Roger Strange, |
Jim Newton, |
The Management
Centre,
|
School of
Business,
|
International trade in silk, which has taken place for several thousand years, has undergone radical change during the last two decades. Major changes have occurred in the nature of the trade policies of both the European Union (EU) and the United States of America towards imports of silk from China. During the same period, there have been revolutionary changes in the technology and structure of silk production in China, which has been the world's dominant producer of raw silk since the late 1970s. The two are connected. The objective of this paper is to examine these changes in order to explore both the formation of EU trade policy, and the influence of practitioners operating at the industry level.
We adopt an analytical framework drawn from the field of International Political Economy (IPE), and one that focuses particularly on bargaining relationships between political entities and market participants. Some background information about the production of silk, and about historical changes in the structure of the industry worldwide, is provided, and recent changes in the structure of the Chinese silk industry are outlined and analysed. Recent developments in EU trade policy with regard to silk imports from China are considered, with particular attention being paid to the conflicting interests amongst European producers and retailers, and contrasted with the US response to the increased volume of silk imports from China.
International trade in silk, which has taken place for several thousand years, has undergone radical change during the last two decades. Major changes have occurred in the nature of the trade policies of both the European Union (EU) and the United States of America towards imports of silk from China. During the same period, there have been revolutionary changes in the technology and structure of silk production in China, which has been the world's dominant producer of raw silk since the late 1970s. The two are connected.
The objective of this paper is to examine these changes in order to explore both the formation of EU trade policy, and the influence of practitioners operating at the industry level. We adopt a framework drawn from the field of International Political Economy (IPE), and one that focuses particularly on bargaining relationships between political entities and market participants. As the changes initiated by the European Union were arguably both greater, and more controversial, than those implemented by the United States, the study concentrates on the formation of EU policy and then considers comparisons with American policy changes.
The European Union unilaterally imposed punitive quotas on the imports of silk fabrics and clothing from China in March 1994, in a sharp departure from previous practice. Whilst similar goods made from other materials, especially cotton and synthetics, had faced trade restraints for many years, the applicable quotas had been negotiated bilaterally within the framework of the GATT Multi-Fibre Agreement (MFA) that dates from 1974. The new EU trade policy was unusual in being formulated without consultation with the Chinese authorities. It was also the first time that silk had been subject to trade restraints across the European Union. Whilst several bilateral agreements had previously been in place between China and individual Member States of the European Union, the majority of EU members had allowed unfettered access to Chinese silk. Indeed, notwithstanding the existence of the bilateral agreements, the UNCTAD/WTO International Trade Centre was able to assert that ‘until 1994, silk was one of the few non-restricted textile materials in international trade’ (International Trade Centre 1997: 26). A final departure from conventional practice was that the administration of the new EU restraints was to be undertaken by the European Commission, rather than by the exporting country. The quotas were greeted with dismay both by the Chinese exporters and by the European retail trade, particularly the large multiple store groups (Watkins & Phillips 1994: 105). Eventually the Commission agreed to negotiate bilaterally with the Chinese authorities, and the quotas for 1995 and 1996 were significantly higher than those for 1994 – see Table 1. Furthermore, the responsibility for administering the quotas was passed over to the Chinese authorities (Islam 1995: 52).
Category |
Type of article |
1993 * |
1994 * |
1995 * |
1996 * |
ex 136 |
Finished silk fabrics |
1008 |
285 |
320 |
333 |
156 |
Blouses, pullovers for women & girls |
144 |
760 |
2588 |
2679 |
157 |
Various garments |
1829 |
5400 |
10250 |
10506 |
159 |
Dresses, blouses, shawls, ties etc. |
4772 |
3020 |
3950 |
3990 |
161 |
Various garments |
12224 |
10770 |
13136 |
13465 |
* Imports quantities (tons)
Notes: Categories marked by ‘ex’ cover products other than those of wool or fine animal hairs, cotton or synthetic or artificial textile products.
Sources: Watkins (1997) p.65.
The US position had been even more liberal, with totally quota-free access being granted for Chinese silk products. This too ended in early 1994, although the restrictions were far less severe and the changes were negotiated bilaterally in consultation with the Chinese authorities. The quotas for 1994 were based upon actual US imports for 1993, unlike the EU quotas where the general levels fell below actual imports for 1992. The Americans, however, left the administration to the Chinese authorities in compliance with MFA norms. One additional aspect of the new arrangements which was common to both importers was the global policy context within which these new barriers to trade were erected. The Uruguay Round of negotiations under the auspices of the GATT had been brought to a close only a few months earlier, ending seven years of debate aimed at greater liberalisation of world trade. The negotiations had resulted in the agreement to phase out the MFA trade restraints on textiles and clothing, albeit over a ten-year transitionary period.
The structure of the paper is as follows. The first section outlines a cross-disciplinary analytical framework adapted from the work of Susan Strange (1988, 1991) and John Stopford and Susan Strange (1991). The next section provides some background information about the production of silk, and about historical changes in the structure of the industry worldwide. Recent developments in the Chinese silk industry are then outlined and analysed. The following section considers the recent developments in EU trade policy with regard to silk imports from China, and attempts to provide a explanation for the changes referred to above. Particular attention is paid to the conflicting interests amongst European producers and retailers. The penultimate section then considers the US response to the increased volume of silk imports from China, and contrasts this response with that of the European Union. Finally, we draw together our conclusions about the formation of EU trade policy.
A CROSS-DISCIPLINARY FRAMEWORK
One of the main contributions of Strange (1988) to the development of IPE was to highlight the importance and role of power in international relations, whether these were inter-State relations or business relations. The objective was to include within the list of relevant variables both economic and bureaucratic/political sources of power. Central to Strange’s work was the notion that power was a unifying variable that could synthesise the study of economics and politics. The later collaboration with John Stopford extended these ideas by focusing on the relations between States and multinational enterprises. These relations were conceptualised as the outcome of a bargaining process, where power resources were used in reaching temporary or permanent agreements or ‘bargains’. The main theme was ‘triangular diplomacy’ (Stopford & Strange 1991: 19), where negotiations were conducted between States and MNEs, between States, and between MNEs. State-state negotiations have, of course, formed the traditional remit of international politics, and competition between international firms a principal concern of economics. The concept of triangular diplomacy tries to bring the three sets of bargaining relationships into a single framework, and then to draw attention to the existence of multiple bargaining relationships. Hence MNE-MNE bargains and outcomes could have an impact on MNE-State relations and so on. Figure 1 tries to illustrate the ideas in graphic form.
Value (m ecu) |
1985 |
1986 |
1987 |
1988 |
1989 |
1990 |
1991 |
1992 |
1993 |
1994 |
1995 |
Raw silk |
195.7 |
169.6 |
125.1 |
162.5 |
273.3 |
164.7 |
102.0 |
89.6 |
76.9 |
97.4 |
n/a |
Silk waste |
14.7 |
11.9 |
16.1 |
19.6 |
30.0 |
19.5 |
25.9 |
15.7 |
14.7 |
29.9 |
n/a |
Semi-finished yarn |
16.0 |
14.7 |
14.2 |
22.3 |
30.1 |
15.6 |
25.4 |
19.8 |
16.9 |
23.6 |
n/a |
Fabrics |
72.3 |
86.3 |
81.6 |
119.9 |
185.0 |
150.6 |
121.6 |
115.3 |
124.1 |
107.9 |
n/a |
Made-up goods |
10.8 |
20.9 |
26.7 |
42.2 |
79.9 |
127.7 |
223.0 |
237.6 |
332.1 |
268.8 |
n/a |
Total |
309.5 |
303.2 |
263.7 |
366.4 |
598.4 |
478.1 |
497.8 |
477.9 |
564.7 |
527.6 |
n/a |
Source: EUROSTAT, given in Watkins (1997) p.62.
The framework presented in this paper adopts the notion of bargaining power as a critical variable, irrespective of whether the bargaining is done face-to-face, through an institutional medium, or through the market-place. Bargaining power may be derived from political or bureaucratic sources, or from control over economic resources. This framework also stresses the importance of both States and MNEs as dominant actors when analysing empirical observations in international business. Strange and Stopford also placed great emphasis on the role of technology as a cause of political and economic change. Hence at least three major sources of change (States, MNEs, technology) are to be considered.
Two extensions to the framework are desirable. First, the concept of ‘States’ should be expanded to include supranational groups such as the European Union, which may add a further dimension to the bargaining processes. Second, a further conceptual extension at the firm level may aid analysis. The increasing use of outsourcing in production systems, both domestic and international, suggests that it may be useful to include the concept of agglomerations of independent or quasi-independent firms which operate as a network, rather than as a hierarchy, to produce the final good or service. When reviewing the growth of multinationals, Stopford and Strange (1991: 16) use the term New Forms of Investment (NFI) for these non-equity and contractual means of organising international production. Dicken (1998: 223) provides a useful review of the literature of the external networks of relationships which increasingly span national boundaries. Through these relationships even small, predominantly local, firms are integrated into a global production network, creating ‘the threads from which the fabric of the global economy is woven’. The existence of networks, rather than large MNEs, may modify negotiating processes, raising the power of smaller firms and potentially counterbalancing the power of the large MNEs. The analysis that follows thus emphasises a range of multilateral bargaining relationships, that include individual States and firms, together with groupings of States and networks of firms.
SILK PRODUCTION AND
TRADE:
HISTORICAL CHANGES IN A GLOBAL
INDUSTRY
In order to fully appreciate the contemporary changes in the international trade in silk, it is useful to have an understanding of both the production processes involved and the historical development that preceded current trade practices. The activities and technology required for the production of raw silk have changed little in almost five thousand years. This essential input is generated by two highly labour-intensive processes, known as sericulture and reeling. Sericulture involves the planting of mulberry trees on which silkworms feed, and the cultivation of the worms and their cocoons. Reeling takes place after the cocoons have been boiled to soften the silk when the threads are wound around a bobbin before being packed in skeins.
The raw silk is then ready for weaving, or for a process known as throwing where several threads are twisted together to form a stronger thread before weaving (Ma 1996: 331). The weaving stage in the production process first creates an unfinished fabric known as ‘greige’ or loomstate cloth, which requires bleaching and then either dyeing or printing to yield the finished fabric. This fabric may then be used to manufacture a variety of garments and other made-up goods (e.g. furnishings).
The peculiar ability of the silkworm cocoon to create such a luxurious fabric was first discovered in China around 2600 BC – by accident, according to legend. Serendipity also explains the location of the discovery, as China was the only known place where the silkworm and the white mulberry tree coexisted naturally. For almost 3000 years the Chinese managed to keep the technology a secret, thus maintaining a monopoly over world production of the commodity. Consequently China also dominated world trade in silk, which was conducted principally with the countries surrounding the Mediterranean. Chinese exports were transported by merchants who also dealt in such goods as tea, porcelain, jade and amber along the route that connected the Orient to Europe. The route came to be known as the Silk Road, as silk was the finest and most expensive of all the commodities traded along its length (Feltwell 1991).
The monopoly was broken by industrial espionage. Persian monks smuggled silkmoth eggs to Constantinople in 552 AD, defying the Chinese death penalty to do so. With the transfer of technology, production soared in Persia, and the technique of sericulture was later spread to Greece, then Italy and, by royal decree, to France. Germany and Spain later joined the ranks of the European silk producers and, two hundred years later, the English monarchy followed the French example. James I was also directly responsible for the creation of America’s silk industry, shipping silkworms to the then British colony of Virginia. The monarchy was seeking ways to improve the new colonial economies of the Eastern Seaboard and silk, which still maintained its high value, was seen as a major contributor. Sericulture rapidly gave way to tobacco in Virginia, but was successfully transferred to Pennsylvania where it flourished for over a hundred years. In the mid-1800s, however, a blight of mulberry trees damaged the raw silk industry so severely that it never recovered (Feltwell 1991). Not only America was affected, as a silkworm disease decimated European sericulture a few years later everywhere except in Italy. However, fabric weaving and finishing operations were well established in both continents and hence demand for the raw inputs remained at high levels (Eng 1984).
The silk production network was thus separated into two location-specific activities. China initially regained its hegemony in the production of raw silk, and weaving and finishing were concentrated in the West. In 1890, silk replaced tea as China’s principal export commodity and accounted for over one-third of export earnings (Li 1982). Partly as a result of this unprecedented demand, Chinese dominance of raw silk production was again threatened, this time by the concentrated expansion of sericulture in Japan. However, the main cause of Japan’s rise in the world’s raw silk trade was a combination of mercantilism and technological innovation. Japan wanted to export in order to generate foreign exchange for industrialisation. Li (1982: 197) estimates that over 40 percent of Japan’s imports of machinery and raw materials were financed by raw silk exports in the period from 1870 to 1930. In the inter-War years, the State subsidised research to improve both the yield and quality of Japanese raw silk, improving the bio-technology of sericulture and mechanising the reeling process (Watkins 1997). The result was a high degree of standardisation of output, which was then suitable for the power looms favoured by the American weaving industry. Chinese silk lacked the same consistency though it remained suitable for the European weavers who continued to rely on hand-powered technology (Ma 1996). By 1910, Japan had overtaken China as the world’s largest exporter of silk, and twenty years later Japanese exports of raw silk had soared to ten times the volume shipped in 1890. By contrast, Chinese exports had risen by only 70 percent over the same period.
The subsequent decline of the Japanese silk industry is attributable to many factors, not least technological innovation. In 1939, the Du Pont company invented nylon (Li 1982). By the end of the War natural fibres had given way to synthetics and silk stockings had been replaced by nylons. After the War, Japanese production steadily declined and domestic consumption not only absorbed available supply, but turned Japan into a net importer. Production of raw silk had peaked in 1938 at 43,150 tons – see Table 2 - but by the 1960s this had fallen to around 16,000 tons annually. The 1960s were also a time of strong domestic demand, principally for the manufacture of kimonos, when around 30,000 tons of raw silk and silk products were consumed each year (Hyvarinen 1993). This shift in demand for Japanese raw silk was partially responsible for the re-emergence of China as the world’s dominant supplier. Chinese production had been a mere 4,855 tons in 1938. By the time the ‘open door’ policies were being introduced in China in 1978, production had risen to 19,000 tons, and China had overtaken Japan. Further expansion in Chinese sericulture meant that, by 1983, China was producing 50 percent of the world’s output of raw silk.
Year |
Total |
China |
India |
Japan |
CIS |
Brazil |
Others |
1938 |
56,500 |
4,855 |
690 |
43,150 |
1,900 |
35 |
5,870 |
1978 |
49,360 |
19,000 |
3,475 |
15,960 |
3,240 |
1,250 |
6,435 |
1983 |
56,576 |
28,140 |
5,681 |
12,456 |
3,899 |
1,362 |
5,038 |
1984 |
56,129 |
28,140 |
6,895 |
10,800 |
3,999 |
1,456 |
4,839 |
1985 |
58,887 |
32,000 |
7,029 |
9,592 |
4,000 |
1,553 |
4,713 |
1986 |
61,253 |
34,700 |
7,905 |
8,240 |
4,000 |
1,664 |
4,744 |
1987 |
64,415 |
37,600 |
8,455 |
7,864 |
4,000 |
1,657 |
4,839 |
1988 |
62,611 |
34,400 |
9,300 |
6,840 |
4,000 |
1,749 |
6,322 |
1989 |
67,447 |
40,700 |
9,757 |
6,078 |
4,010 |
1,696 |
5,206 |
1990 |
71,781 |
44,000 |
11,580 |
5,800 |
4,009 |
1,693 |
4,699 |
1991 |
75,530 |
48,486 |
10,500 |
5,527 |
4,000 |
2,077 |
4,940 |
1992 |
88,749 |
60,570 |
13,000 |
5,100 |
4,000 |
2,296 |
3,783 |
1993 |
100,174 |
71,845 |
14,000 |
4,254 |
4,000 |
2,325 |
3,750 |
1994 |
100,654 |
72,000 |
13,418 |
3,901 |
3,000 |
2,535 |
5,800 |
1995 |
n/a |
76,400 |
n/a |
3,228 |
n/a |
2,448 |
n/a |
Notes: All figures are in tons.
Sources: 1938-78 Hyvarinen (1993) from International Silk Association
1983-95 Watkins (1997) from International Silk Association
CONTEMPORARY CHANGES IN CHINESE SILK PRODUCTION
The dramatic rise in China’s production capacity was an integral part of the radical changes in the global silk industry that occurred over the last two decades. These changes in silk production can be explained by reference to the IPE framework, as they were the product of a mix of political, economic and technological factors. The origins, as is well known, lie in the Chinese economic reforms that date from 1978, and which were codified in the National Plans starting in 1981. This drive to decentralise State control in specific industries focused particularly on the textiles and garments sector, given its size and economic importance (Watkins 1997). Silk was a small part of the sector in volume terms, but significant in value especially with regard to exports. Prior to the reforms, all silk trading had been centralised by the State (Glasse 1995). Furthermore silk production was concentrated at the upstream end of the production chain, in sericulture and reeling, with the result that raw silk and yarn accounted for the bulk of China’s silk exports.
By the late 1980s, two forces for change co-existed. The market reforms permitting decentralisation had encouraged new entrants, not only into sericulture, but further downstream in weaving. Joint ventures were encouraged and foreign firms, many from Hong Kong, introduced new technologies and management into China’s traditional weaving mills and sparked an increase in new weaving enterprises. Increased domestic demand for raw silk for weaving together with speculation in cocoon markets (no longer prohibited under the decentralised system) pushed up prices in 1988. The price of raw silk doubled in one year (Watkins 1997). In 1989, however, the Chinese economy was severely damaged by the fallout from Tiananmen Square. Partly as a result, the State resumed intervention in the silk industry, at first offering subsidies to silk farmers to increase the output of cocoons. The area under mulberry cultivation rose sharply to reach 805,000 hectares in 1992, more than double the area cultivated in 1987. The State also resumed centralised control both over markets for upstream production and over the export of silk products (Glasse 1995). Unfortunately, as the increased supply began to come onstream, world demand fell as a result of the Gulf War and the attendant recession. China had a glut of raw silk, based on a sericulture industry that provided income for 20 million households.
The solution to this oversupply lay in organisational and technological change within the industry in China. First, downstream production of finished products was expanded significantly, with silk replacing more traditional materials in the country’s huge garment manufacturing industry. China progressed from being the world’s largest supplier of raw silk, to also being a major exporter of silk apparel. At the same time, technological changes were incorporated into the process of fabric production in China, which drastically changed the nature of the material. The technique was known as ‘sandwashing’, which created a soft comfortable fabric characterised by a peachskin look, that was quite unlike the lustrous finish of traditional silk. The new fabric possessed two unique attributes: it was well-suited to casual and even sports wear, and it was easy to care for as it was machine-washable (International Trade Centre 1997). An American invention, sandwashed silk was introduced first in the US domestic market, where it caught a vogue for elegant, comfortable leisure wear. Later innovations in the process, however, were made in Hong Kong, where it was adapted for volume production and subsequently transferred to China (Watkins and Phillips 1994). A further critical advantage of the new cloth was that it masked imperfections in the original greige fabric so that sub-standard stocks could be absorbed and prices cut.
CONTEMPORARY CHANGES IN EU TRADE POLICY
The rapid shifts in the patterns of both silk production and silk trade from the early 1980s until the mid-1990s culminated in the imposition of the quotas described in the opening paragraphs of this paper. This section of the paper is an attempt to determine the causes of these changes in trade policy through the application of the triangular bargaining model outlined earlier.
As the most remarkable changes to the status quo were made by the European Union, this forms the most suitable starting point for this esquire. The main distinguishing features of the new policy were, first its unilateral imposition, second the very low levels of quota, and third the resentment that was generated in several powerful member countries of the European Union. In order to understand this situation, it is necessary to extend the bargaining model to take account of the role of the European Commission in trade policy. This adds an additional level, as well as a further actor, to the arena. Thus bargaining takes place between individual member states and the Commission, as well as between States, and between the Commission and non-member States.
The introduction of the EU-wide quotas was prompted by the completion of the European Single Market in January 1993. This had an influence on policy towards external trading partners as well as affecting the role of the administration in Brussels. As Glasse (1995: 96) notes ‘French and Italian silk printers and finishers had for several years been protected by national quotas. But the introduction of the single market meant that these quotas had either to be abandoned or applied to the EU as a whole’. With the dismantling of internal customs and border controls in the Union, goods could otherwise enter a quota-free member country and then be simply transshipped to France or Italy thereby avoiding national attempts at protection.
Yet despite this thrust towards centralism, individual States continued to bargain fiercely with the Commission. This was especially true of Italy with regard to silk trade. Indeed, Watkins (1997) considers that the European quota regime was a direct outcome of a campaign mounted by Italian silk manufacturers from late 1991. The Italian silk industry, based principally in Milan and Como, included the major part of Europe’s remaining silk manufacturing capacity, and employed around 200,000 people in the early 1990s (Watkins and Phillips 1994). With no commercial sericulture left, Italy has traditionally been one of the world’s largest importers of raw silk and, more recently, greige fabrics as essential inputs to its fashion business (Hyvarinen 1993). Of the other European manufacturers of silk products, which include France, Germany, Switzerland and the United Kingdom, only France had an operation of any comparable significance, largely because of an emphasis on haute couture. In the negotiations with the Commission, France sided with Italy in the attempt to impose Union-wide quotas on Chinese products (International Trade Centre 1997).
The major factors driving the Italian campaign in 1991, however, were not so much the prospect of the Single Market, but rather a sharp drop in international demand for high-quality finished silk goods, and a rise in the supply of finished garments from China – see Table 3. Italy relied heavily on exports, mainly to the United States but also to Germany and other members of the European Union. Como’s industry was especially export-dependent, with around two thirds of output destined for foreign markets (International Trade Centre 1997: 31). The recession sparked by the Gulf War had curbed demand for high value luxury goods, especially in the United States, to such an extent that Italy’s silk industry was operating at around only half of its potential capacity (Watkins 1997). As discussed above, the nature of Chinese output was also in a state of flux around this time. The traditional concentration on raw silk, yarn and greige weaves was giving way to dyed and printed fabrics and to made-up silk garments. Whilst the bulk of this output was not in competition with the quality goods from Italy and France, the changes were suggestive of, at least, a potential threat.
Volume (tons) |
1985 |
1986 |
1987 |
1988 |
1989 |
1990 |
1991 |
1992 |
1993 |
1994 |
1995 |
Raw silk |
5213 |
4963 |
4392 |
4500 |
5097 |
3402 |
2475 |
2792 |
2752 |
4006 |
3393 |
Silk waste |
978 |
1143 |
2058 |
1854 |
1943 |
1782 |
3149 |
2721 |
3076 |
4710 |
2907 |
Semi-finished yarn |
490 |
476 |
601 |
688 |
665 |
406 |
750 |
701 |
1051 |
1467 |
1051 |
Fabrics |
1324 |
1717 |
1849 |
2593 |
2858 |
2401 |
1821 |
2155 |
2241 |
2588 |
2436 |
Made-up goods |
121 |
321 |
539 |
848 |
1181 |
1794 |
2900 |
3521 |
5080 |
4683 |
2518 |
Total |
8126 |
8620 |
9439 |
10483 |
11744 |
9785 |
11095 |
11890 |
14200 |
17454 |
12305 |
In addition, Watkins (1997) has suggested that the objectives of the Italian producers were not confined to short and medium-term protection during a downturn. At a higher level was the attempt to counter China’s dominant position as the world’s monopolist in the supply of raw silk, and also the potential dominance of China as a manufacturer of garments. To counter this threat, the Italians wanted to find a means to overturn the changes in China that had led to the rise in garment production. This would require the reintroduction of centralised control in China, replacing the devolution of control that had been progressing for several years. To achieve such ends would require a much stronger front than the Italian silk industry alone could present. The Italians needed the European Commission to back their demands. What is curious about the campaign is Italy’s vulnerability to the loss of Chinese raw silk inputs. However, if the Commission were to be the principal negotiator with China, then Italy would no longer appear as the main protagonist and could be shielded from any retaliation.
Within the European Union, opposition was mounted initially by Germany. With no silk manufacturing capability, German interests would not be served by protectionist measures. This opposition resulted in a delay to the implementation of the Italian proposals during 1992 and 1993. During this period, Chinese exports of ready-made garments expanded rapidly, partly as a result of increased involvement with European retail chains. British chains especially, including the UK’s largest retailer, Marks and Spencer, had developed close links with OEM suppliers based in China for the supply of their own-brand garments. Naturally, this expansion was made possible by the quota-free access for silk products to the United Kingdom. Despite this close relationship with their suppliers, however, British firms were slower than their continental European counterparts in mounting opposition to the Italian proposal. This was surprising as the new restrictions would both damage their commercial interests by cutting supply, and would put at risk their investment of management time in the OEM relationships. With the announcement of the quotas in 1994, intra-EU bargaining became intense and resentful. British retailers were harshly critical of the measures which, given the punitive nature of the quotas, inflicted both direct losses from orders that could no longer be shipped, as well as the opportunity losses accruing from foregone future sales. One UK retailer was quoted in the London Financial Times complaining that the total EU quota for silk shirts could be filled by just a single British importer.
The severity of the conflict of interest between EU member states makes it even more remarkable that the quota levels should be set so low. Whilst bureaucratic bungling could be a possible hypothesis, other explanations may be advanced through reference to the bargaining model. The multilateral negotiations over China’s GATT membership, which had begun in 1986, were intensifying in the early 1990s as China hoped to become a founder member of the WTO. If China were to achieve its aims, the GATT Most Favoured Nation (MFN) principle would prevent further discrimination and, indeed, direct ensuing policy changes towards more liberal ends. On one level, the European Union’s action could be interpreted as introducing additional protectionism whilst there was still the opportunity to do so. However, it may also be interpreted as an attempt to gain valuable concessions from China in areas that would benefit a wider constituency of European businesses. The accession of China to GATT/WTO, particularly if China were to accede as a developing country, would allow China greater access to developed markets, whilst only slowly dismantling protection against imports. The strong possibility of a surge in Chinese imports to the European Union, coupled with the loss of bilateral bargaining power that would follow China’s accession, would both favour the introduction of a restrictive trade policy, even if it were only transitory.
Later developments lend support to this explanation. First, the rapid rise in silk imports from China that took place in 1993 acted as confirmation of China’s potential export capability. Such a capacity for expansion might not be confined to silk, or even to all types of garments, but existed in a wide number of other products, especially labor-intensive goods. More significant, however, was the sudden increase in quota levels less than a year from their inception. Britain and Germany had been joined in intra-EU negotiations by the Netherlands, and this greater degree of opposition was one factor in the revision. However, China and the EU had been involved in what Watkins (1997: 69) described as ‘intensely active negotiations’ that included silk, but also encompassed other areas of trade relations. As a result, China offered concessions on the protection of intellectual property rights. This was followed by the formal agreement in January 1995 that permitted far more generous levels of silk quotas for 1995 and 1996.
CHINA-US SILK TRADE: REACTION AND COMPARISON
No examination of China’s silk industry would be complete without reference to its biggest customer, the United States. The world’s largest importer of silk, the United States accounted for the purchase of around half of China’s total exports of finished silk goods and garments in 1993 (International Trade Centre 1997). US policy changes relating to the Chinese silk trade are therefore of considerable importance in themselves. In addition, the formulation of these changes offers a useful basis for comparison with the process within the European Union.
A major difference between the United States and the European Union was the almost total lack of a silk processing industry in the former. With the collapse the silk stocking industry during the War years, silk weaving and finishing disappeared. After the War, America became a major customer for European silk goods, especially those from Italy and France which achieved a dominant position in key segments of the US import trade (International Trade Centre 1997). US demand for raw silk, therefore, has been virtually non-existent since the late 1930s.
Chinese penetration of the US market gathered pace after the market reforms and, by the 1980s, fashion houses such as Liz Claiborne were sourcing silks from China. Between 1989 and 1993, following the changes in production in China, imports soared. Growth was spearheaded by shirts and blouses, the main product lines. In 1989, 850,000 dozen of these garments were imported. In 1993, the total had jumped to 11 million dozen. The total value of Chinese imports into the United States rose from US$369 million in 1990 to US$1.4 billion in 1993, and accounted for 71 percent of America’s total silk imports (Watkins 1997). The increase in imports was a function of lower prices, the new easy-care properties of sandwashed silk, and fashion trends that valued the use of the new fabric in casualwear.
At the American end of the production network were the fashion houses, department store groups, discount stores and mail-order houses who relied mainly on OEM production in China for their goods (International Trade Centre 1997). Whilst these operations stood to gain from a continued liberal policy in silk goods, open international markets were threatening to other segments of the American garment industry. On the one hand, the higher end of the silk business, whose product was sourced mainly from Europe, was threatened by a cheapening of the image of silk. On the other hand, the wider garment industry was faced with the threat of the cheaper silks becoming a substitute for traditional fabrics. In the latter segment, both output and jobs fell in 1993 (Watkins 1997). The actors in the bargaining arena in the United States were thus fewer than in the European Union, and the patterns of interaction less complex. The main conflict lay between the silk importers and the domestic manufacturers of cotton and synthetic garments. The former argued against the imposition of quotas on the grounds that there was no indigenous silk industry of any size to protect, whilst the latter, supported by the haute couture importers, argued for protection. The rapid increase in import penetration in the three years from 1990 lent support to the protectionist arguments, and the quotas were imposed in 1994 (International Trade Centre 1997). The quota levels, however, were comparable with the actual 1993 imports, unlike the restrictive EU levels, suggesting that the pressure for protectionism was balanced by the interests of the importers. A final mitigating factor was that the US quotas were agreed bilaterally, thus permitting the Chinese side a role in the bargaining.
CONCLUSIONS
The objective of this paper has been to explore the formation of EU trade policy, by focusing on the recent changes in EU policy with regard to silk imports from China. We have constructed an explanation for these changes by making use of an analytical framework, adapted from the work of Susan Strange and John Stopford, which stresses the importance of bargaining power between the various actors. The framework appears well suited to the complexities of trade diplomacy and indeed to other areas of international business.
We have found that EU interests have been strongly divided, as well as being aligned with specific national interests. Not only has it been the importers against the silk producers, but the bargaining has been conducted between the British, together with the Germans, opposing the Italians who were allied with the French. The creation of the Single European Market has added further dimensions to the issue. On the one hand, it has provided a technical argument in support of EU-wide trade restraints. On the other hand, the European Union has undoubtedly had higher level objectives that it wished to achieve in its trade negotiations with the Chinese, than simple protection of the indigenous silk manufacturing industry. This contrasts starkly with US policy formation, as there has been no evidence to suggest that the US quotas were imposed in order to achieve other aims.
Two final points may be made. The first is to stress the importance of the silk industry as a source of foreign exchange for China, with earnings in 1993 of $3.76bn. Furthermore, in contrast to much of Chinese manufacturing industry that is geared to export but which is dependent upon foreign capital and foreign supplies of parts etc., both the technology and the inputs to silk production are largely Chinese with the result that the net foreign exchange benefit is substantial. The second is to draw attention to another curious aspect of EU trade policy regarding imports of Chinese raw silk. In January 1980, the then European Community (EC) and China implemented an Agreement on bilateral trade in textile products. The Agreement permitted significantly increased access for Chinese textile products to the EC market, but also included a number of concessions to EC wishes including an undertaking by China (Article 11) to supply minimum guaranteed quantities of certain textile raw materials (sure as raw silk, cashmere, and angora) to the EC industry at the normal trade price (Strange 1998: 68). The Textile Agreement has been subsequently amended, renegotiated, and extended, but it still includes this minimum restriction on imports of raw silk. Why and how this requirement came to be enshrined in the Agreement, and if/how the negotiations on import quotas on finished silk goods are influenced by the existence of this requirement are questions for further research. The EU garment manufacturers clearly need supplies of raw silk from China, as China is to all intents and purposes a monopoly supplier, and this clearly puts a bargaining constraint on the process of establishing import quotas on finished garments.
Dicken, Peter. Global Shift: Transforming the World Economy. Second Edition. (New York: Guilford Press, 1998)
Eng, Robert Y. ‘Chinese Entrepreneurs, the Government, and the Foreign Sector: the Canton and Shanghai Silk-Reeling Enterprises, 1861-1932.’ Modern Asian Studies, 18.3 (1984) pp.353-370.
Feltwell, John. The Story of Silk. (New York: St. Martin’s Press, 1991)
Glasse, James. Textiles and Clothing in China: Competitive Threat or Investment Opportunity? EIU Special Report No.2638. (Textiles Intelligence Limited, 1995)
Hyvarinen, Antero. ‘Trends in the World Silk Market.’ Textile Outlook International, no.47 (May 1993) pp.9-59.
International Trade Centre, UNCTAD/WTO. Silk Review 1997: a Survey of International Trends in Production and Trade. (International Trade Centre, 1997)
Islam, Shada. ‘A New Deal: European Union Increases Quotas for Asian Exporters.’ Far Eastern Economic Review, 158.5 (2 February 1995) p.52.
Li, Lillian M. ‘Silks by Sea: Trade, Technology, and Enterprise in China and Japan.’ Business History Review, 56.2 (Summer 1982) pp.192-217.
Ma, Debin. ‘The Modern Silk Road: the Global Raw-Silk Market, 1850-1930.’ The Journal of Economic History, 56.2 (June 1996) pp.330-355.
Stopford, John M. and Susan Strange with John S. Henley. Rival States, Rival Firms: Competition for World Market Shares. (Cambridge: Cambridge University Press, 1991)
Strange, Roger. ‘EC Trade Policy Towards China.’ In Roger Strange, Jim Slater and Limin Wang (eds) Trade and Investment in China: the European Experience, pp.59-80. (London: Routledge, 1998)
Strange, Susan. States and Markets: an Introduction to International Political Economy. (Cambridge: Basil Blackwell, 1988)
Strange, Susan. ‘An Eclectic Approach.’ In C.N. Murphy and R. Tooze (eds) The New International Political Economy, pp.33-49. (Boulder CO: Lynne Rienner, 1991)
Watkins, Philippa and Pearson Phillips. ‘Global Changes in the Market for Silk and Silk Products.’ Textile Outlook International, no.55 (September 1994) pp.93-127.
Watkins, Philippa. World Markets for Silk: Forecasts to 2000. (Wilmslow: Textiles Intelligence Limited, 1997)