horsesEuromed countries Animal products necTurkey Vegetable oils and fatsIndia Raw milkUnited States Wool, silk-worm cocoonsCanada ForestryMexico FishingArgentina Bovine meat productsBrazil Meat products necChile Dairy productsRest of Latin America Processed riceNo WTO Sugar

Food products necBeverages and tobacco productsManufacturesMineralsTextile sectorWood productsPaper products, publishingPetroleum, coal productsChemical, rubber, plastic productsOther manufacturingMetal productsMotor vehicles and partsElectronic equipmentServicesWaterConstructionTradeCommunicationTransport servicesFinancialOther services

37

APPENDIX 2 The 2004 baseline

In order to evaluate the market access level for all products and services, including agri-food and manufactured goods, we adopted an imperfect competition closure for the model. More specifically, we adopted the approach suggested by Francois (1998) in order to model economies of scale and monopolistic competition. Regarding the former, scale economies are introduced in an otherwise standard specification through a new exogenous variable (OSCALE), while the output augmenting technical change variable (AO) is declared endogenous for the relevant sectors.

As far as the latter is concerned, in each region, when an industry j is assumed to be monopolistically competitive, this means that individual firms produce unique varieties of good j, and hence are monopolists within their chosen market niche. Given the demand for variety, the demand for each variety is less than perfectly elastic. However, while firms are thus able to price as monopolists, free entry drives their economic profits to zero, so that pricing is at average cost. Moreover, since consumers decide over different varieties and a non-nested structure for import demand is adopted, the Armington assumption is dropped. In practice, the substitution parameter between the domestic and the composite imported commodities (ESUBD) has been set equal to the “Armington elasticity” among imported commodities from different sources (ESUBM), where the latter is calibrated from the distance between average and marginal cost (“cost disadvantage ratio”).

Concerning the service sectors, since the database does not include any protection measures, we introduced estimates of ad-valorem equivalent tariffs drawn from the literature (Park, 2002). Version 6.2 of the database was aggregated for this application to include 22 regions, 39 products and 5 endowments (Table A1).

Since the paper focuses on the EU, Japan and the US protection structure, the regional aggregation aims to highlight the most relevant regions for the bilateral trade policies of these countries. The product aggregation is as detailed as possible, taking into account the estimates available for the scale economies and tariff equivalents. As a consequence, the largest number of products refer to the primary sector, but this is an interesting feature of the model, since these products present the highest levels of tariff protection (Bureau and Salvatici, 2004).

A number of changes have been introduced into the model in order to update the baseline from 2001 to 2004. Furthermore, we decided to include some policy changes already decided, but to be implemented after 2004, such as sugar sector reform or the EBA initiative in the case of the EU. More specifically, GDP, population, labour force and total factor productivity were shocked taking into account the changes between 2001 and 2004. Concerning policy shocks, particular consideration was given to the Common Agricultural Policy (CAP), which has undergone significant modifications over this period: the residual implementation of the “Agenda 2000” reform, and the Fischler reform of 2003 (Bach et al., 2000; Brockmeier et al., 2001; van Meijl and van Tongeren, 2002). Moreover, the enlargement of the EU, and the related extension of the CAP to ten new members was taken into account by removing import tariffs between the EU and the CEECs, and through the alignment of export, output and input subsidies or taxes. Finally, a set of shocks was introduced into the model to take into account the change in the preferential policy pursued with the EBA framework, allowing all imports from the LDC countries to access the EU market duty free from 2009.

The reductions of intervention prices in the sectors of rice, sugar, cereals and dairy products3 were approximated through changes in the corresponding import taxes. With regard to the Fischler reform, given the model’s characteristics, it was only possible to consider the decoupling of direct payments, i.e. their switch to non-crop-specific payments. This measure, which is considered the most important among those introduced by the 2003 reform, is represented in the model through a homogeneous subsidy to land use, captured by an additional variable, whose level is determined endogenously on the

3 For raw milk, output quotas were modelled by setting production exogenously at the level of the base period, and checking after each step undertaken whenbuilding the 2004 baseline, that this limit was effectively binding. This prevents the quota from acting as a minimum rather than a maximum constraint on output.

38

basis of the expenditure arising in the baseline from the granting of crop-specific subsidies. It is worth mentioning that in order to have a proper modelling of the CAP, the previous shocks have been implemented distinguishing among the old (EU15) and the new (EU10) members of the EU: only at the end of the baseline creation procedure, was a new aggregation (EU25) created in order to compute the protection indexes for the enlarged EU.

Concerning other countries’ policies, some of the provisions of the 2002 FSRI Act were included in the baseline, following mostly Bouët et al. (2005b). A reduction in land productivity was introduced to take into account the increase in the acreage conservation programme; output subsidies were increased for cereals and dairy products, but decreased in the case of soybeans.

Also the decoupled payments of the PROCAMPO program in Mexico were increased in 2004, taking into account the rates applied for farmers with more and less than five hectares of land (FAO, 2005) through a weighted average increase based on sizes reported by Eastwood et al. (2004). In addiition, the recent introduction of direct payments in 13 provinces of China was also taken into account (FAO, 2005) as an ad valorem subsidy to land use in cereals, rice, and oilseeds, taking into account the share of the relevant provinces in total arable land.

Finally, the integration of China into the WTO has been implemented by updating the level of tariffs to 2004, based on data from the TRAINS database. In order to be coherent with the GTAP database, the tariffs introduced into the model are weighted by trade flows.

FAO COMMODITY AND TRADE POLICY RESEARCH WORKING PAPERS

2006

22 Threshold cointegration in the sugar-ethanol-oil price system in Brazil: evidence from nonlinear vector error correction models George Rapsomanikis and David Hallam 21 Estimating price elasticities of supply for cotton: a structural time-series approach Ben Shepherd 20 Market access and preferential trading schemes: Evidence from selected developed and developing countries Piero Conforti and Luca Salvatici 19 The role of agriculture in reducing poverty in Tanzania: A household perspective from rural Kilimanjaro and Ruvuma Alexander Sarris, Sara Savastano and Luc Christiaensen 18 Producer demand and welfare benefits of rainfall insurance in the United Republic of Tanzania Alexander Sarris, Panayiotis Karfakis, and Luc Christiaensen 17 Household vulnerability in rural Tanzania Alexander Sarris and Panayiotis Karfakis 16 The use of organized commodity markets to manage food import price instability and risk Alexander Sarris, Piero Conforti and Adam Prakash 15 The impact of domestic and international commodity price volatility on agricultural income instability in Ghana, Vietnam and Peru George Rapsomanikis and Alexander Sarris 14 Linkages between domestic and international maize markets, and market-based strategies for hedging maize import price risks in Tanzania Alexander Sarris and Ekaterini Mantzou 13 Food import risk in Malawi: simulating a hedging scheme for Malawi food imports using historical data Wouter Zant 2005 12 The effect of direct payments of the OECD countries in world agricultural markets. Evidence from partial and general equilibrium frameworks Piero Conforti

11 The impact of import surges: country case study results for Senegal and Tanzania Ramesh Sharma, David Nyange, Guillaume Duteutre and Nancy Morgan 2004 10 Agricultural trade liberalization in the Doha round. Alternative scenarios and strategic interactions between developed and developing countries Piero Conforti and Luca Salvatici 9 The EU cotton policy regime and the implications of the proposed changes for producer welfare Giannis Karagiannis 8 The impact of domestic and trade policies on the world cotton market Daneswar Poonyth, Alexander Sarris, Ramesh Sharma and Shangnan Shui 7 Price transmission in selected agricultural markets Piero Conforti 6 The marketing potential of date palm fruits in the European market Pascal Liu 5 World markets for organic citrus and citrus juices: Current market situation and medium-term prospects Pascal Liu 4 Agricultural Policy Indicators Timothy Josling and Alberto Valdés (also issued as ESA Working Paper No. 2004/4) 2003 3 Quantifying appropriate levels of the WTO bound tariffs on basic food products in the context of the Development Box proposals Ramesh Sharma 2 The WTO and environmental and social standards, certification and labelling in agriculture. Cora Dankers 1 The Brazilian ethanol programme: impacts on world ethanol and sugar markets Tatsuji Koizumi

market access and preferential trading schemes: evidence

Documents