The Case of Trade with the Netherlands

Vietnam’s Trade in the Wake of its WTO Membership

The Case of Trade with the Netherlands

Master Thesis International Economics and Business

Vietnam’s Trade in the Wake of its WTO Membership

The Case of Trade with the Netherlands

Master Thesis International Economics and Business

The Hague, 21 December 2006

In cooperation with:

Ministry of Agriculture, Nature and Food Quality Department of International Affairs; Global Affairs The Hague, the Netherlands

Written by:

T.T.K. Tran (Kim) Student number: 1283472 Rijksuniversiteit Groningen Faculty of Economics

International Economics and Business Groningen, the Netherlands

t.t.k.tran@student.rug.nl

Supervisor:

Dr. G.J. Lanjouw

Rijksuniversiteit Groningen Faculty of Economics

International Economics and Business

Groningen, the Netherlands

ACKNOWLEDGMENT

This thesis is the end product of nearly six months of hard work, but also of many learning moments.

In a broader context, this thesis is also the result of five years of studies in International Economics and Business at the University of Groningen. Soon, I will make my way to the labour market with a degree in my pocket. In preparation of mingling with other professionals, this thesis was undertaken at the Dutch Ministry of Agriculture, Nature and Food Quality in The Hague in combination with an internship.

The Ministry of Agriculture, Nature and Food Quality was a logical choice, since exports to Vietnam largely consist of agricultural products. Furthermore, the Ministry is involved in so-called partnership projects with Vietnam, whereby the focus is on market-access improvement. Currently, there are partnership projects in the field of seafood. With the approaching WTO accession of Vietnam the bilateral trade relationship might change. The aim of this thesis is to quantify the WTO impact on exports in both directions. Moreover, recommendations are made to the Ministry to expand partnership projects to products other than seafood.

The Ministry contributed to this thesis in many ways. For that reason, I would like to take this opportunity to express my gratitude to a few people by name. First of all, I sincerely wish to thank Hans Hoogeveen and Marcel Vernooij for giving me the opportunity to learn and experience how the Ministry is organized.

Furthermore, I am most grateful to Karin Olsthoorn and Omer van Renterghem for bringing me into contact with the Ministry of Agriculture.

Also to be mentioned Bart Vrolijk, Rubert Konijn and Carla Boonstra who gave valuable comments on a draft version of my thesis and gave me constructive suggestions leading to improvements thereof, for which I am very grateful. Moreover, I would like to express my gratitude to the department of International Affairs, for making me feel welcome and for providing information on the subject. I have really enjoyed working at the Ministry in my cosy little room, with some pleasant distraction just outside of my doorway.

Last, but certainly not least, I would like to thank dr. G.J. Lanjouw, my thesis supervisor from my University, for reading my thesis and for giving critical and constructive comments. His guidance was indispensable for this research.

Many thanks to all the others, that I have not mentioned by name, I highly valued your support.

The Hague, the Netherlands, 21 December 2006

Tan Thien Kim Tran

ABSTRACT

Vietnam is eager to accede to the World Trade Organization (WTO), since

this institution promises economic growth to its Members. It had pushed for a

2004 membership, but failed. The aim of this research is to quantify the

foregone exports, disaggregated into agriculture and non-agricultural exports

between the Netherlands and Vietnam, and vice versa, as a consequence of

Vietnam failing to accede in 2004. Moreover, the differential impact of WTO

on agriculture and non-agriculture is considered as well. This research found

insignificant result for agricultural exports in both directions and non-

agricultural exports from Vietnam. Significant results were found for non-

agricultural exports from the Netherlands to Vietnam, these are expected to

increase substantially.

TABLE OF CONTENTS

I. INTRODUCTION... 8

1.1. Vietnam and International Trade... 8

1.2. Aim of this study ... 10

II. HISTORICAL BACKGROUND ... 11

2.1. Vietnam’s Integration into the World Economy ... 11

2.2. Trade Patterns ... 13

2.3. Trade with the Netherlands... 15

2.3.1. Public Private Partnership projects in Vietnam ... 16

III. THEORETICAL MODEL... 19

3.1. Computable general equilibrium model ... 19

3.2. Partial equilibrium ... 20

3.3. The gravity model... 21

IV. MODEL ... 24

4.1. Research Questions and Hypotheses ... 24

4.2. Model Specifications ... 25

4.3. Variables and Data... 27

4.3.1. Exports ... 27

4.3.2. Production... 28

4.3.3. Consumption ... 29

4.3.4. Factors of production... 29

4.3.5. WTO dummy... 31

4.3.6. Bilateral trade agreements... 31

4.4. Statistical Methods ... 31

4.4.1. Endogeneity... 31

4.4.2. Heteroskadicity ... 32

4.4.3. Collinearity and Multicollinearity ... 32

4.4.4. Unit Roots ... 32

V. RESULTS... 33

5.1. Agricultural exports... 33

5.1.1. China to the Netherlands ... 33

5.1.2. The Netherlands to China ... 35

5.2. Non-agricultural exports... 37

5.2.1. China to the Netherlands ... 37

5.2.2. The Netherlands to China ... 38

VI. APPLICATION TO VIETNAM ... 40

VII. CONCLUSIONS ... 42

VIII. RECOMMENDATIONS AND LIMITATIONS ... 44

7.1. Recommendations ... 44

7.2. Limitations... 44

Appendix A: Comparison China and Vietnam before acceding to WTO... Error! Bookmark not defined.

Appendix B: Bilateral Trade Agreements between the Netherlands and Vietnam ...Error! Bookmark not defined.

Appendix C: Correlation Matrix...Error! Bookmark not defined.

Appendix D: Commodity List; SITC Rev.1...Error! Bookmark not defined.

LIST OF FIGURES AND TABLES

Figure 2.1. Trade in Vietnam in 2004 ... 13

Figure 2.2. Vietnam’s trade with major partners in 2004 ... 14

Figure 2.3 Exports from the Netherlands to China 1995... 17

Figure 2.4. Exports from the Netherlands to China 2004... 17

Figure 2.5. Agricultural exports from the Netherlands to China... 17

Figure 2.6. Agricultural exports from China to the Netherlands ... 17

Figure 2.7. Exports from China to the Netherlands 1995... 18

Figure 2.8. Exports from China to the Netherlands 2004... 18

Figure 4.1. Average years of education for China and the Netherlands ... 30

Figure 6.3. Non-agricultural exports with WTO membership in 2004 ... 40

Figure 6.4. Actual non-agricultural exports compared to 2004* and 2004**... 41

Table 1.1. China’s share of world trade 2001-2004 ... 9

Table 2.1. Trade flows between the Netherlands and Vietnam... 15

Table 5.1. Regression output for agricultural exports from China to the Netherlands. ... 34

Table 5.2. Non-stationary variables ... 35

Table 5.3.Regression output for agricultural exports from the Netherlands to China. ... 35

Table 5.4. Non-stationary variables ... 36

Table 5.5. Regression output for non-agricultural exports from China to the Netherlands. ... 38

Table 5.6. Regression output for non-agricultural exports from the Netherlands to China. ... 39

Box 3.1. A gravity equation ... 21

Box 4.1. Regression equations for RQ1 ... 26

Box 4.2. Regression equation for RQ2... 27

Box 4.3. Baxter and Kouparitsas’ factors of production ... 29

I. INTRODUCTION

1.1. Vietnam and International Trade

Undoubtedly, Vietnam will be the next great Asian Tiger, given its economic performance (BBC News, 2000). Vietnam’s real GDP grew by 7.7 percent in 2004, despite the avian influenza outbreak and was surpassed in the region only by China and Singapore. Vietnam is of increasing importance to the region. As an illustration, Newsweek (2005) reported that it is expected that large multinationals will flee China’s higher manufacturing costs, reducing margins and intensifying price competition and start to invest in Vietnam instead, An example of this are Mabuchi Motor Co., the world’s largest producer of small motors, and Fujitsu Ltd. They have relocated parts of their operations from China to Vietnam and considered this to be the preferred South East Asian country due to its lower-cost manufacturing base.

Furthermore, the upcoming accession of Vietnam to the World Trade Organization (WTO) will open up the economy to investors and liberalize trade. WTO membership requires Vietnam to strengthen legal institutions and to bring these in line with international practices and WTO rules.

Restructuring of legal systems and the economy will increase transparency in Vietnamese rules and regulations; this reduces the risk and uncertainty faced by foreign investors (Drabek and Payne, 1999).

Consequently, upcoming WTO membership will be a big leap forward in integrating Vietnam into the world economy, this would attract more investment capital for establishing and developing processing industries that obviously are desirable to further push the industrialization programs of Vietnam. As in China, WTO membership can be expected to be positive for trade expansion and economic growth.

The result could be as substantial as US$5.4 billion in new foreign direct investment in Vietnam in 2006 (Newsweek, 2005).

The agricultural sector is the most important sector in Vietnam, since this sector provides employment to seventy percent of the Vietnamese population; many families are dependent on incomes earned in this sector. The output of this sector accounts for over twenty percent of GDP.

Vietnam is now the world’s top exporter of black pepper, the second largest exporter of rice, coffee and cashew nuts and the fourth largest exporter of rubber and the seventh largest for tea. The largest growth is in the forestry

¹

product processing industry with total export turnover of more than US$ 1 billion, accounting for 25 percent of the export turnover of the agricultural sector as a whole (EU Report on Vietnam, 2005 ).

The agricultural sector is important to the bilateral trade relationship between the Netherlands and Vietnam as well, since agricultural products account for over sixty percent of Dutch exports to Vietnam. In general this sector is important to the Netherlands, since the Netherlands is the world’s third largest agricultural exporter, after the US and France (Dutch Ministry of Agriculture, Nature and

1 Forestry is included in agriculture. The definition of agriculture will be discussed in chapter IV.

Food Quality, 2006). Owing to its trade in agricultural products, the Netherlands managed to maintain a surplus in its bilateral trade account. Both countries would gain from increased bilateral agricultural trade.

Currently, various tariff and non-tariff barriers apply to bilateral trade between Vietnam and the Netherlands, in broader aspect the EU, and Vietnam, especially for the agricultural sector.

Traditionally, agriculture is a sensitive sector in WTO negotiations

²

with many hurdles having to be taken before reaching consensus between member states on both tariff and non-tariff regulations.

However, WTO negotiations will continue to aim at further reducing trade barriers in this sector in order to be more market conform. This will have an impact on trade in agricultural products.

Conditional upon trade barriers being further reduced, Vietnamese WTO membership will offer new opportunities to the Netherlands in the agricultural sector.

WTO impact on economic development and trade is evident in China. WTO accession of China in 2001 has boosted its foreign trade (Tong, 2006) as indicated in table 1.1, where trade is disaggregated in exports and imports. Since China has opened its domestic market to all other WTO members, it can export commodities needed in China, including advanced capital goods, general merchandise, agricultural produce, materials and energy products (Tong, 2006). Simultaneously, imports have increased rapidly. The reason for this import increase is straightforward; its increased production requires additional energy and materials. As a result, China is now the third largest trading country. In addition, with more income being generated by economic growth, Chinese people will want to consume more. In fact, for the first time in Chinese history, consumption has become a force that is driving economic development (Tong, 2006).

Source: WTO, International Trade Statistics 2004; values are in US$ billions.

It is very likely that Vietnam will follow in China’s footsteps and will develop in a similar manner, since endowments are quite comparable between both countries. Similarities are scarcity in land

³

(Smil, 1999) and capital (Tong, 2006) and abundance of labour. Trade, as a consequence of

2 At the time of writing, the Doha Round negotiations are suspended until further notice.

3 Even though China has much land, only 14 percent can be used for agriculture. The potential for expanding cultivated land in China is extremely limited. Urban industrial, commercial, and residential uses are competing with agriculture for China’s limited land resources (Smil, 1999).

Year World Exports

China Exports

China's Share in World Exports

Ranking World Imports

China Imports

China's Share in World Imports

Ranking

2001 6162.4 266.2 4.3 6 6410.5 243.6 3.8 6

2002 6432.9 325.6 5.1 5 6709.1 295.2 4.4 6

2003 7482.0 438.4 5.9 4 7794.3 413.1 5.3 3

2004 9123.5 593.4 6.5 3 9511.9 561.2 5.9 3

Table 1.1. China’s share of world trade 2001-2004

WTO accession, i.e. lower trade barriers, will tend to reflect these characteristics and hence Vietnam will import land-intensive products and export labour-intensive products. Similar trading patterns have appeared in China, as was to be expected on the basis of the Heckser-Ohlin (H-O) model

⁴

.

1.2. Aim of this study

Vietnam applied for WTO membership in January 1995. Ten years of negotiations have passed, and still Vietnam has not acceded to WTO. By looking at the case of China, Vietnam sees what to expect as soon as it is a WTO member.

This research estimates what the agricultural exports from and to Vietnam and the Netherlands would have been, if Vietnam had already been WTO member in 2004. Initially Vietnam pursued 2004 as the accession year. The focus is WTO membership impact on bilateral trade between the Netherlands and Vietnam using China as a reference. The research question will be: “What impact would WTO membership of Vietnam in 2004 have had on agricultural

⁵

exports between the Netherlands and Vietnam, and vice versa, in that same year?” Another area of interest is whether WTO will have a bigger impact on agricultural than on non-agricultural trade, as literature led us to expect. Thus, the second research question will be: “Is the impact of WTO membership substantially and significantly larger for agricultural exports in 2004?” Furthermore, the content of trade in China is analysed to detect any shifts in trade pattern.

The outline of the paper is the following. Chapter II discusses the historical background of Vietnam and its trade pattern with the Netherlands. Chapter III presents the theoretical model, which covers literature on the impact of trade liberalization on trade and welfare. Chapter IV specifies the applied model, its variables and data. Chapter V presents the results of the regression analysis.

Concluding remarks are stated in chapter VI. Furthermore, chapter VII gives recommendations for the Dutch Ministry of Agriculture’s policy. Limitations of this research are given in the final chapter VIII.

4 The H-O theory states that country pairs will trade more, the more different are their factor endowments.

5 This research extends WTO definition of agriculture with fishery and forestry. The definition will be further explained in chapter IV.

II. HISTORICAL BACKGROUND

Currently, Vietnam has a market economy with a socialist orientation (EU Report on Vietnam, 2005). It is the result of twenty years of reforms. Before Vietnam arrived at this point, it had passed many stations. This chapter discusses the different historical reforms of economical structures. In addition, the general trade patterns of Vietnam are examined. The last chapter is devoted to trade with the Netherlands. Prospects are given for trade between the Netherlands and Vietnam, using China as a reference.

2.1. Vietnam’s Integration into the World Economy

From late 1970s until the 1990s, Vietnam was a member of Comecon

⁶

, and therefore heavily dependent on trade with the Soviet Union and its allies. After the fall of Communism in Eastern Europe, and consequently the end of Comecon, Vietnam lost its traditional trading partners

⁷

. Hence, Vietnam was forced to find new trading partners and to liberalize trade.

In 1986, the Sixth Party Congress of the Communist Party of Vietnam formally abandoned communist economic planning and began introducing market elements as part of a broad economic reform program called Doi Moi, meaning renovation. The need for restructuring arose because Vietnam was not growing enough rice under collectivized agriculture to be able to feed itself. Vietnam had to import 1.5 millions of tons of rice and starvation conditions were prevalent. As a result of several land reform measures, Vietnam is now the second-largest rice exporter in the world (Ministry of Agriculture, Nature and Food Quality, 2005).

Vietnam’s 1995 admission to ASEAN

⁸

was the final act of regional political reconciliation following the Third Indochina War

⁹

(1978-1991) pitting communist Vietnam against non-communist ASEAN over the future of Cambodia. Despite a political and economic gap separating Vietnam from its new ASEAN partners, accession to the group and Hanoi’s adoption of ASEAN standards of international peace and cooperation, symbolically heralded Vietnam’s entrance into the western market-oriented global system.

Just as important to Vietnam’s economic development goals was the normalization of relations with the US (Weatherbee, 2005). This occurred in 1995, after four decades of conflict. Trade and

6 The Council for Mutual Economic Assistance (Comecon) was an economic organization of communist states and a kind of Eastern Bloc equivalent to – but more inclusive than – the European Economic Community.

7 Russia still has a stake in Vietnam, they have signed a strategic partnership pact. In exchange for aid in the modernization of the Vietnamese armed forces, Vietnam, in turn, would assist Russia in its bid for a seat at the Asia-Europe meeting (ASEM).

8 Association of South East Asian Nations (ASEAN) established on 8 August 1967 with currently ten members:

Brunei Darussalam, Cambodia, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, Thailand and Vietnam.

9 The Third Indochina War: The Khmer nationalist forces’ resistance backed by ASEAN, China, and the United States, to the Vietnamese December 1978 invasion and occupation of Cambodia. It ended with the 1991 peace agreement that created the United Nations Transitional Authority in Cambodia (UNTAC).

investment between the two countries has increased ever since, and Vietnam continues to give full cooperation in the search for Americans killed or missing in action during the American Indochina War (1961-1975).

Initiatives to secure the trade relationship between the Netherlands – and also with the EU as a whole – and Vietnam led to different bilateral agreements: the Investors’ Protection Agreement (1994); the Tax Treaty (1995); the Economic Cooperation Agreement (1995); Market Access Agreement (2004). Vietnam also benefits from preferential access to EU markets under the Generalized System of Preferences (GSP)

¹⁰

. In the former GSP scheme (1995-2005), China used to be the largest GSP beneficiary (35.8% of imports under GSP). However, under the renewed GSP scheme China’s improved competitiveness means that 80 percent of its past GSP exports are now phased out of GSP. With China losing its status of GSP beneficiary, Vietnam will gain a competitive edge over China, with lower trade barriers into the EU.

Vietnam’s full entrance into the global economy still awaits adoption and implementation of measures required for it to join WTO. Vietnam pushed for membership in 2004, but [Vietnam] is still a long way from reaching its goal of becoming WTO member. Since acceding to WTO failed in 2004, Vietnam is foregoing additional economic growth that could have been achieved had it obtained WTO membership earlier. Vietnam’s goal of joining by 2005 was not feasible either due to unsuccessful bilateral negotiations

¹¹

with several of the sixty-three interested WTO members

¹²

. According to the Working Party on the accession of Vietnam to the WTO, Vietnam needs to ‘quantum jump’, in order to finish more bilateral talks, despite the tremendous efforts and progress Vietnam has already made.

To give an indication of the gap to bridge: China has committed to open 57 percent of its sectors/modes, whereas Vietnam offered only to open 39 percent of its sectors in their bilateral negotiation with the US (World Bank, 2003).

At the time of writing this report, the bilateral agreements were done, the multilateral agreement nearly, thus Vietnam’s membership was in sight. By the time of finalizing this research the date of ratification is set on 13 January 2007. So finally the negotiation process is indeed coming to an end after more than ten years.

All efforts by the Vietnamese government to comply with international standards will make Vietnam more transparent to potential foreign investors, which in turn can be translated into strong economic performance in the near future. Vietnam’s trade patterns will now be discussed in the next chapter.

10 GSP is a preferential trade regime to assist developing countries. The EU GSP is the most widely used of all developed country GSP system. The newly revised system is valid for the period 2005-2008.

11 Of which in total 28 countries had requested for actual bilateral negotiations.

12 The US, Australia, New Zealand, Mexico, Honduras and Dominican Republic.

2.2. Trade Patterns

¹³

Vietnam’s foreign trade in 2004 performed impressively, with export values growing at a post- Asian crisis (1997) record rate of 28.9 percent, reaching US$ 26 billion (figure 2.1). Imports rose by a robust 25.0 percent, to US$ 31.5 billion. The trade deficit reached an all-time record of US$ 5.5 billion, equivalent to 3.9 percent of GDP (EU Report on Vietnam, 2005).

A trade deficit may reflect the needs of a rapidly expanding economy. The largest imports were in machinery and equipment, petroleum products and other capital goods and production inputs.

This points towards brisk investment and the potential for higher export-oriented production, but also flags the weakness of Vietnam’s economy, namely its dependence on imports for the lion’s share of the inputs needed for its key export industries (EU Report on Vietnam, 2005).

Figure 2.1. Trade in Vietnam in 2004

Figure 2.1. Trade in Vietnam in 2004

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0

1995 1996

1997 1998

1999 2000

2001 2002

2003 2004

US$ billion

Total Trade Imports Exports

In 2004 export growth was led by Vietnam’s traditional export sectors: crude oil (up 48.3 percent to US$ 5.6 billion), textiles and garments (up 17.2 percent to US$ 4.3 billion), and footwear (up 14.8 percent to US$ 2.6 billion). Exports in three other product groups – seafood, wood products and electronic goods – also yielded more than US$ 1 billion each. The latter two grew strongly, at 60.2 and 85.9 percent respectively, signalling Vietnam’s partial success in diversifying its export structure and moving up the value chain. Nonetheless, labour-intensive and low value-added production continues to dominate the export sector, and the economy at large. Moreover, agricultural commodities (without the burgeoning fisheries sector) and crude oil together still account for around

13 All trade flows in this chapter are seen from the perspective of the Netherlands; meaning exports from and imports to the Netherlands.

Source: - IMF, based on General Statistical Office of Vietnam (GSO).

half of total exports, and thus Vietnam’s exposure to fluctuating world market prices remains considerable.

Vietnam’s biggest trading partner outside ASEAN is the EU, representing 13.3 percent of total trade (figure 2.2), followed by Japan, China and the US. EU’s share has remained relatively stable at 13-14 percent in the last decade. Trade flows with the US, which had exploded following the coming into force of the US-Vietnam Bilateral Trade Agreement in 2002, are becoming more mature; they now represent 11.1 percent of total (up from only 2.2 percent in 1995). Trade with China also grew strongly. China is set to overtake Japan as second trading partner (figure 2.2). The shares of ASEAN and Japan have both declined by around 5 percent in the last decade, and now stand at 19.3 percent and 12.4 percent respectively. In ASEAN’s case, this reflects the existence of non-tariff barriers on both sides, as well as similarities in export structures and the low competitiveness of some Vietnamese products.

The trade relationship between the Netherlands and Vietnam has to be looked at in a broader context, since the bilateral agreement was negotiated between the EU and Vietnam. As already noted, outside ASEAN, the EU takes the No.1 trading position with Vietnam. The EU is the fourth largest exporter to Vietnam, following ASEAN, China and Japan, but ranks second for exports from Vietnam, below the US (figure 2.2). In general the EU is a bigger importer from than exporter to Vietnam in terms of trade value and growth rates, which leads to a bilateral trade deficit for the EU.

Figure 2.2. Vietnam’s trade with major partn ers in 2004

The International Trade Centre (ITC)

¹⁴

of UNCTAD and WTO compiled the top 200-export products

¹⁵

from Vietnam. The trade flow in 2004 to the EU mainly consists of footwear and headgear,

14 ITC is the technical cooperation agency of the United Nations Conference on Trade and Development (UNCTAD) and WTO for operational, enterprise-oriented aspects of trade development.

Figure 2.2. Vietnam's trade with major partners in 2004

0 2000 4000 6000 8000 10000 12000

exports imports total trade

in US$ mln

EU-15 ASEAN US Japan China

Source: EU Report on Vietnam, 2005

which accounted for 70 percent of total exports to the EU. Other products were vegetable products (9%) and furniture (7%). The Netherlands is an important importer of furniture, footwear and vegetable products. These products are the top three most traded products in Vietnam. The Netherlands takes at most 10 percent of the footwear trade going to the EU. Since the Netherlands can serve as a gate to other EU countries, with its Main Port Rotterdam, therefore it should keep an eye on other traded products in order to reap the benefits from increased trade between the EU and Vietnam.

2.3. Trade with the Netherlands

Given the fact that Vietnam had already been granted preferential access to EU markets, WTO impact will be relatively larger on exports from the Netherlands to Vietnam than exports in the opposite direction.

Trade flows of 2004 between the Netherlands and Vietnam are shown in table 2.1. Imports from Vietnam are substantially larger than exports to Vietnam. The difference between the growth rates of agricultural and non-agricultural imports declined drastically from 60 (7 - 67 percent) to 1(33 - 34 percent) for respectively agricultural and non-agricultural imports over the period of 2002-2004. In other words, the growth rates have moved closer to each other. However, the value of trade in non- agricultural products is still more than three times higher than trade in agriculture. Hence, it is not very likely that trade in agricultural products will catch up, since growth rates have converged largely.

However, as soon as WTO members have reached consensus about agricultural reforms, growth rates for agricultural products trade can very well increase, given that trade in these products would be more attractive with lower trade barriers.

Source: Trademap.com.

The same opportunities are present for agricultural export flows from the Netherlands to Vietnam. Currently, agriculture already constitutes a substantial share of total exports to Vietnam,

15 The Top 200 exported products list rank countries according to their import value from Vietnam according to product codes, these products are currently important for the Vietnamese exports.

2002 2003 2004

In US$ ‘000 EX IM EX IM EX IM

Agricultural products 39,702 49,414 60,194 82,422 93,168 109,709

Growth rates agr. prod. - - 52% 67% 55% 33%

Non-agricultural prod. 38,785 258,626 53,714 277,086 68,069 371,444 Growth rates non-agr.

prod. - - 38% 7% 27% 34%

Table 2.1. Trade flows between the Netherlands and Vietnam

namely 40 percent. In addition, exports in agricultural products are growing at a relatively high rate, while growth in non-agricultural products is much lower. Altogether, the Dutch agricultural export sector is performing well, and WTO membership of Vietnam would bring new opportunities, which signals potential for the Netherlands.

Main imported agricultural products from Vietnam to the Netherlands are cashew nuts, fish and coffee, accounting for 40, 18 and 11 percent, respectively, of total agricultural trade. Agricultural exports are dominated by food preparations and milk and cream, which represent respectively 35 and 27 percent of total agricultural trade.

2.3.1. Public Private Partnership projects in Vietnam

The Dutch Ministry of Agriculture, Nature and Food Quality is involved in a number of so- called Public Private Partnership projects in Vietnam. This concept was developed at the World Summit for Sustainable Development in Johannesburg, and is defined as a voluntary cooperation arrangement between the actors from government and non-governmental sectors (NGOs, the business community) that agree to work together to achieve a common objective to carry out a specific task, by sharing the risks, responsibilities, resources, competencies and benefits. These projects aim at increasing market access for seafood products from Vietnam onto the European markets. The partnerships focus on capacity building in order to enhance quality related to food safety, health, sanitary and veterinary and environmental measures. Currently, partnerships only exist in the seafood sector. With the upcoming WTO membership, trade flows and the comparative advantage of Vietnam might shift from agricultural products to non-agricultural products, following China’s footsteps.

Obviously, the parties involved in the partnerships should be aware of these potential shifts, in order to seek new market opportunities.

However, Vietnam needs time and capital to make any shifts in its trade, due to constraints on the supply side. As a consequence, no predictions can be made on potential shifts by looking at WTO impacts only. Yet, by looking at any shifts in bilateral trade between the Netherlands and China, we can predict what direction Vietnam will be heading for after WTO accession. Now, the shifts in trade patterns between the Netherlands and China are going to be analysed below, to predicts any shifts in the trade pattern for Vietnam after accession.

Figures 2.3 and 2.4 present agricultural shares in exports from the Netherlands to China. As can be seen, even after WTO accession, these shares have remained rather stable.

When looking at content of agricultural exports

¹⁶

from the Netherlands in figures 2.5, it can be seen that only three product groups dominate exports from the Netherlands, namely food and live animals; crude materials, inedible except fuels; and manufactured goods classified chiefly by materials (like rubber and wood). Crude materials, product group two, experienced the highest growth rate.

16 List of products of SITC Rev.1 is in Appendix C.

Figure 2.3. Exports from the Netherlands to China 1995

28%

72%

agr non agr

Figure 2.4. Exports from the Netherlands to China 2004

30%

70%

agr non agr

Source: UN ComTrade Source: UN ComTrade

Figure2.3 Exports from the Netherlands to China 1995 Figure 2.4. Exports from the Netherlands to China 2004

Source: UN ComTrade

Figure 2.5. Agricultural exports from the Netherlands to China Figure .6. Agricultural exports from China to the Netherlands

Doing the same analysis for exports from China shows us that the export share of agricultural products decreases compared to exports in 1995 (figures 2.7 and 2.8). The same three products groups dominate content of agricultural exports from China, however, by far the biggest share in exports are manufactured goods, which are classified by material (figure 2.6).

This analysis tells us the following about bilateral trade between the Netherlands and Vietnam:

- the share of agricultural products in exports from the Netherlands stays rather stable, whereas the share of agricultural products in exports from Vietnam is likely to decrease;

Figure 2.6. Agricultural exports from China to the Netherlands

$0

$200

$400

$600

$800

$1,000

$1,200

$1,400

$1,600

$1,800

$2,000

1995 1998 2004

in mln Manufact goods

classified chiefly by material

Animals &

vegetables oils &

fats

Crude materials, inedible, except fuels

Beverages &

tobacco

Food & live animals

Figure 2.5. Agricultural exports from the Netherlands to China

$0

$100

$200

$300

$400

$500

$600

$700

$800

$900

1995 1998 2004

in mln Manufact goods

classified chiefly by material

Animals &

vegetables oils &

fats

Crude materials, inedible, except fuels

Beverages &

tobacco

Food & live animals

Source: UN ComTrade

- manufactured products will take up a large share in exports from Vietnam;

- agricultural exports from both the Netherlands and from Vietnam are expected to grow. More specifically, exports of the same agricultural product groups. Apparently both countries export different products within the same product groups.

Figure 2.7. Exports from China to the Netherlands 1995

Figure 2.8. Exports from China to the Netherlands 2004

Next to the existing seafood partnerships, the Dutch Ministry’s partnerships can be extended to other product groups that promise much growth potential. These other products would be crude materials, like oil, flowers and plants (product group 2), and manufactured goods, like wood and cork (product group 6). A comprehensive product list is in Appendix A.

Further research is needed to see what other products, besides seafood, would be successful within the food and live animals category to devote partnership projects to.

Figure 2.7. Exports from China to the Netherlands 1995

25%

75%

agr non agr

Figure 2.8. Exports from China to the Netherlands 2004

10%

90%

agr non agr

III. THEORETICAL MODEL

In recent years, quantitative analysis of the effects of trade liberalization policies (e.g.

regional, bilateral and multilateral agreements) on economic outcomes has grown sharply. There are three sets of quantitative tools – computable general equilibrium (CGE), partial equilibrium and gravity models. This chapter will focus on these three models, the variables used in the models that predict WTO impact on trade and relevant literature.

3.1. Computable general equilibrium model

CGE models are computer-based simulations of future economies according to a specified set of policy changes (Piermartini and Teh, 2005). A general equilibrium analysis explicitly accounts for all the links between sectors of an economy – households, firms, governments and countries. It imposes a set of constraints on these sectors so that overall expenditures do not exceed income.

Income, in turn, is determined by the rate of return on factors employed in production. Consequently, the impact of a policy change will affect the whole economy, i.e. consumption, production and trade.

This model is built upon interconnected equations, reproducing the economy and simplifying assumptions. With international trade, the model will include links with other countries, which will also have their own set of equations.

There have been a large number of CGE simulations of trade liberalization. Regardless of the set of equations used in the model or the applied assumptions, usually CGE simulations show overall welfare gains from multilateral trade liberalization. However, the range of the estimated welfare gains is between 0.4 percent 7.2 and percent of world GDP in 1997 (Piermartini and Teh, 2005). These differences arise from the coverage of the simulation, the depth of the liberalization, whether all countries contribute to trade liberalization, whether dynamics are included and the structure of the model itself.

Keuschnigg and Kohler (1997) have used a CGE model to compare the effects of unilateral and multilateral trade liberalization. They found that both types of liberalization produce markedly different patterns of results. The most remarkably difference is that under multilateral liberalization export demand increased due to lower foreign tariffs, this causes terms of trade improvement and, therefore, prevents the welfare loss implied by the unilateral

¹⁷

scenario. The welfare loss under the unilateral scenario is the result of increased competition in the domestic market and the price wedge between domestic producer prices and foreign consumer prices in the export markets.

Countries acceding to the WTO would experience increases in export demand, as a consequence of lower trade barriers. Early studies conducted by the WTO estimated gains worth

17 Unilateral liberalization implies lowering trade barriers by just one country.

US$500 billion annually from the Uruguay Round

¹⁸

(UR). A study by the OECD (2003) estimated gains of US$200 billion only from agricultural liberalization. But preliminary estimates were significantly higher than estimates produced after the UR was concluded

¹⁹

. One of the most important explanations for this discrepancy was that actual commitments contained in the final agreement implied a substantially lower degree of liberalization than assumed in policy experiments conducted in those studies, especially relative to agriculture. Later studies conducted at the end of the UR on the basis of actual agreement, revised these estimates downwards.

After the UR negotiations were concluded, Hertel et al (1996) estimated a global welfare gain from agricultural and manufacturing trade liberalization of US$260 billion per year in 2005 (in 1992 prices), equivalent to 0.4 percent of expected global GDP. Much of that estimated benefit of WTO would accrue to developing countries, especially more open ones and those making the largest liberalization commitments. Estimated gain to East Asia’s developing countries, for example, was a 4.7 percent boost in GDP, primarily because of large gains from textile and clothing import liberalization by industrial countries and their own reforms. Francois et al. (1996) had come to the same conclusion; again Asian developing economies were projected to gain proportionately much more from the UR Agreements than other regions.

While CGE models are very useful in predicting economic outcomes after policy changes, these models are criticized for two reasons. First, CGE models are “theory with numbers”. In order to generate numerical results, CGE models choose parameters in an ad hoc way (typically from various earlier unrelated empirical analyses) to estimate the impact of trade policy changes on consumption, production and trade (Baier and Bergstrand, 2004). Second, ex post

²⁰

CGE models tend to considerably underestimate trade volume effects of trade liberalization, and often by “an order of magnitude” (Baier and Bergstrand, 2004).

3.2. Partial equilibrium model

A partial equilibrium model is similar to CGE models in explaining the economy after a policy change according to computer simulations. Usually focusing only on one part or sector of the economy, while ignoring interactions with other markets. All other factors that can affect this market are assumed constant, like resource constraint of the economy. A partial equilibrium model is applied when sectoral policies are analysed.

18 UR has been the start of incorporation of agricultural policies under GATT discipline. Their inclusion was considered necessary, because agricultural protectionism had become extremely distortionary by the 1980s.Main components of UR’s Agreement on Agriculture are: reductions in farm export subsidies, increases in import market access and cuts in domestic producer subsidies.

19 UR negotiations were launched at Punta del Este, Uruguay in 1986 and concluded in Geneva in 1993. Signed by Ministers in Marrakech, Morocco in 1994 and started in January 1995.

20 Ex post analysis uses historical data to conduct analysis of past trade policy effects. After estimation, the model is used for simulations, relying on the assumption that past impact of a policy may give guidance about what can be expected from a change in future policy.

A number of partial equilibrium models have been developed to simulate international trade policy changes. Take for instance the Agricultural Trade Policy Simulations Model developed by UNCTAD. These simulations of agricultural trade liberalization produce the greatest variance in results, due to different assumptions, equations and data used to build models, as mentioned earlier.

Negative results are explained by the worsened reallocation of resources within countries, because elimination of export subsidies would not necessarily improve allocation of resources while other major distortions remain in place (UNCTAD, 2003). Most researches with negative results show that net food-importing developing countries are the victims of agricultural liberalization (Francois et al, 2003; UNCTAD, 2003). Yet, other researches show positive results from trade liberalization in agriculture, Anderson et al. (2003) even showed that gains from trade liberalization in agriculture are larger than gains derived from trade liberalization in manufacturers.

3.3. The gravity model

Gravity models were first applied to international trade by Tinbergen (1962) and Pöyhönen (1963), who proposed that the volume of trade could be estimated as an increasing function of national incomes of the trading partners, and a decreasing function of the distance between them. Various combinations of macroeconomic variables, such as gross domestic product and population with geographic distance, are powerful predictors of trade potential. Gravity models have been used extensively in empirical literature on international trade (Havrylyshin and Pritchett, 1991; Rose, 2002;

Subramanian and Wei, 2003). The gravity model is one of the most empirically successful tools in economics (Anderson and Van Wincoop (2003). Historically, the most commonly applied design is shown in Box 3.1.

Box 3.1. A gravity equation

m ij ij M j m

ij

Y

i

Y z

^m

X β

^{β β}

^{( )}

^β

ε

1

2 1

0

∏

=

=

⁽¹⁾

Where:

X

ij = total exports from i to j.

β

0 = a constant

Y

i ,

Y

j = the countries’ incomes.

m m

ij M m

z )

^β

(

=1

∏

= a set of measures impacting either negatively, i.e. trade costs, or positively, e.g.

common border, on trade flows from i to j.

ε

ij = random error term.

The gravity equation can be adjusted to each specific international issue by including or excluding specific variables in the equation. Applications of the gravity equation fall into four categories of issues in international trade, namely: estimating the cost of a border; explaining trade patterns; identifying effects related to regionalism and lastly predicting trade potential. Estimating the effect of WTO membership would fall into the last category. However, the basis of gravity equations remains unchanged in all four categories, namely that bilateral trade is estimated as a function of importer and exporter GDP and bilateral distance.

Recent papers on the impact of WTO membership on trade using gravity models produce conflicting results. The coefficient of WTO membership ranges from 0.04 to 0.6 depending on the dependent variable used in gravity models and their focus: sector or aggregate level; and industrial or developing member.

There is also contrasting literature concluding that WTO has been completely ineffective in promoting world trade (Rose 2002). Rose used the gravity model in assessing the influence of WTO on world trade and concluded that bilateral trade cannot be strongly and dependably linked to WTO membership. Subramanian and Wei (2003) extended Rose’s research and found robust evidence that the institution has had a powerful and positive impact on trade. They claimed that Rose’s analysis can be misread seriously, because Rose’s analysis is incomplete on two grounds. First, on econometric grounds, Rose’s analysis needs to include country fixed effects, in accordance to Anderson and Van Wincoop (2003). Second, on economic grounds, it needs to be further elaborated to take account of the asymmetric influence

²¹

of the WTO on trade liberalization. Once Subramanian and Wei took these factors into account, they found that WTO has promoted world trade. They estimated that WTO (and its predecessor, GATT) might have increased world imports by about 44 percent or about US$3 trillion in 2000 alone.

Baxter and Kouparitsas (2006) have analysed which variables exactly determine bilateral trade flows, next to standard gravity variables. Potential trade determinants include the following:

endowments of the factors of production, including land, labour, and capital; the level of economic development; various measures of barriers to trade; exchange rate volatility; currency-union membership; and similarity of industrial structure. Baxter and Kouparitsas found two robust determinants of trade. Specifically, the product of endowments in the two countries is positively

21 Subramanian and Wei (2003) emphasize three types of liberalization asymmetries. Firstly, between developed and developing countries. Developed countries have far greater liberalization obligations than developing countries. The special treatment of developing country is codified under the principle of special and differential treatment. Second asymmetry is between developing countries that joined WTO before and after the UR. As the UR progressed, it became clear that one of its objectives was to narrow the gap between developed and

developing countries in terms of their obligations to liberalize trade barriers. This objective was particularly noticeable in defining the terms of accession of new WTO members, namely those that joined after the UR negotiations had commenced. Third asymmetry is between sectors where WTO has been effective in bringing down trade barriers and those – notably agriculture, textiles and clothing – where it has been less effective.

related to bilateral trade. Furthermore, the same stage of development

²²

appears to be a robust determinant of trade as well. Results on the other variables were mixed.

The gravity model can also be used to analyse trade at sectoral level. ITC analyses market trends extensively by calculating trade potentials for developing countries and economies in transition with the gravity model. As an illustration, ITC analysed the five major importers (US, UK, Japan, Namibia and Zimbabwe) of South African products. They found that, even though the US were already the largest importer, their imports still had high untapped potential, according to natural country characteristics, especially in the motor vehicles and textiles sectors.

All trade potential at sector level adds up to aggregate trade potential of a country. The variables in the gravity equation at sector level, however, are different from variables in the gravity equation at the aggregate level. Instead of countries’ incomes, Y

i

and Y

j ,

variables Y

i,a

and c

j,a

are used, respectively these are agricultural production in country i and agricultural consumption in country j

.

These variables will be used in this research. The variables will be further explained in chapter IV.

The advantage of gravity models is the high explanatory value of the model in explaining bilateral trade flows and that it provides an easy method to test the role other variables play in affecting trade. Despite the advantages, gravity models are often criticized, because they lack a theoretical foundation. The lack of theoretical underpinnings significantly weakens the applicability of the model. One cannot simply plug an additional policy variable into the standard gravity model and interpret the coefficient of this variable as the response of trade to a change in policy without a theoretical framework in mind. However, these foundations were subsequently developed by, among others, Anderson (1979) and Bergstrand (1998), who derived gravity models from models of monopolistic competition, and Deardorff (1998), who demonstrated that the gravity model could be derived within Ricardian

²³

and H-O frameworks.

22 Following IMF classifications reported in the World Economic Outlook (2000), countries are split into two groups, ‘developed countries’ and ‘developing countries’.

23 The Ricardian model relies on differences in technology across countries to explain trade patterns.

IV. MODEL

In chapter III researches have applied both (partial) CGE and gravity models to estimate the impact of WTO-membership on trade. CGE models are extensive models with many equations simulating the economy. Within the time span and knowledge available, these models would be too demanding. Therefore, this research uses variables of the standard gravity model, being robust determinants of trade, according to Baxter and Kouparitsas (2006) and variables taking into account any formal trade treaty, e.g. WTO and bilateral trade agreements. Since the aim of this research is to determine the foregone exports between the Netherlands and Vietnam, and vice versa, these are the only two countries in the sample. This is different from gravity models. The following chapter explains the model used in this research, after the research questions and hypotheses are stated. Then, the variables and data are clarified and finally the statistical methods are described.

4.1. Research Questions and Hypotheses

The initial deadline of WTO accession was set at 2004 by Vietnam. Since Vietnam failed to accede to WTO in that year, it is missing out on additional economic growth that would be caused by earlier WTO membership. The objective of this research is to determine the impact of WTO- membership on agricultural bilateral exports between the Netherlands and Vietnam, and vice versa, in 2004, if Vietnam would have acceded in that year already. Consequently, the first research question will be:

RQ1: What impact would WTO membership of Vietnam in 2004 have had on agricultural exports between the Netherlands and Vietnam, and vice versa, in that same year?

The qualitative hypothesis for RQ1 is that WTO membership is expected to have a positive impact on bilateral trade between the Netherlands and Vietnam, since WTO membership has had a positive impact on international trade in China (Tong, 2006). As explained in chapter I and II, Vietnam and China are quite similar, thus similar impact of WTO membership can be expected. So, Vietnam and the Netherlands would have exported more than actual exports in 2004, if Vietnam had already been WTO member in 2004. The quantitative hypothesis for RQ1 is unknown, since literature has shown many different outcomes. This increase in exports is yet going to be quantified in this research.

Another subject of interest is the possible differential impact of WTO membership on

agricultural and non-agricultural trade. Export patterns may change as a consequence of the different

impact of WTO membership on sectors. This leads us to the second research question:

RQ2: Is the impact of WTO membership substantially and significantly larger for agricultural exports in 2004?

Initially, it is expected that WTO membership impact is relatively larger for agricultural exports from the Netherlands to Vietnam and non-agricultural exports from Vietnam to the Netherlands, than non-agricultural exports from the Netherlands and agricultural exports from Vietnam, assuming that trade patterns develop according to the H-O model, i.e. Vietnam exports more labour-intensive products and imports land-intensive products. However, the literature indicates that WTO will have a bigger impact on non-agricultural trade. Since agriculture, among other sectors, is a highly protected sector, with average tariffs and non-tariff barriers well above the average for the industrial sector as a whole, and with remaining significant peak tariffs, particularly in agriculture (Subramanian and Wei, 2003). The impact differential for agricultural and non-agricultural trade is ambiguous; therefore, the hypothesis on RQ2 is undetermined.

4.2. Model Specifications

The impact of WTO membership on bilateral agricultural trade in 2004 between the Netherlands and Vietnam is determined by estimating the impact of WTO membership of China on bilateral agricultural trade with the Netherlands. Here, it is assumed that WTO membership will have a similar impact on Vietnam as on China. This assumption is supported by the similarities between Vietnam and China: both countries have similar resource endowments – scarce capital and abundant labour – and similar economic and political structures. In both socialist market economies the reform process towards WTO standards has been incremental and experimental; step by step and learning from experience. WTO accession involves a structural change, there has to be more reliance on legal commitments next to the reduction in trade barriers. For comparison some key statistics on China in 2001 (the year of accession) and Vietnam 2006 (expected year of accession) are presented in Appendix A. This research will not go into further detail on Chinese developments, since that is beyond the scope of this paper.

We assume trade in a two-country world. This assumption is to exclude the possibility of

trading with other countries, our focus is purely on bilateral trade between the Netherlands and China

as a proxy for trade with Vietnam. In most researches the gravity model is used in samples larger than

100 countries, whereas this research is only interested in the specific trade relationship between the

Netherlands and either China or Vietnam. The rationale is that as soon as the countries have decided to

trade with one another, other countries are not considered to matter any more. Hereafter, distance is

not a determinant of trade anymore. For this reason the gravity variable – distance – drops out of the

(2) (3)

model. The Baxter and Kouparitsas’ variable of “the same stage of development” is omitted for the same reason as well.

The regression equation will consist of variables from the standard gravity model, endowments of the factors of production, a WTO membership dummy, plus a bilateral trade agreement variable. The theoretical chapter has shown that these variables influence bilateral trade.

Each variable is discussed below.

The impact of WTO membership on agricultural trade between the Netherlands and Vietnam is estimated using the WTO coefficient in the regression analysis of exports between the Netherlands and China. Equations (2) and (3) are used to answer RQ1.

RQ2 is answered by running another regression on non-agricultural trade data between the Netherlands and China. In this way we can distinguish the relative WTO influence on agriculture and non-agriculture. If WTO has a greater impact on agricultural trade, then WTO coefficient for agriculture will be higher than for non-agricultural trade. Accordingly, the following equations will be used for this regression analysis:

Box 4.1. Regression equations for RQ1

ε β

β β

β β

β β

ε β

β β

β β

β β

+ +

+ +

+ +

+ +

=

+ +

+ +

+ +

+ +

=

ch nl a

ch ch

nl ch nl ch

nl a

ch a

nl a

ch nl

nl ch a

ch nl

ch nl

ch nl

ch a

nl a

ch a

nl ch

bilatra WTO

l k

edu c

Y EX

bilatra WTO

l k

edu c

Y EX

, ,

6 , 5 , 4 , 3 , 2 , 1 0 , ,

, ,

6 , 5 , 4 , 3 , 2 , 1 0 ,

,

Where:

a

=

j

EX

i_{, ,} volume of agricultural exports from location i to j.

β

k = regression parameters of variable k a

=

Y

i_, agricultural production ofexporting country i.

a

=

c

j_, consumption of agricultural products in the receiving country j.

j

edu

i, = product of education in the two countries:

F *

it

F

jt j

k

i, = product of physical capital per worker in the two countries, equivalent to

edu

i_{, .} j j

l

i, = product of arable land in the two countries, equivalent to

edu

i_{, .} j a

WTO

j, = whether country i, j are WTO members.

bilatra

_i,j= number of bilateral trade agreements between country i and j.

(4) (5)

4.3. Variables and Data

First of all, it is necessary to define agriculture. The WTO definition of agriculture

²⁴

will be extended with fishery and forestry products, in order to be in compliance with the working field of the Dutch Ministry of Agriculture, Nature and Food Quality, such that this research could be used as guideline in the Dutch Ministry of Agriculture’s projects in Vietnam. Non-agricultural production or trade is simply total national production or trade minus agricultural production or trade.

As noted before, the independent variables in the regression equation are: the variables common to the standard gravity models; production and consumption at sector level (except for distance), Baxter and Kouparitsas’ factor-endowment variables, bilateral-trade-agreements variables and a WTO dummy. The dependent variable is exports.

Time series data is collected for the Netherlands and China including Taiwan over the period of 1983-2004. Hong Kong and Macau are excluded from the analysis, because their history is different and they are more developed, which is also reflected in their trade patterns. By including Hong Kong and Macau, the WTO effect on China would be overshadowed by the economic performances of Hong Kong and Macau. Therefore, the WTO effect will be best detected in China mainland. Taiwan should have been excluded from the sample for the same reason as Hong Kong and Macau. However, all Chinese data available include Taiwan, since Taiwan is not considered to be independent. Therefore, using data on only China mainland is not possible.

Each variable is transformed in the regression equation to make sure that the statistical tests are reliable. The transformations are further explained below.

4.3.1. Exports

Since the primary focus of this research is to analyse the impact of WTO on agricultural exports from the Netherlands to Vietnam, and vice versa, exports will be the dependent variable. Data is export data and not imports. The distinction between imports and exports can be very small, since one country’s exports are the other’s imports. In other words, imports and exports are two sides of the

24 WTO Agreement on Agriculture covers the following products: HS chapter 1 to 24 less fish and fish products plus several other products within HS chapters.

Box 4.2. Regression equation for RQ2

ε β

β β

β β

β β

ε β

β β

β β

β β

+ +

+ +

+ +

+ +

=

+ +

+ +

+ +

+ +

=

ch nl n

ch ch

nl ch

nl ch

nl n

ch n

nl n

ch nl

nl ch n

ch nl

ch nl

ch nl

ch n

nl n

ch n

nl ch

bilatra WTO

l k

edu c

Y EX

bilatra WTO

l k

edu c

Y EX

, ,

6 , 5 , 4 , 3 , 2 , 1 0 , ,

, ,

6 , 5 , 4 , 3 , 2 , 1 0 , ,

Where n stands for non-agricultural products instead of a, agricultural products.

same coin, and both are expected to increase as a consequence of WTO membership. However, in practice, there is always some discrepancy between reported exports and imports. This can be explained by:

- time lags: an export reported in December of a given year could reach its destination in January of the following year;

- insurance and freight costs: usually these costs are added to the exports in the importing country;

- partner-country mismatch: the place of origin or of final destination could be reported differently by the two reporting countries (e.g. consider an export that leaves country A, is unloaded at the port in country B, and is then transported by land to country C. The exporter A could correctly state the country of final destination as country C, while the importer C could declare that the origin of the goods was country B

²⁵

.

The discrepancy between import and export should be taken into account when drawing conclusions on the basis of trade data.

This research employs the Netherlands as the reporting country, to ensure data reliability and conformity to international standards. This may have consequences for data, since exports are usually reported on a Free On Board (FOB) basis and imports on the basis of Cost, Insurance and Freight (CIF). For this reason, exports from China tend to be overvalued. On the other hand, UNSTAT noted that there are exports data that show correspondence to the data on imports in their database (UN, 2006), despite the possible discrepancy between the two.

Agricultural export flows are extracted from the UN Commodity Trade Statistics Database (UN Comtrade) by adding up sections 0, 1, 2, 4 and 6 in the SITC Rev.1 classification

²⁶

and non- agricultural trade flows are the remaining sections. The SITC Rev. 1 classification can be found in Appendix A.

4.3.2. Production

The share of agriculture in GDP is used for the production variable of agriculture, taken from UNSTATS database. It is expected that as production increases trade tends to increase, since increased supply implies lower prices, consequently higher demand. The relationship between production and trade is, therefore, positive. It is assumed that the price elasticity of demand is > 1 in absolute value, implying that lower prices generate more sales revenue.

25 The coverage of UN statistics (http://unstats.un.org.unsd/comtrade/help/uReadMeFirst.aspx).

26 SITC Rev.3 stands for Standard International Trade Classification, Revision 3.