Impactos en el campo académico

The apparel industry is widely considered the quintessential ‘starter’ industry of

export-oriented industrialisation in labour-abundant developing countries.1 _High

labour intensities of production and low entry barriers in terms of low fixed costs and well-diffused technology place it ahead of most other industries as a gateway to

1 _{In common usage, textile and apparel are often lumped together as a single industry. However, for the}

purpose of a meaningful discussion of the role of these products in the process of economic development, it is important to distinguish clearly between the two. This is because, in contrast to clothing, textile (yarn and fabric) production is more capital and skill intensive, and thus production remains largely confined to high-income (developed) or upper-middle-income countries. The focus of this paper is on clothing, although textile receives attention in the discussion on industrial upgrading within the global apparel value chain.

industrialisation for these countries (Jones 2006). However, the abilities of countries to enter this industry and expand exports purely based on their comparative advantage remained severely constrained for over three decades (from 1974) by a complex system of export quotas imposed by the major importing countries under the MFA (Keesing & Wolf 1981; Krishna & Tan 1998; Srinivasan 1998).

The MFA, which came into effect in 1974, restricted trade in textile and apparel exports from developing countries through a complex maze of country- and product- specific quotas, imposed by the individual importing countries. It went through four successive extensions of five-year periods in 1978, 1982, 1986 and 1990, with each round encompassing a wider range of products and countries.

The introduction and subsequent tightening of MFA quotas on exports from the newly industrialising countries (NIEs) in East Asia played an important role in the global spread of the apparel industry. In response to binding MFA quotas, entrepreneurs from these countries (especially those from Hong Kong and Taiwan) moved production to low-wage countries in Asia and subsequently to even high-cost locations in Latin America and Africa. Thus, many countries with relatively high labour costs and limited technical or business skills ‘benefited’ from the guaranteed market access provided by the MFA because it severely distorted the global spread of the clothing industry in two ways.

First, there was a clear developing-country bias in the implementation of quota restrictions (Srinivasan 1998). This resulted in a notable reallocation of world production from quota-constrained developing-country suppliers to relatively low-cost countries in the developed world such as Italy, Portugal and Spain, which were high- cost countries by developing country standards. Second, there were inter-country

differences in restraint levels, with a bias in favour of first comers and the politically powerful (Keesing & Wolf 1981). The three established exporters in East Asia—Hong Kong, South Korea and Taiwan—were able to maintain international competitiveness even after their comparative advantage had eroded because the ‘quota rent’ was an effective buffer against escalating domestic cost pressures on profit margins. By contrast, quotas were imposed on new exporting countries at levels well below those enjoyed by established exporters, purely based on initial rapid growth regardless of the low starting base, effectively short-circuiting the industrialisation process at a very early stage.

As a part of the Uruguay Round that concluded in 1994, the MFA was replaced

by the Agreement on Textiles and Clothing (ATC),2 which was designed to incorporate

trade in textiles and clothing into the General Agreement on Tariffs and Trade (GATT) of the World Trade Organization (WTO) so that protection could only take the form of bound non-discriminatory tariffs. The ATC put in place a program for eliminating MFA quotas in four stages in 1995, 1998, 2002 and 2004. In the implementation process, most importing countries did not go beyond the minimum liberalisation required in the first three stages, retaining the bulk of the quota restrictions to the very end of the transition period. Given this ‘back-loading’ of implementation, the final removal of quotas with effect from 1 January 2005 represented a systemic trade policy change that entailed considerable adjustment for all stakeholders. Some quantitative restraints on apparel exports from China (‘China safeguards’) continued to remain place in the European Union (EU), Turkey, and the USA after this date, as permitted

2 _{In this paper we use the terms ‘clothing’, ‘apparel’ and ‘garments’ interchangeably.}

under the China WTO accession protocol. These safeguard quotas were also eliminated by the end of 2008.

Following the phasing out of MFA quotas and the ending of the China safeguard, international buyers are now free to source apparel from any country, subject only to the system of tariffs. Therefore, the cost competitiveness is of course much more important in export success in the post-MFA era because ‘quota rents’ no longer distort market prices. However, since they are no longer constrained by country-specific quotas, buyers demand many more attributes in addition to prices, such as product variety, quality and timely delivery. Moreover, since there are no limits on the amount procured from a given country (or a firm), scale economies (the volume factor) could counterbalance export competitiveness based purely on cost competitiveness. Thus, global apparel production is likely to become more concentrated among the most capable firms in a handful of low-cost production sites (Fung, Fung & Wind 2007).

The importance of these non-price factors in export success in the post-MFA era has been further elevated by the ongoing process of lean retailing, a business strategy which has become widespread in the apparel trades in developed countries since the mid-1990s (Abernathy, Volpe & Weil 2006; Evans & Harrigan 2005; Harrigan & Barrows 2009). Lean retailing involves replenishing the range of clothes on offer on the shop floor in very short cycles (rather than seasonally, as was traditionally done), while defraying the inventory risk by holding low stocks. For products subject to rapid replenishment, direct costs related to labour, textile inputs, shipping and tariffs are balanced against the cost associated with lead times, inventory maintenance and their attended risks. In the process of lean retailing, the buyers, therefore, increasingly

require suppliers to undertake tasks such as labelling, packaging and barcoding that were traditionally done in the buyer’s warehouses or distribution centres (Abernathy, Volpe & Weil 2006; Fung, Fung & Wind 2007). Suppliers therefore need to adhere to

‘flexible manufacturing’3 in order to enable them to respond swiftly to changing

demands, while cutting batch sizes and reducing inventories.

A key determinant of a firm’s success in flexible manufacturing is the backward integration of the production process—domestic availability of high quality fabric at competitive prices. Without quota restrictions on the amount of production in a particular country, it can become cheaper for a country, which produces both textiles and clothing to compete on world markets, thus avoiding the transport costs of inputs, time delays and the management time needed to coordinate the fragmented supply

chain (Audet 2007; Audet, Tokatli & Kızılgün 2009). During the MFA era, the textile

industry did not migrate to developing country locations as fast as the clothing industry. In the post-MFA period, there are no longer artificial obstacles (quotas) to prevent the emergence of a high-quality textile capacity in developing countries and stronger clusters of expertise. While low wages can still give developing countries a competitive edge in world markets, a strong domestic textile base can now play an important role in determining international competitiveness in the fashion-oriented and time-sensitive apparel markets by ensuring a quick turnaround time.

3 _{Flexible manufacturing refers to the ability to customise a product, to produce to order, or to shift}

quickly from production of one model to another on the same line in order to serve relatively small, specialised niche segments (Abernathy, Volpe & Weil 2006).