The Multi-Fibre Arrangement - A Thread of Protectionism

Over the years economic theory has widely documented the potential gains from trade among two or more countries. Ricardo (1963) proposed that countries that specialised in the good in which they had a comparative advantage and then traded such goods, could yield substantial benefits from this behaviour. Comparative advantage in the production of a good is equivalent to the concept that the opportunity cost to the country of producing that good is relatively low (Brenton et al, 1997). The Heckscher-Ohlin model proposes that the source of this comparative advantage lies in the countries' relative stocks of factors of production. This means that a country with a relatively abundant capital supply will find it cheaper to produce goods whose production requires much capital and little labour compared to a good calling for a more unskilled, labour-intensive production process. The theorem thus states that each country will specialise in and export the good that is intensive in its abundant factor (Williamson and Milner, 1991).

Since the end of World War II, the world's economies have paid more attention to the importance of the global market and the gains to be reaped from pursuing one's comparative advantage and engaging in international trade. Successive rounds of the General Agreement on Tariffs and Trade (GATT) have heralded increasing liberalisation of the international market. Even by 1974, tariffs on trade in manufacturing had been cut to near 6%, from an average level of 40% in the 1940s (World Bank, 1987).

However, one important exception to this trend of freeing up international trade has been the textile and clothing market. The textile and clothing industry is one whose production process is highly intensive in unskilled labour. Since this is a factor with which developing countries are relatively well-endowed, exports of textile and clothing products are argued to have been the "obvious choice for Third World countries in the drive to industrialise"¹. Yet from the beginning of the 1980s, the developed world has placed increasingly restrictive protectionist measures against their imports of textiles and apparel from developing countries under the auspices of the Multi-Fibre Arrangement (MFA).

Given that both theory and behaviour have supported the process of pursuing one's comparative advantage and engaging in free trade, one might expect that blocking trade via the MFA in the one area where the developing world has a comparative advantage must have had a devastating effect on their efforts to develop and grow economically.

This discussion explores this issue with specific reference to two developing countries - the Republic of Korea and Bangladesh. The experience of the Republic of Korea - one of the newly industrialising countries (NICs) - would seem to suggest that the protectionist forces of the MFA had little negative impact on the economy. The way in which the Republic of Korea reacted to the MFA measures will be used to outline how an economy responds to the imposition of quotas. In contrast Bangladesh appears to have been seriously affected by the MFA restrictions, in line with one's expectations, and the reasons for this will be explored. Bringing the experience of these two countries together (despite seemingly contradictory evidence) points to the harmful effects of protectionist measures such as the MFA on the ability of developing countries to gain from international trade.

From 1963 to 1976, imports of textiles and clothing by developed countries from the less-developed world increased by 14.1%, and by 1986 the developing world's share of world exports in this industry reached 33.4% (Trela and Whalley, 1990). Since production of these goods was in line with the comparative advantage of the developing countries, they were able to supply at a lower unit labour cost than the companies in the developed countries (World Bank, 1995). As a result, import-competing firms feared that this rise in imports from the cost-competitive developing world would threaten their jobs and the viability of their textile and clothing industry. These producers thus lobbied for protectionist measures against imports of textile and clothing products.

Under such intense pressure government policy makers in the developed countries (e.g. UK, US and France), conceded to the demands. Trade in textiles and apparel was not subject to the normal GATT free trade rules, until the Uruguay Round in 1991 (Grimwade, 1996).

Thus the first Multi-Fibre Arrangement was devised in 1974. It was an extension of a previous arrangement known as the Long-Term Agreement (LTA), which originated from the "Short Term Arrangement Regarding International Trade in Cotton Textiles" (STA). The MFA is made up of a series of bilaterally negotiated quota restrictions on trade in textiles and clothing between individual developed country importers and developing country exporters. Under the quota, the exporter is allowed to supply a certain volume of textile and clothing products up to a specified ceiling, and it is up to the exporter to allocate the quota allowance among its domestic producers.

Since 1974, the MFA has been renegotiated four times and each modification has brought with it increasingly restrictive measures - covering a broader range of products, and reducing any flexibility provisions in the system. In 1986, when MFA IV was negotiated, coverage was extended to additional fibres, silk, ramie, linen and jute. By 1994, MFA IV involved eight importers and 31 developing, Central and Eastern Europe countries. During the Uruguay Round, the decision was taken to phase out the MFA over a ten year period, in order to bring trade in textiles and clothing gradually into line with the rest of industrial trade under formal GATT rules and procedures (Hoekman and Kostecki, 1995). The following section will focus on the experience of the Republic of Korea under this MFA regime until 1994.

In 1945, Korea was divided into two economic units, North and South Korea. South Korea with an extremely high population density - was among the world's poorest countries (Todaro, 1997). Yet despite possessing limited agricultural and natural resources, economic growth over the past 30 years has been spectacular. From 1960 to 1994, real GDP per capita increased from US$690 to US$10,656 (UNDP, 1997). Annual growth of GNP per capita was 7.3% for the period 1965-1980, reaching 8.2% in the years 1980-1993 (UNDP, 1997).

This growth was principally due to the pursuit of industrialisation via export promotion. During the period 1965-1993, exports grew by over 20% per annum (Todaro, 1997), of which the export of textiles and clothing played the major role (Hamilton and Kim, 1990). In the early 1970s, this industry's share of total exports was 38.7%, and by 1975, it was contributing to nearly 20% of the total value added of the manufacturing sector and employing 25% of its workers (Hamilton and Kim, 1990).

In the 1980s then as with the other NICs - Hong Kong and Taiwan - exports of textiles and clothing from the Republic of Korea came under MFA restrictions. In 1981 (under MFA II), 73% of Korean exports of such goods were subject to MFA quotas, and by 1987 (under MFA IV) this had risen to 97%. A high quota utilisation rate indicates that the quotas were binding, in other words restraining the volume that suppliers wanted to export. From 1981 to 1987, these rates were over 90% with regard to Korean exports to both the United States and the United Kingdom. Estimates of the 'import tariff equivalents' of the US MFA restraints on Korea range from 30% in 1982 to 44% by 1984 (Hamilton and Kim, 1990). Given the severity of restrictions imposed on exports from the Republic of Korea, it is not surprising that MFA quotas have been cited as the biggest obstacle ever faced by Korean textile and clothing producers (Hamilton and Kim, 1990).

As noted in the introduction, the experience of the Republic of Korea highlights the typical reactions of a country to the imposition of a quota. First, since export quantities are restricted, share of the protectionist markets is expected to fall - as shown by Table 1, which deals with the two main export markets, US and EC, for the three NICs.

Table 1: Changes in the import market share of suppliers in textile products that were under binding restrictions for the Republic of Korea, Hong Kong and Taiwan in the US and EC markets, 1981-87.

Source: Erzan et al. (1990)

Second, producers seek to diversify production into goods that are not subject to quota restrictions. However, the imposition of quotas speedily followed this process such that by 1987, out of a total 111 categories of clothing textile products, 75 were under bilateral quotas.

Third, given that the MFA quotas were placed on volume rather than value of exports, product upgrading provided a means of raising the income earned on a consignment of goods. By increasing the quality and hence the value of goods, producers could gain without increasing the quantity of goods exported. Experience of developing country suppliers as a whole indicates the use of product upgrading to circumvent the quantity restrictions of quotas (Erzan et al., 1990).

Fourth, overseas subcontracting was pursued by the Republic of Korea in order to seek out other geographic locations that were not subject to the bilateral agreements. Over the 1980s, the number of foreign investments made by the Korean textile and clothing industry rose significantly, between 1983 and 1988 for example, the number of clothing projects increased from two to 21, and many of these investments were located in Bangladesh. However as soon as the suppliers in these new locations became successful exporters, MFA quotas were imposed (Erzan et al, 1990).

Fifth, during the early periods of the MFA, exporters were allowed some flexibility in the use of the quotas, for instance advance use of the next year's quota, or carrying over an unused portion of the previous year's quota. This lessened the severity of the impact of the MFA on the production process and allowed some room for Korean exporters to respond to the changing market demand conditions. Unfortunately, with subsequent renegotiations of the MFA (MFA III and IV), these flexibility provisions were reduced and finding ways to circumvent the quotas became increasingly difficult (Trela and Whalley, 1990).

However, despite these problems, during the 1980s the volume of Korean textile and clothing production doubled, while the volume of their exports trebled (Hamilton and Kim, 1990). By 1986, the Republic of Korea was second only to Italy as the largest net exporter of clothing and textiles in the world (Chisolm et al., 1986). The country succeeded in moving from near-destitution in the 1940s to being one of the 10 largest trading economies in the world (Todaro, 1997).

The one main feature that adequately accounts for the ability of the Republic of Korea to grow despite MFA protectionism concerns the country's most abundant asset - labour, which in contrast to many other developing nations, is highly educated and productive. Even back in 1970, adult literacy was already 86% of the population (UNDP, 1997). Chisolm et al. (1986) noted that in the 1980s, Korean employees worked longer hours than anywhere else in the world.

Thus Trela and Whalley (1990) argued that with this relatively high skilled labour force, and the significant amount of entrepreneurial flair therein, the country was well-equipped to diversify out of the single process, unskilled, labour-intensive production into more capital-intensive production. In this way the country engaged in a structural shift in its comparative advantage towards more skill- and capital- intensive processes (Hamilton and Kim, 1990. An example of this can be seen in the increase in recent years of imports of chemicals for producing man-made fibres and of textile machinery - reflecting the change in the nature of textile production away from unskilled-labour-intensive processes.

Perhaps then, as some economists have argued, the trade restrictions on textiles and clothing have actually encouraged the developing countries to pass through the stages of industrialisation more quickly by forcing them to adapt their comparative advantage to suit the global market conditions. Thus, they have advanced by improving the labour market quality in line with more skill-intensive production (Trela and Whalley, 1990). If this were the case however, how is it that a country like Bangladesh, which is similar in many ways to the Republic of Korea, has not managed to replicate the gains from the export-led strategy of the first NICs (Chisolm et al, 1986)?

In the 1960s and 1970s, Bangladesh , like the Republic of Korea of the 1940s was one of the poorest and most densely populated countries in the world, had a very large pool of labour and was lacking in non-energy minerals and other natural resources (Economist Intelligence Unit -EIU, 1996). From 1965 to 1980, GNP per capita contracted by 0.3% per annum, 88% of the labour force worked in subsistence agriculture, adult literacy was only 24% of the population, and life expectancy at birth was a mere 39.6 years (UNDP, 1997).

For many decades Bangladesh relied heavily on its exports of raw jute and jute products. However with the constant threat of serious flooding that can instantly destroy crops, declining jute fibre prices and a significant decrease in world demand, the contribution of the jute sector to the economy's ability to grow and develop has deteriorated (Spinanger, 1986). Thus attention turned to the role of the manufacturing sector in driving the much needed export-oriented growth.

In particular, the textile and clothing industry was focused on, its requirements being "consistent with Bangladesh's comparative advantage".² In 1978, less than twelve garment companies existed. By 1985, with the help of Korean investment, there were 450 companies in operation and 300 in the pipeline. 140,000 workers were employed, with a capacity of three million garments a year (Spinanger, 1987). The industry was earning US$116 billion in 1985, and contributed to 12% of total national export earnings (World Bank, 1995). While the exports of raw jute grew by only 2.6% between 1984-85, the exports of garments grew by a spectacular 71.3% (EIU, 1986).

The huge success of this industry reflects the ability of the entrepreneurs in Bangladesh to recognise where its comparative advantage lay - i.e. where the factor intensity (unskilled-labour-intensive) aligned perfectly with the relative factor supplies of the economy (Spinanger, 1987). However, as early as 1984, France and the UK imposed MFA quotas on Bangladesh and later the US did likewise (World Bank, 1987).

These bilateral agreements were extremely restrictive, unlike the more flexible restraints placed on the NICs in the early stages of the MFA programme. For example the US arrangement allowed only a 6% growth rate in MFA imports from Bangladesh - yet from 1981 to 1984, this measure had shown a 386.4% rate of growth (World Bank, 1987)! Furthermore this agreement was very detailed, restricting imports down to 7-digit SITC (Standard International Trade Classification) categories - for example a quota was placed not only on shirts, but on shirts made from dyed yarn in particular sizes - "so detailed an arrangement would make diversification into uncontrolled goods well nigh impossible".³ In the years 1985-87, over 90% of US imports of textile products from Bangladesh were subject to quotas, and the average rate of quota utilisation was about 95% (Erzan et al, 1990).

The immediate impact of the MFA restrictions on Bangladesh was to cause a large number of factories to close. Consequently, workers were laid off leaving thousands of females (representing 84% of the industry's workforce) destitute. Banks stopped lending due to uncertainty over the future, and investments in the industry ceased (Chisolm et al, 1986). Scarce managerial ability was absorbed in the administration of the quota allocation system, and efforts to diversify into unrestricted products were blocked by the restrictiveness of the quotas.

In more recent years, producers have looked to product upgrading as the direction for future growth of the industry. The silk and leather industries are areas where trade restrictions are minimal, and thus Bangladeshi entrepreneurs have diverted production in this direction (World Bank, 1995). In this light, surely Bangladesh can follow the Republic of Korea's experience, as the latter found ways to adapt to the MFA quotas by moving to higher quality production in more capital- and skill-intensive products?

Unfortunately, nearly 30 years after the initiation of growth in exports of textiles and clothing, Bangladesh has not benefited from anything like the growth experience of the NICs. Bangladesh is still one of the poorest countries in the world, as indicated by the comparative profile below.

Source: UNDP (1997)

Note in particular that the Bangladeshi adult literacy rate was only 37% in 1994 (UNDP, 1997). This simple statistic points to a fundamental difference between the experience of Bangladesh and that of a country like the Republic of Korea in trying to expand and diversify its exports in the face of MFA restrictions. Korea, with its highly educated labour force, was readily equipped to enter into higher quality, more skill-intensive production processes. Bangladesh on the other hand still maintained a comparative advantage in terms of unskilled labour, and had not yet "acquired the expertise to diversify".⁴ This is made evident by the situation today where an acute shortage of skilled labour has emerged as the garment industry moves towards product upgrading - in contrast to the still-abundant supply of unskilled workers (World Bank, 1995).

Hence while the Republic of Korea was ready and able to be encouraged by the MFA regime to adapt to its comparative advantage, Bangladesh was forced to seek other methods of export promotion, for instance leather and silk, where its most abundant factor could not be fully absorbed. In this way, unless one has unique advantages such as Korea's highly educated labour force, the MFA restraints "seem likely to push countries up the ladder of comparative advantage faster than market forces would take them into products which are too capital- and technology-intensive for their present resource endowments^' (World Bank, 1987).

Thus, instead of the MFA encouraging Bangladesh to speed up its process of industrialisation, as suggested earlier, it has frustrated its comparative advantage. Trela and Whalley (1988) estimated that if all developed countries' restrictions on textiles and clothing were removed, exports of these products from Bangladesh could increase by 70%. The welfare effects for Bangladesh of removing the MFA quotas were estimated at US$0.223 billion. While significant, these figures are likely to have underestimated the true cost of the MFA protectionism measures for Bangladesh, in terms of retarding its economic development (Grimwade, 1996). This is indicated by the fact that Bangladesh's poverty status in the world has scarcely improved over the last 3 decades (UNDP, 1997).

Furthermore, it is not argued here that countries like the Republic of Korea did not suffer at all from the MFA regime - far from it. According to Trela and Whalley (1988), the growth of the Republic of Korea's textile and clothing export market could have been even greater than it was under MFA restrictions - an increase of nearly 210% with an estimated welfare gain of US$0.817 billion.

It can be seen that countries do take into account where their comparative advantage lies when pursuing a development strategy. In the case of developing nations who have an abundant supply of unskilled labour, the textile and clothing industry - intensive in this factor - has played a significant role in their quest for export-led growth (Trela and Whalley, 1990). Unfortunately for the success of these development strategies the Multi-Fibre Arrangement was introduced as a protectionist measure by developed nations in the 1970s.

This investigation has looked at the ways in which an exporting country - Republic of Korea - responds to the imposition of MFA quotas on its production, in terms of among other measures, product diversification, and overseas subcontracting. The unusual ability of Korea to adapt to the quotas by moving into more capital- and skill-intensive production was observed. The case of Bangladesh, however, highlighted the fact that altering one's comparative advantage to suit market conditions is not always immediately possible. In this way, the MFA cannot be seen as a favourable force in terms of encouraging the transition to higher skill-intensive methods of production. As Bangladesh has illustrated, the country may simply not be ready. With the majority of the population working in agriculture, Bangladesh needed time to accumulate an industrial workforce before any further advancement of skill was possible (Chisolm et al, 1986).

This discussion has highlighted the retarding and growth effects of the MFA on the NICs, exemplified by the Republic of Korea, and the devastating impact of the MFA on a developing country like Bangladesh which is "attempting to exploit its primary resource - vast supplies of labour - to build itself an industrial base" (Chisolm et al, 1986, p.37). It was thus a happy moment for all involved when with the conclusion of the Uruguay Round in 1994, the MFA's thread of protectionism was finally severed.