FIRST DRAFT

RISE OF READYMADE GARMENTS INDUSTRY IN BANGLADESH: ENTREPRENEURIAL INGENUITY OR PUBLIC POLICY

by MOHAMMED ALI RASHID Professor of Economics North South University

Paper presented at the Workshop on Governance and Development organized by the World Bank and BIDS at Dhaka on 11-12 November 2006

I. INTRODUCTION The 100 percent export-oriented readymade garments (RMG) industry of Bangladesh has witnessed remarkable growth since its inception in the late 1970s. Paradoxically, this flagship industry of Bangladeshi private entrepreneurial talent took roots through the first export consignment of shirts from Bangladesh made by the state-trading agency, the Trading Corporation of Bangladesh (TCB), in the mid-1970s under countertrade arrangements and the destination was some East European countries. Subsequently, however, private entrepreneurs entered the industry and phenomenal growth took place in RMG exports from Bangladesh. Export of RMG increased from US $40 thousand in 1978 –79 to US $6.4 billion in 2004-05. The industry has also provided employment to nearly 2 million workers, most of them women drawn from the rural areas. Explosive growth of RMG exports is of course not unique to Bangladesh. The annual compound growth rate of RMG export industries in Indonesia (31.2%), Mauritius (23.8%), and Dominican Republic (21.1%) compares favourably with that of Bangladesh 1 (81.3%) over the 1980-87 period . However, while initial conditions were favourable for export growth in the countries noted above, this was far from true in the case of Bangladesh. This makes research into the factors responsible for the observed striking growth of RMG exports from Bangladesh a compelling case study in economic 2 development . Academics and researches have generally attributed the remarkable growth of RMG exports from Bangladesh to favourable external conditions, notably the Multi-Fibre Arrangement (MFA) bilateral quota system imposed by developed apparel countries, and 3 low wages in Bangladesh . The role played by supportive government policy has also been noted in these studies. However, the role played by the RMG entrepreneurs in the 4 dynamic growth of this industry has been largely overlooked . Thus, Khan and Hossain (1989) refuse to accept that the Bangladesh entrepreneurs had made any contribution to 5 the rapid growth of RMG exports from Bangladesh . This apparent disbelief in the dynamics and creativity of Bangladeshi RMG entrepreneurs seems to have sprung from the more general notion, widely held by many scholars during the 1970s and 1980s, that 6 Bangladesh lacks entrepreneurial resources . A major objective of this paper is to analyse the emergence and sustained growth to 7 date of the export-oriented RMG industry of Bangladesh , paying particular attention to the governance aspects of such growth. As indicated above, the rise of the RMG industry in Bangladesh occurred during a time when the economy’s health was in poor shape and it was plagued by various problems, both structural and policy-induced in nature, which constrained, among other things, growth of manufacturing industries in the country. How did the RMG industry overcome these problems and flourish while growth of most other industries was stymied by these unfavourable factors, many of which had their roots in poor governance? The paper attempts to answer this question by analysing the problems of governance which industries in general suffered from and then explaining how the RMG industry managed to overcome many of these problems which ensured its healthy growth.

1

In this regard, special attention will be paid to an examination of the role played by the RMG entrepreneurs, in particular how the social background of the entrepreneurs in this industry, many of whom were former civil servants, military officials and politicians, enabled them to overcome the adverse effects of misgovernance. While inappropriate government policy can spell disaster for an industry, by the same logic, sound economic policy of the government can stimulate industrial growth. Hence, a second objective of this paper is to analyse the role of supportive government policies in promoting the growth of the RMG industry in Bangladesh. In addition to policies which are supportive, deliberately weak or minimum government regulation of an industry in the form of non-implementation or weak implementation of existing government laws and regulations may spur growth of that particular industry, even through it would clearly reduce social welfare and hence can be construed as misgovernance. Has this happened in the case of the garment industry? This paper will also examine this issue. There is an unanimity of opinion among scholars that the changing structure of the global apparel industry has been uniquely fashioned by the MFA Quota system used by developed importing countries which restricted export supplies of garments to these markets from established supplying countries and thereby paved the way for the emergence of new garments exporting countries. This migration of the export-oriented garment industry from established suppliers to new suppliers was, as one would expect, directed by the existence of low wages in the new exporting countries. An important feature of the quota-restrained apparel market, which economic theory teaches us to expect, has been the emergence of quota premium or quota rent where the quota is binding. This quota rent clearly is an addition to normal profit that would occur in an unrestrained market. This paper investigates the importance of the MFA Quota system as a contributory factor to the rise of the RMG industry in Bangladesh. More importantly, an attempt is made to assess the extent to which quota rents enabled the RMG enterprises to absorb the higher cost of doing business resulting from misgovernance, a luxury which was not available to other industries. Since May 2006, the RMG industry of Bangladesh has been beset with very serious labour unrest problems which has resulted in large-scale damaging of garment factories by 8 the workers and has at times appeared to threaten the very existence of this industry . The major bone of contention between the RMG factory owners and the workers has been the allegedly low level of wages paid in this industry, particularly wages paid to unskilled workers, together with other issues like late payment of wages, lack of security of workers resulting from absence of a formal contract between the worker and the employer, nonpayment of maternity and other benefits to female workers, etc. These issues, which form part of what is commonly known as ‘compliance with social standards’, have also posed problems for Bangladesh’s RMG industry on the external front for the past few years during which time foreign buyers of Bangladesh’s garments have insisted on strict social compliance on the part of RMG enterprises in Bangladesh as a pre-condition for their importation from this country.

2

The fourth and final objective of this paper is to determine whether weak or lack of social compliance of Bangladesh’s RMG industry at present, to the extent that it is true, can be traced back to the government’s lenient and perhaps ‘indulgent’ attitude towards the country’s overwhelmingly important foreign exchange earning industry during its formative years, in the form of lack of monitoring and regulation and strict enforcement of the country’s laws pertaining to social standards. If found to be true, this can be construed as misgovernance because the government, perhaps unwittingly, would in this case have encouraged the RMG industry to move along a path of long-run unsustainable growth, the lack of sustainability originating from a disgruntled workforce and dissatisfied importers, not to speak of exploitation of labour, particularly women workers who dominate this industry. The paper is organized as follows. Section II looks at the highly regulated policy regime that was in operation when the garment industry was developing in Bangladesh. In Section III we analyse the structure and growth of the RMG industry. Section IV is devoted to an analysis of the favourable external market conditions and low labour cost that stimulated growth of the garment industry in Bangladesh. Section V considers the role played by public policy in the growth of this industry. The role of the entrepreneur is examined in Section VI. The long-run sustainability of the industry in the context of compliance with social standards is investigated in Section VII. Concluding remarks are presented in Section VIII. II. UNFAVOURABLE INITIAL CONDITIONS The experiences of the East Asian and subsequently the South-East Asian countries pinpoint the importance of propitious structural factors and, perhaps more importantly, an appropriate and conducive policy environment, both at the macroeconomic and sectoral levels, for achieving rapid industrial growth. Judged by these criteria, the initial conditions obtaining in Bangladesh during the first decade or so after its independence on 16th December 1971 was definitely not favourable for attaining a healthy pace of industrial growth. Yet, despite these odds, the RMG industry took roots in the late 1970s and began to flourish in Bangladesh while most other industries were virtually stagnating. In this section we examine the structural and policy-induced constraints to industrial development in Bangladesh during the 1970s. Subsequently, we shall analyse how the RMG industry managed to overcome these impediments which had generally obstructed the progress of other industries. Bangladesh experienced disequilibrium in both the internal and external accounts throughout the decade of the seventies. The economic situation was precarious during the first three years after independence, although it eased somewhat in the second half of the decade when the task of rehabilitating a war-ravaged economy had been more or less completed. Large fiscal deficits were the order of the day since the government had no choice but to resort to deficit financing to finance rehabilitation works and to reinvigorate the level of economic activity in the country. As a result, the money supply increased rapidly and there was very high inflation. The balance of payments position was precarious

3

in the face of a rising import bill and a low level of exports. Foreign exchange reserves 9 were at a very low level . The average annual GDP and per capita GDP growth rates during the First FiveYear Plan period 1973-78 were estimated by B.B.S to be 5.2 percent and 2.6 percent, respectively. The manufacturing sector grew at 8.3 percent per year during this time while 10 the share of this sector in GDP was about 10 percent . During the Two-Year Plan period 1978-80, the growth rates of GDP and per capita GDP declined to 3.9 percent and 1.6 percent respectively, while the growth rate of manufacturing fell to 6.1 percent. Growth rates were even lower during the first three years after independence; GDP and per capita GDP growth rates were 2.75 percent and –0.10 percent, respectively, during 1972-75. During this time, the poverty headcount index stood at 88 percent, while the average 11 inflation rate was 47 percent . Average per capita income during 1972/73 – 1979/80 was Taka 3377. The domestic savings rate was under 2 percent at this time; having declined from 8.4 percent in the 1960s, while the investment rate declined from 11.6 percent in the sixties to 8.1 percent in 1974/75, although it increased subsequently to a little over 16 percent in 1979/80. It is important to note that the private-investment-GDP ratio increased marginally from 4 percent in 1974/75 to 6.5 percent in 1979/80, reflecting hesitancy on the part of private investors to increase their investment; this trend in private investment was offset by a more rapid increase in the public investment-GDP ratio from 4.1 percent to 10.2 percent during this period. Dependence on foreign aid was very high during the 1970s; capital inflow (foreign savings) as a proportion of GDP increased from 28 percent before 12 independence to 87.6 percent in 1974/75, although it declined to 76 percent in 1979/80 . After the creation of Pakistan, modern industries were set up through many incentives and under the shelter of high effective protection. These incentives were, however, mainly aimed at the West Pakistani capitalists for industrialization of West Pakistan, and were not provided for the industrialization of East Pakistan (which now constitutes Bangladesh). The little industrialization that took place in East Pakistan was made possible by the availability of indigenous raw materials, such as jute and paper, but these were developed either under the direct ownership of West Pakistani entrepreneurs or under public sector ownership, to be transferred later to the private capitalists of West Pakistan. A large proportion of Bangladesh’s manufacturing sector was thus owned by West Pakistani industrialists. At the time of independence of Bangladesh, Bangali industrialists owned only 18 percent of the assets of modern industry, 47 percent was owned by the West Pakistani capitalists, 34 percent by the state-owned East Pakistan Industrial Development Corporation (EPIDC), and 1 percent by the non-Pakistani foreign 13 investors . During the Pakistan period, Bangladesh experienced a mild rate of industrialization. As of 1969/70, the year immediately preceding Bangladesh’s independence, the industrial structure was dominated by those industries the production in which was essentially based on domestically available raw materials. These were jute and cotton textiles, tea processing, sugar vegetable oils, cigarettes, paper and newsprint, matches, urea fertilizer, and cement. These were all traditional manufacturing activities and in general did not depict any significant rate of growth in the two decades of the country’s association with 4

Pakistan. There was very heavy concentration of manufacturing employment, production and value-added in only a few sectors. After Bangladesh gained independence in December 1971, the government formed by the Awami League under Sheikh Mujibur Rahman chose to maintain tight control over the economy, partly due to its socialist orientation. Important parts of the private economy including most banks were nationalized. Limits were imposed on private investment and on foreign direct investment. This period ended in the military coup on August 15, 1975 in which President Mujibur Rahman was assassinated. The new administration of Khondokar Mushtaq Ahmed (August-November 1975) started making moves away from socialist policies of the previous government to market-oriented ones. After an unstable period of coups and counter-coups, General Ziaur Rahman came to power and ruled the country from 1975 to 1981. Zia reversed many of the socialist policies of the Awami League administration and was generally considered to be more pro-market and laissez faire leaning even through his administration would frequently intervene in the economic development process. During the first half of the 1970s, industrialization efforts in Bangladesh were pursued within a planning framework; a wide array of direct controls and indirect economic policies were deployed to guide and influence the pattern of new investment in industrial activity in the desired directions. The major direct controls were investment sanctioning, import licensing and exchange controls, and allocation of areas of activities for private and public investment. The main indirect policy instruments were tariffs and QRs, investment incentives, policy on foreign investment, provision of industrial finance, and incentives for promoting export activity. Since the mid-1970s , however, important changes were introduced in these policies with a view to developing a broader and diversified industrial base, spearheaded by the private sector 14. Bangladesh inherited its industrial policy framework from Pakistan which focused on bureaucratic control of a largely private industrial sector with emphasis on import substitution and near exclusion of foreign investment. With independence, the new government of independent Bangladesh nationalized all industries abandoned by West Pakistani entrepreneurs and also all Bangladeshi-owned enterprises (including jute export trade, insurance, and banking) with fixed assets exceeding Taka 1.5 million. All large-scale industries, in particular, jute and cotton textiles and sugar, were totally nationalized. As a result, the public sector’s share in the fixed assets of the modern industrial sector went up to 92 percent. In the first industrial policy statement of January 1973, the role of the private sector was restricted by limiting permissible investment to Taka 2.5 million Foreign direct investment was allowed only in collaboration with the public sector. The overall policy was clearly aimed at fostering and maintaining public enterprises in large-and medium-scale industry, limiting private sector activity to only small industries. Since 1973, the government’s policy shifted towards encouraging private sector activity in manufacturing and reducing, through disinvestment, the dominant role of the public sector.

5

Since the response of the private sector to these policy changes was still very poor (presumably because of domestic and international stagflation and inadequacy of foreign exchange allocation), the industrial policy was further revised in December 1975 and important changes in policies and institutions were initiated to improve the investment climate. Investment approval and loan disbursement procedures were simplified. The investment ceiling was raised to Taka 100 million and then finally withdrawn in September 1978. The list of industries reserved for public sector investment was reduced to eight; investment in all other areas was open for private investors. Historically, an important instrument used by the government to influence private industrial investment in Bangladesh has been the import control system. Bangladesh followed an extremely restrictive import policy as a means of rationing scarce foreign exchange. With a few exceptions, licences were required for all imports. Complexity of the licensing system, deficiency in its administration, cumbersome foreign exchange budgeting procedures, rigid allocation of licences and time-consuming procedures, all resulted in inordinate delays and presented a serous obstacle to efficient industrial production. During the 1970s, the government pursued an import-substituting strategy of industrialization. Consequently, high tariffs were used to protect domestic industries. This posed serious problems for industries which relied on the use of imported raw materials and intermediate inputs. The problem was exacerbated by the existence of serious anomalies within the general tariff structure; some intermediate inputs were banned on had much higher tariffs than the final products. Together with special tariff exemptions based on end-use, the anomalous tariff structure resulted in low or even negative effective rates of protection for some industrial process and unduly high levels of protection for others. This distorted the structure of protection and fostered allocative inefficiency, often 15 according the highest protection to industries with lowest value-added . The effect of inappropriate government policy delineated above is clearly seen from an examination of the structure and pattern of industries in Bangladesh during the decade 16 of the 1970s . The size of the manufacturing sector was small and dominated by small and cottage industries. Thus in 1981-82, while large-scale units employed only 18 percent of manufacturing labour force and generated 58 percent of value-added, 82 percent of manufacturing employment and 42 percent of value-added originated in small and cottage industries. Large-scale industry was dominated by the public sector with a share of 46.1 percent in value-added and 13.7 percent in employment; the corresponding shares of the private sector were 11.4 percent and 4.1 percent. In 1979-80, textiles (which covers cotton and jute) was the largest industry with 36.3 percent of manufacturing value-added, followed by chemicals (14.1 percent of value-added), tobacco (12.7 percent of valueadded), and food (9.1 percent of value-added). These were the very industries which were dominant in the pre-independence period as well, indicating the virtual absence of the emergence of new industries during this period. It is also important to note that jute goods and tea continued to be the major manufactured export commodities, while most of the other industries were oriented towards the protected domestic market (e.g. cotton textile, urea fertilizer, paper, etc.).

6

The small size and slow growth of the manufacturing sector during the 1970s can be attributed to several factors which had constrained industrial investment, particularly private investment. These impediments to rapid industrial growth can be classified into two 17 groups: (1) structural constraints, and (2) policy-induced constraints . The structural constraints include : (i) the small size and poor growth of the domestic economy, (ii) need for finance, (iii) inadequate infrastructure, (iv) low level of technological development. The policy-induced constraints include : (i) regulatory barriers, (ii) dominance of the public sector, (iii) weak labour laws and labour institutions, and (iv) inappropriate trade policy. Too much regulation of the economy has been a major policy-induced constraint to industrial growth in Bangladesh. Private entrepreneurs have always expressed frustration about the regulatory environment and the associated bureaucratic sloth and corruption. As indicated above, the government used a variety of instruments such as investment schedules, project profiles, priority lists, and discouraged list to guide and regulate private investment. The system, while failing to channel private investment in a desired track, instead worked as a major disincentive and barrier to growth of private investment. The complexity of regulations not only made it difficult for a poorly trained civil service to make support services available, but also resulted-in widespread corruption in the administrative system which raised manufacturing costs. In sum, the poor growth of the manufacturing sector during the decade of the 1970s was to a large extent attributable to poor governance manifested in over-regulation of the economy, the bureaucratic inefficiency and corruption that accompanied government control, and poorly designed policies and their even poorer implementation. Cumbersome investment sanctioning procedures, lack of credit and its high cost, high cost of imported raw materials, inefficient import licensing procedures, militant labour unions, etc. all raised the cost of manufacturing and strongly discouraged private industrial investment. III. THE RISE OF THE READYMADE GARMENT INDUSTRY IN BANGLADESH Generally, an industry initially develops in response to domestic demand, and then subsequently turns to export once it becomes mature. The evolution of the garment industry in Bangladesh, as in most CBI and Sub-Saharan countries, has not followed this pattern. Instead of growth being spurred by domestic demand, the rise of the RMG industry in Bangladesh can be ascribed to growing demand in developed countries for cheap apparel. Having said this, one should not forget that there was a small domestic garments industry in the – then East Pakistan (now Bangladesh) during the sixties which catered to the demand for apparel in the – then West Pakistan. The Mercury shirts, a company located in Karachi, sourced a few consignments of shirts during 1965-68 made by some tailoring outfits operating in Dhaka, and then exported these to some European countries. There were a few tailoring groups in Dhaka who made a small quantity of export-quantity shirts and children’s wear on specific orders. They received orders from 18 and supplied to Karachi-based firms . However, there was little investment in this industry during those days because of the very limited size of the domestic market. It was the global trend of relocation of production of garments from high-wage to low-wage countries, together with the bilateral MFA Quota system, that acted as the main driving

7

force for the emergence and subsequent growth of the RMG industry in Bangladesh. Supportive government policies also played an important contributory role in this regard. As noted earlier, it was the state trading agency, the TCB, that made the first export of RMG from Bangladesh in the mid-seventies. The first consignment of private sector export of garments from Bangladesh took place in 1977-78 when M/S. Reaz Garments Ltd. exported men’s shirts worth 13 million French Franc to a Paris-based firm. At that time, there were only 9 export-oriented RMG units in Bangladesh. Along with Desh Garments, some other apparel-producing enterprises were set up in 1979, bringing the total number of firms in the industry to 22. By 1980, the number of firms in the industry increased to 47, but total RMG export from Bangladesh was less than US $ 1 million (Table 1). Despite the small export volume in the initial years, by the early eighties, the export-oriented RMG industry was well on its way to a historic place in the annals of industrial development in Bangladesh. During the early eighties, the government issued licences to many entrepreneurs for the duty-free importation of machinery to produce garments for export purposes. Consequently, the number of firms in the garments industry increased rapidly and reached 19 632 in 1984-85 . Export increased from US $ 1.3 million in 1980-81 to US $ 116.2 million in 1984-85, a phenomenal growth indeed, as shown in Table 1. Bangladesh exported its garments to the North American and European markets in the early eighties; at that time, its exports were not subject to MFA Quotas in these markets, hence the spectacular growth. However, the very rapid growth of imports of apparel from Bangladesh prompted the US, Canada and the European countries to impose MFA Quotas on Bangladesh’s 20 garments exports in 1985 . This had temporarily showed down growth of the RMG industry in Bangladesh; there was a restrained increase in the number of firms from 744 in 1985-86 to 804 in 1989-90. Export growth had also slowed down somewhat during 198590 period. However, the industry bounced back from 1990 onwards.

8

Table 1 : Growth in the Garment Industry of Bangladesh Percentage Year Export (in US $ million) change 1978 - 79 0.04 1979 – 80 0.25 84.00 1980 – 81 1.32 430.00 1981 – 82 3.50 62.14 1982 – 83 6.37 82.14 1983 – 84 31.57 395.60 1984 – 85 116.20 268.07 1985 – 86 131.48 13.15 1986 – 87 298.67 127.16 1987 – 88 433.92 45.28 1988 – 89 471.09 8.57 1989 – 90 624.16 32.49 1990 – 91 866.82 38.88 1991 – 92 1182.57 36.43 1992 – 93 1445.02 22.19 1993 – 94 1555.79 7.67 1994 – 95 2228.35 43.47 1995 – 96 2547.13 14.11 1996 – 97 3001.25 17.83 1997 – 98 3781.94 26.01 1998 – 99 4019.98 6.29 1999 - 2000 4349.41 8.19 2000 – 01 4859.83 11.74 2001 – 02 4583.75 -5.68 2002 – 03 4912.12 7.21 2003 – 04 5686.09 15.83 2004 – 05 6417.72 12.91 Source : Export Promotion Bureau The number of garments factories shot up to some 3000 in 2002 and by 2005 it reached 21 3560 . By 2004-05, exports had reached US $ 6.4 billion. The average growth rate of exports during 1990/91-1997/98 was almost 21 percent per year. However, the growth rate of exports declined in the 1998/99-2004/05 period mainly due to exogenous factors like flood in 1988, the global recession in 2000, and the 11 September incidents in the US the following year. The share of garments export in total export earnings of Bangladesh climbed from 3.89 percent in 1983-84 to 75.67 percent in 1998-99 and has stayed around this level since then. It is important to note that the RMG entrepreneurs were not experienced in the technical, supervisory or managerial aspects of garment manufacturing. They were investors who were initially approached by an overseas exporter or a commission agent who offered to sell the raw materials and arrange the sale of the finished product to an

9

overseas buyer. Thus, the profit was shared between the foreign buying agent and the Bangladeshi manufacturer in the early days of the industry. Subsequently, many of the large RMG units established direct links with foreign buyers and the role of the buying agent became redundant in this case. Due to their low level of skills and risk aversion, the Bangladeshi manufacturers initially concentrated production on a limited number of easy22 to-make items with a minimum value-added . The woven RMG industry of Bangladesh has been heavily dependent on the use of imported inputs, particularly fabric, because of the above-noted trend of foreign buying houses and direct importers to specify the sourcing of fabric and also due to the inability (during the early days of the RMG industry) of the local primary textile sector (PTS) to supply fabrics of the required quality specifications. As a result, local value-addition in the woven RMG sector ranged between 25-30 percent in the eighties but subsequently rose to 23 reach about 40-45 percent in more recent times . Thus, net foreign exchange earning of the RMG industry during the eighties and early nineties was much lower than the gross foreign exchange earnings reported in Table1. An important development in the RMG industry of Bangladesh was the growth of knitted garments export in the early 1990s. The garment industry was almost completely dominated by woven garments during the 1980s. Export of knit wear to the tune of US $ 131.20 million was the first major consignment of knit RMG export from Bangladesh. After this, knitwear export from Bangladesh grew rapidly; during the ten year period 1993/94 – 2003/04, the annual growth rate of knitwear exports was about 23 percent, which was higher than the annual growth rate of 10.6 percent achieved by woven garments export during this period. The share of knit garments exports in total apparel export from Bangladesh rose from 17 percent in 1993-94 to over 43 percent by 2004-05, while the share of woven garments gradually declined. While woven garments exports of Bangladesh are concentrated in the North American market, knitwear exports are mainly sold in the EU market. Local value addition in the knit sector is about 70 percent. According to the BGMEA, there were 4300 member firms in 2004-05 of which 2275 were woven garment units, 700 were knitting factories, and 525 were producing sweaters; however, 1300 of these units were closed. In addition, there were 560 units which were exclusive members of the Bangladesh Knitwear Exporters and Manufacturers Association (BKMEA). Thus, the total number of active firms producing garments in Bangladesh in 2005 was 3560, of which 47 percent was woven units, 42 percent knit units, and 11 percent producing sweaters. Total employment generated by these enterprises is 24 estimated to be 1.9 million production workers, or 2 million if all employees are included . Most of these workers (an estimated 80 percent) are women drawn from the rural areas. The distribution of employment among different size-class of firms is important from the perspective of the present study. EPB data, as cited in World Bank (2005), show a relatively high degree of export concentration at the firm level. Of 2387 exporting RMG units in 2004, the top 500 firms exported 74 percent of total garments export, while 81 percent of total RMG exports was made by the top 650 firms. These firms belong in the large category (production capacity of 5,000 to 10,000 dozens per month). The remaining

10

19 percent of garments export was made by 1737 firms which can be considered as small (production capacity of less than 5,000 dozens per month). The large firms source their own fabric and sell directly to foreign buyers, while the medium-sized firms work on a cutand-make (CM) basis for importers or buying agents. The small firms either sell to buying agents or work as sub-contractor for the larger firms. This means that out of the total number of 3560 firms in the RMG industry, about 1200 firms were either not exporting at all or were working as subcontractors. The same EPB firm-level data show that 1.2 million workers were employed in the 2387 firms that were exporting garments. This means that about 0.8 million workers are employed in the 1200 firms that are engaged in sub-contracting activity (i.e. they are not direct exporters) and hence are not required to fulfil the social compliance conditions of foreign buyers. IV. FAVOURABLE BANGLADESH TRADE ENVIRONMENT AND LOW LABOUR COST IN

Two factors which played important roles in the growth of the RMG industry in Bangladesh were : (i) a favourable exporting environment created by the MFA Quota system, and (ii) low wages of Bangladesh. How these two factors have contributed to the rise of the garment industry in Bangladesh will be analysed in this section. Up until the early 1960s, the apparel industry in the developed countries like the US and UK relied on domestic subcontracting; cutting and stitching operations were subcontracted to small garment factories mostly relying on the use of cheap female labour, while large-scale merchandising was undertaken by larger firms. Subsequently, as industrial wages began to rise, the apparel retailers in the developed countries found it more profitable to relocate production to lower-wage developing countries; in most cases, such out-sourcing took the form of subcontracting arrangements between the retailer in the developed country and the garment manufacturer in the developing country. The retailers in the developed countries placed work-orders to the off-share garment manufacturers, often through buying agents, and they also helped the garment makers in various ways to produce and ship the merchandise. Such sub-contracting reduced the risk of doing business with foreign partners since it did not require any foreign direct investment. In essence it was a triangular trade between the garment manufacturers and the foreign buying agent on the one hand, and the retailer and the buying agent on the other. It was in the US that the process of relocation of the apparel industry first began when domestic producers found it difficult to compete with apparel imports from low-wage developing countries. The American producers first moved production to low-wage neighbour Mexico, and then subsequently to other low-wage countries like Japan in the 1950s, and a few years later to East Asian NICs, namely Hong Kong, Singapore, South Korea and Taiwan who became large exporters of garments in the 1960s and 1970s. As wages began to rise in the NICs, production was outsourced to other developing countries where wages were lower, like Philippines, Indonesia, Malaysia, China, India, Sri Lanka and Bangladesh. A similar pattern of outsourcing of production to low-wage developing

11

countries is also observed for the European countries. Like the US, the European producers moved to low-wage developing countries like India, China, Bangladesh and the North African countries. This pattern of relocation of apparel production arising from wage differentials was hastened further by the Multi-Fibre Arrangement (MFA) under which a developed garment importing country imposed quota restrictions (in addition to tariffs) through bilateral negotiations with an exporting country. The MFA Quota system, which was approved by the GATT even trough it was a violation of its principle of non-discrimination, was designed to protect the domestic textile and clothing industry of the developed importing countries. Since 1974, global textile and clothing trade has been regulated by the MFA until its complete phase-out on 31st December 2004 under the Uruguay Round Agreement on Textile and Clothing (ATC). Even through the MFA was primarily designed to protect the apparel industry of the developed importing countries like the US and the EU countries, it also resulted in the spread of the export-oriented garment industry from countries whose exports were restrained by MFA quotas to quota unrestrained low-wage developing countries. In the seventies, large garment exporters like Hong Kong, South Korea, Taiwan and Singapore came under increasing MFA Quota restrictions in the developed importing countries. Also, as noted earlier, at this time these countries were experiencing rising labour costs. To overcome these two disadvantages and to retain their shares in the global apparel market, these countries (and other countries in a similar position) relocated their production, sometimes under joint-venture, to those countries that enjoyed quota-free status. As an LDC, Bangladesh enjoyed quota-free status in the 1970s, and wages were also low. These two features attracted ‘quota-hopping’ foreign producers to Bangladesh who collaborated with Bangladeshi entrepreneurs to export garments to North America, the EU, and elsewhere. Daewoo, a major South Korean apparel exporter, collaborated with Desh Garments of Bangladesh in 1980 and this resulted in the first major consignment of RMG export from Bangladesh. The simple technology of making garments, basically involving ‘cutting and making’ (CM), together with relatively low investment cost, made it easier for the Bangladeshi entrepreneurs to move into this industry, and rapid growth took place, as mentioned earlier. Even though Bangladesh enjoyed quota-free status in the North American market, its exports were subject to the payment of import duties. However, in the EU market, Bangladesh was not only exempted from MFA quotas but it also enjoyed preferential market access under the EU’s Generalised System of Preference (GSP) which allowed duty-free access to Bangladesh’s garment exports. Thus, Bangladesh’s exports of garments enjoyed quota-free and duty-free access to the EU market. This was another important factor which contributed to the emergence of the RMG industry in Bangladesh. These favourable external market conditions, together with low labour cost, were two important factors which favoured the rise of the RMG industry in Bangladesh. As noted above, both the number of firms in the industry and the value of export grew rapidly.

12

However, the fabourable market situation was jolted in 1985 when several importing countries, most notably the US, imposed MFA quotas on Bangladesh’s garments exports. This resulted in a temporary slow-down in the growth of the number of firms in the industry and also export value. Between 1985-86 and 1988-89 the number of firms increased from 744 to 754, which is a much slower growth compared to 1979/80 – 1984/85 period when the number of firms increased from 47 to 632. The average growth rate of export value also fell from 191 percent during 1979/80 -1984/85 to about 48 percent between 1985/86 – 1988/89. But growth picked up from 1989 onwards. One reason behind this was that even after the imposition of quota in 1986, the buyers preferred Bangladesh because (i) it had very large quota, and (ii) its FOB price for garment of a particular specification was lower than the price offered by most other supplier-countries. Structural problems along with policy-induced constraints had obstructed industrial growth in Bangladesh in the seventies and early eighties. How did the RMG industry overcome these problems to register remarkable growth while other industries had stagnated? Part of the answer lies in the favourable external market situation in the case of garments, which was not the case with other industries of Bangladesh. Other factors responsible for this impressive growth of the RMG industry were low labour cost, supportive government policy, and entrepreneurial talent of the RMG producers. The influence of each of these factors will be analysed in turn and in the process the role of governance will also be highlighted. One factor that had hindered industrial growth in the seventies and eighties was the small size of the domestic market. Other than jute goods and tea which were export industries, most of the other manufactured goods were produced for the protected domestic market. Growth of these import-substituting industries was hindered by the small size of the domestic market. Growth of the two major export industries, jute manufactures 25 and processed tea, was low because of declining would demand . World demand for jute goods was declining primarily due to the development of synthetic substitute of jute fibre, while demand for tea was low mainly due to the inferior quality of Bangladeshi tea compared to say Sri Lanka and Kenyan tea. Since the RMG industry catered to the demand of a global market, its growth was not constrained by the limited size of the domestic market. Moreover, unlike jute goods and tea, world demand for Bangladesh’s garments was growing rather than declining. This was particularly true when Bangladesh’s garments exports were not restrained by MFA quotas. Even when quotas were introduced in 1985, these quotas were large and growing 26 annually . Hence, favourable external market conditions was an important contributory factor behind the growth of the RMG industry in Bangladesh. At this point it should be noted that the Bangladesh government of that period had some positive role to play in the fixation of liberal quotas for Bangladesh through negotiation with the US and other major apparel importing countries, and this may be viewed as a positive contribution of governance towards the growth of the RMG industry. As important feature of any quota regime, including MFA quota, is the emergence of quota rent or quota premium. Quota rent is a distortion introduced by quota restrictions and

13

raises the market price above what it would have been in the absence of a quota . That is, the realised market price includes the quota premium if the quota is binding. The more restrictive a quota is, the higher will be the quota rent. Bangladesh’s exports of apparel were restrained by MFA quotas from 1985, and hence in the post-1985 period till 31st December 2004 (when MFA quotas were completely phased out under the GATT/WTO ATC), Bangladesh’s garment exporters were enjoying quota rent of varying magnitude for different categories of garments. An important issue in this regard is the extent to which these quota rents helped RMG exporting firms to overcome some of the problems originating from poor governance at that time. To get some idea of the impact of quota rents, we use the data on quota prices of different categories of garments exported by Bangladesh in 2003, which is cited in Annex 28 Table 2 of World Bank (2005) . The major garment items exported by Bangladesh are shirts and trousers; Table 2 shows that the quota price of these items are quite high. The mean quota price was US $ 6.97 per dozen in 2002 when phasing out of MFA quotas under the ATC had progressed sufficiently and the global apparel market was less quotarestrained than in the 1980s and 1990s. This implies that the quota prices would have been higher in the 1980s and 1990s when the market was much more restrained by MFA quotas. However, to get some idea about the magnitude of quota rent earned by Bangladesh’s garment exporters in say 1989/90, we multiply the volume of production of 29 garments in that year by the mean quota price in 2002 . It is found that the total quota rent earned by Bangladesh’s garment exporters in 1989-90 would have been approximately equal to US $ 135-140 thousand, or Taka 4.75 – 4.90 million (using the – then exchange rate of US $ 1 = Tk 35). Table 2 : Average Quota Prices of Selected Garments Items Exported by Bangladesh, 2002 Category Unit Description Average Quota Price (US $) 1.06 13.77 11.61 3.76 11.56 21.65 5.08 0.40 1.62 0.36 9.31 6.97

27

237 Dozen Play Suits, Sun Suits 334/634 Dozen Other Coat MB 335/635 Dozen Coat W&G 336/636 Dozen Dresses 338 Dozen Knitted Shirts, M&B 347/48 Dozen Trousers, Slacks & Shorts M&B 351/651 SME Night wears & Pajamas 352/652 Dozen Underwears 638/639 Dozen Knit Shirts, M&B 641 Dozen W&G shirts & Blouses, not knit 647/648 Dozen M/b, W/G Trousers, Breeches & Shorts MEAN Source : Annex Table 2, World Bank (2005)

14

Thus, a substantial amount of quota rent was available to RMG entrepreneurs to absorb the high costs of doing business resulting from misgovernance and to spend on ‘greasing the wheels’ of an inefficient and corrupt bureaucracy. Such ‘extra money’ was available only to garment exporters to spend as ‘speed money’ to get things done in a highly regulated economy and was not available to other industries in Bangladesh. This, at least partly, explains why the garment industry alone thrived in Bangladesh during the 1980s when other industries, mostly import-substituting in nature, stagnated in the regulated environment. The RMG industry is a low-technology and labour-intensive industry. Hence, global relocation of this industry has been partly dictated by the availability of cheap labour in a potential supplying country. Being a labour-surplus economy, Bangladesh has always enjoyed low wages. The comparative wage position of the Bangladeshi apparel industry in 1981, the time when global relocation of the RMG industry was occurring at rapid rate, is shown in Table 3. The wage advantage of Bangladesh is clearly seen from the table. The large wage differential between Bangladesh on the one hand, and South Korea, Hong Kong and Thailand on the other, explains why these (and other countries) relocated their garment production to Bangladesh in the early 1980s. Table 3 also shows the comparative wage advantage (in terms of hourly wages) of Bangladesh with respect to its South Asian competitors in the global market like India, Pakistan and Sri Lanka even in 1994. Table 3 : Wages in the Apparel Industry in Selected Countries Hourly Wage in Country Annual Wage in 1981 (US $) 1994 (US $) Bangladesh 305.7 0.24 South Korea 2096.9 5.00 Hong Kong 4121.5 3.90 Indonesia 664.9 0.40 Malaysia 1363.4 1.20 Thailand 1708.3* 1.00 India 626.0 0.60 Pakistan 1228.4* 0.40 Sri Lanka 0.45 Nepal 0.25 Source : Islam (2001) for annual wage date, and Siddiqi (2004) for hourly wage data. Note : * The data for Thailand is for 1982, and for Pakistan it is for 1985.

There is another dimension of the wage structure in the RMG industry in Bangladesh which deserves attention. Wages in Bangladesh’s apparel industry have been lower not only in comparison to other competitor countries, but also in comparison to most other domestic industries. Thus, for example, a comparison on the basis of wage data provided in the Statistical Yearbook of Bangladesh, 1998 shows that the average monthly wage of skilled RMG factory workers was 1.4 to 2 times lower than that of similar factory 30 workers in the textile and other sectors . Real wage indices also show the comparatively

15

low wages in this sector, as seen from Table 4. Furthermore, UNIDO data show a very modest increase of 11.5 percent in annual wages in the apparel industry of Bangladesh over the period 1981-1992, from US $ 305.7 to US $ 340.9, i.e. an average annual 31 increase of about one percentage point . These factors appear to have been primarily responsible for the lower labour cost in the garment industry compared to most other industries in Bangladesh. First, the garment industry is characterized by a predominance of female workers who are prepared to accept lower wages than their male counterparts due to their lower opportunity cost of time. Second, the minimum wage fixed by the government has been lower for the garment industry compared to other industries. Third, garment entrepreneurs have succeeded in reducing labour costs by generally depriving workers of “decent” working conditions through side-stepping national labour standards. Each of these factors is discussed below. About 80 percent of workers in the garment industry are women. It has been generally argued that entrepreneurs in the garment industry have preferred to hire more women because of prevailing beliefs that women are nimble and patient, more controllable than men because of their docile nature, less likely to join trade unions, and better at sewing. In addition to these reasons, another important reason why women have been preferred is that they are willing to accept lower wages than men. In most cases, employment in the garment industry has given them the first opportunity to earn wages, which are typically several times higher than what they could have earned doing domestic 32 service, one of the very few other job opportunities open to them .

16

Table 4 : Real wage Indices of Industrial Workers (all employees) (Base : 1985-86 = 100) BSIC 1986 Name of Industry code 1989-90 311-312 Food manufacturing 133 313 Beverage industries 77 314 Tobacco manufacturing 71 315 Animal feed manufacturing 80 321-322 Textile manufacturing 112 323 Wearing apparel (except footwear) 86 324 Leather and leather products 96 325 Leather footwear (except rubber & plastic) 44 326 Ginning, pressing and baling of fibres 104 331 Wood and wood cork products 102 332 Wooden furniture and fixture manufacturing 50 341 Paper and paper products 107 342 Printing and publishing 124 351 Drugs and pharmaceuticals 102 352 Industrial chemicals 96 353 Other chemical products 96 354 Petroleum refining 174 355 Misc. petroleum, cool products 138 356 Rubber products 79 357 Plastic products NEC 119 361 Pottery, china and earthenware 128 362 Glass and glass products 115 369 Non-metallic mineral products 111 371 Iron and steel basic industries 73 372 Non-ferrous metal basic industries 81 381-382 Fabricated metal products 83 383 Non-electrical machinery 92 384 Electrical machinery 73 385 Transport equipment 86 386 Scientific measuring instruments & equipment 155 387 Photographic and optical goods 85 393-394 Photographic and optical goods 89 Total All industries 112 Source : C.M.I., BBS

Indices 1990-91 124 61 47 78 99 87 94 46 94 89 54 96 89 117 81 53 168 102 75 106 116 55 91 90 65 84 101 66 88 125 89 84 96

1991-92 143 99 48 78 98 84 99 39 91 110 59 127 118 120 91 60 197 109 81 150 83 128 118 88 101 88 90 66 88 131 108 78 96

Under such circumstances, women workers in the RMG industry have been prepared to accept lower wages than what labour market conditions would seem to warrant, and RMG industry employers have taken advantage of this. The Minimum Wage Board set up by the government periodically fixes the minimum wage for different industries. For some unknown reasons, the average minimum wages

17

fixed for the RMG industry has been the lowest (Table 5). This has allowed the owners of garment firms to pay their workers lower wages than in most other industries. Table 5 : Average Minimum Wages of RMG Industry as Percent of Average Minimum Wages of 11 Selected Industries Industry Minimum Wage of RMG Industry as % of Minimum Wage of This Industry Jute Bailing 83 Match 83 Re-Rolling 70 Printing 92 Cinema 86 Cold Storage 72 Pharmaceutical 79 Shoe 83 Petrol Pump 77 Fishing Trawler 57 Road Transport 72 Source : Siddiqi (2004) Low labour costs in the garment industry of Bangladesh reflect not only low wage rates, but also low non-wage benefits and poor working conditions. A number of studies have shown that violation of Factory Act 1965 and Factory Rules 1979 are widespread in 33 the garment sector . Labour law requires that all worker receive an appointment letter, since this protects labour rights; studies however show that only 5 percent of garment workers, most of them men, receive this document. Violation of this law has not only resulted in widespread job insecurity in the garment industry, it has also enabled garment factory owners to pay lower wages to workers since a worker without a formal appointment letter is treated as a ‘casual’ worker and hence paid lower wages; it has also made it possible for employers to dismiss workers without prior notice, which is illegal. Studies also show that a large number of women garment workers, estimated to be more than 32 percent, did not receive the minimum wage of Taka 930 as fixed for the lowest category workers (helpers) by the Minimum Wage Ordinance, 1994. Female garment workers are also deprived of any severance payment even though they are frequently dismissed. By depriving the workers of the minimum wage and severance pay, garment employers have been able to reduce labour cost and increase their profits. This has been a major source of conflict between employers and employees in the garment industry. Studies show the widespread use of child labour in the garment industry of Bangladesh till the mid-nineties even through it is a violation of the core ILO labour standard relating to minimum age of a worker, which was ratified by the Bangladesh government. These child workers were paid wages below the already low wages prevalent in the garment industry. This was another way in which labour costs were kept down by the garment enterprises even through it was illegal. According to the BGMEA, the practice

18

of employing children below the age of 14 has ceased since 1995 when a project was taken up by the BGMEA in collaboration with the ILO and UNICEF to rehabilitate child 34 labour from the garment factories of Bangladesh . Garment factory workers, particularly the female workers, have also been deprived of non-wage benefits by their employers, and this has also helped the latter to squeeze labour costs. Under the Factory Act 1965, formal sector workers are entitled to various non-wage benefits including accommodation, transport, meal and medical allowance, bonuses, pension, provident fund and insurance benefits. Female workers have been deprived of maternity benefit payable under the Maternity Benefit Act of 1950. Due to the absence of a formal appointment letter, however, garment industry workers, particularly 35 women, are largely deprived of these benefits . In sum, MFA quota system and low labour cost were two important factors which were responsible for the emergence and growth of the RMG industry in Bangladesh. The quota rents associated with MFA quotas had aided Bangladesh’s RMG entrepreneurs to absorb, at least partly, the higher costs of production resulting from bureaucratic inefficiency and corruption in the highly regulated economic environment prevailing in the country during the late 1970s and 1980s when the industry was in its growing phase. At the same time, low labour costs in Bangladesh had attracted quota-restrained foreign garment producers to come to this country and act as a ‘catalyst’ in the rise of the garment 36 industry in Bangladesh . Low labour cost in Bangladesh’s garment industry was partly the result of surplus labour situation in the country, and partly originated from the violation of national and international labour standards by the garment factory owners. This nonobservance of labour standards dictated by national labour laws was possible because of the inability of the government to adequately monitor the proper implementation of relevant labour laws by garment factories, and also partly attributable to the corrupt nature of some labour ministry officials who deliberately overlooked these anomalies after being paid some bribe money. On both counts, this can be viewed as a failure in governance which resulted in robbing the garment workers of social protection. V. SUPPORTIVE GOVERNMENT POLICIES In contrast to the public sector-led import-substituting industrialization strategy pursued during the first few years after independence, the industrialization philosophy of the government changed rather dramatically from the late 1970s when the emphasis was on export-oriented growth to be spearheaded by the private sector. Towards this end, various policy reforms were implemented in the 1980s and 1990s. Some of these reformed policies contributed considerably to the growth of the RMG industry in Bangladesh. The most significant policy reforms in the industry sector were introduced through the New Industrial Policy (NIP) announced by the government in June 1982. The aim of the NIP was to stimulate industrial development through the private sector’ and to that end it made fundamental changes in the industrial policy environment and promotional instruments. The highly regulated policy environment of the previous period began to give way to a process of policy decontrol and the dominance of market forces in the allocation

19

of resources. The NIP was revised in 1986, and the Revised Industrial Policy (RIP) further strengthened the process of liberalization set in motion by the NIP. The Industrial Policy 1991 reiterated the objectives of the NIP and the RIP of achieving a rapid expansion of the private sector and transforming the economy into a competitive market economy. The investment sanctioning procedure was gradually simplified and liberalized. The sanctioning authority delegated to government departments and financial institutions was increased substantially, while controls exercised by the Department of Industries and the Chief Controller of Imports and Exports (CCIE) were streamlined. Automatic registration was provided to projects financed from non-government sources, and from 1991 no government sanction was required for private investment in ‘free’ sectors, provided the industry was set up with the entrepreneur’s own fund. These reforms made it much easier for new investors to enter the RMG industry. Import licensing procedures were liberalized in the 1980s. In 1983-84, imports financed with cash licence at official exchange rate were gradually reduced and industries were required to import the rest of their requirements from the secondary exchange market (SEM) where the exchange rate better reflected the opportunity cost of foreign exchange. Even though this increased import cost, garment manufactures (and other entrepreneurs) benefited because of the hassle-free availability of foreign exchange which did not require the payment of speed money. Import licensing was shortly abolished and imports were permitted against letter-of-credit (L/C) authorization forms to be accepted by banks designated by the industrialists. A very important change in import regulations was the shift in 1986 from a “positive list” to a “negative list” supplemented by a “restricted list”. This greatly benefited industries dependent on imported inputs, like the RMG industry. During the 1980s, a number of incentives were introduced to encourage export activities. Some of them were new like the Bonded Warehouse Facility (BWF), while others 37 like the Export Performance Licence (XPL) Scheme were already in operation and were improved upon. Also, rebates were given on import duties and indirect taxes, there were tax reductions on export income, and export financing was arranged. Under the XPL scheme, exporters of non-traditional products received import licences for specific products over and above their normal percentage allotment based on the f.o.b. value of their exports. The transferability of these licences (‘called Import Entitlement Certificate) was an added incentive to exporters as these could be sold at a premium in the market. In 1985-86, the XPL Scheme was replaced by the XPB Scheme where the import entitlement rates were raised and the scope of the scheme was expanded to cover nearly all exports. Under the Duty Drawback System, exporters of manufactured goods were entitled to get refund of duties and taxes paid on imported inputs used in export production, and also all excise duties paid on exported finished goods. For certain fast-moving items such as RMG, a notional system of duty payments was adopted in 1982-83. Under this system, exporters were exempted from paying duties and taxes on imports used in export production at the time of importation, but were required to keep records of raw and

20

packaging materials imported. The duties and taxes payable on the imports were kept in a suspense account. Liability to pay the amounts in suspense were removed on proof of exports. Exporters, however, experienced difficulties in taking full advantage of the provisions of the Duty Drawback Scheme because of complicated procedures. So, the government introduced the Special Bonded Warehouse (SBW) Scheme which was arguably the most efficient system to exempt exporters from import duties and taxes. This scheme was initially allowed only for garment manufacturers who exported 100% of their output, although the facility was subsequently extended to 100% export-oriented specialized textile and leather goods manufacturers. The SBW Scheme, which was administered through the use of passbooks, input-output coefficients and stock books, permitted fairly speedy administration of the import procedures and guaranteed garment exporters duty-free imports of their import requirements. Thus, the SBW Scheme provided a level playing field for Bangladesh’s garments exporters vis-a-vis their competitors in other countires. Furthermore, it led to some cost reduction as funds were not tied up for paying duties. Export credit (pre-shipment and post-shipment working capital) for non-traditional export items like garments was available in local currency from commercial banks at concessional interest rates. In 1986-87, additional export financing facilities were introduced through the provision of the back-to-back L/C system. The problem of financing working capital was minimised for garment manufacturers with the introduction of this facility. The back-to-back L/C operates in a way that does not require the garment manufacturers to invest money to open import L/C or make payment to the fabric suppliers from their own resources. Payment is made by the original buyers (retailers) against what is called ‘Master L/C’. All risks are covered by the Master L/C opened by the original buyers of the garments. The operation of the back-to-back L/C ensured the garment exporters almost 70 percent of the working capital they needed. Most RMG exporters agree that the operation of the back-to-back L/C is the single-most important factor responsible for the phenomenal growth of garment export from Bangladesh after the 38 favourable international environment . Some other government policies, of more recent vintage, have also provided assistance to garment export from Bangladesh. In 1999-2000 cash assistance was provided to garment exporters to the extent of 25 percent of f.o.b. value of export; it was reduced to 15 percent in 2002-03, and 5 percent subsequently. From 2004 garment manufacturers have been exempted from the payment of VAT on import of inputs and export of garments. These policies were however introduced when the garment industry was already well established and therefore had little or no role to play in the rise of the industry. Leakage of fabrics imported on a duty-free basis by some garment manufacturers into the domestic market during the 1980s had a significant effect on raising profitability of 39 the industry since at that time the import duty on fabric was as high as 150 percent . This was possible since the garment manufacturers ordered excess fabric by using false

21

estimates of fabric requirements. Import of 15 percent excess fabric was legally allowed to cover damage, wastage, etc. during the manufacturing process. However, garment manufacturers imported additional fabric more than this 15 percent legal limit by bribing concerned customs officials. Consequently, much more fabric than necessary was imported duty-free, and the excess cloth was than sold in the open market (such as Islampur in old Dhaka). This practice was however largely discontinued by the 1990s when the customs duty rate on imported fabric was substantially reduced, thereby reducing the premiums on selling cloth imported on a duty-free basis. This is a classic example of how poor implementation of policy, caused by corruption on the part of both the businessman and the bureaucrat, can help an industry’s cause even through it hurts the national cause (lower sale of locally produced fabric in this case). The discussion in this section clearly points to the positive contribution made by policy reforms to the growth of the RMG industry in Bangladesh. In particular, two policies– the SBW facility and the back-to-back L/C system- led to significant reduction in cost of producing garments and enhanced competitiveness of Bangladesh’s garments exports. It also allowed garment manufacturers to earn more profit which, when necessary, could be used to overcome difficulties arising from weak governance. Furthermore, poor governance, reflected in the leakage of duty-free imported fabrics in the domestic market, paradoxically enough also helped the garment manufacturers to earn extra ‘profit’ and thereby enabled them to absorb the ‘high cost of doing business’ – a fall-out of bad governance. VI. THE ROLE OF THE GARMENT ENTREPRENEUR In the literature on the garment industry of Bangladesh there seems to be a general indifference on the part of scholars regarding the role of the entrepreneurs in the rise of this industry in this country. Such apathy is perhaps attributable to a belief that the country was suffering from an acute shortage of entrepreneurs in the post-independence period due to the outflow of West Pakistani entrepreneurs who had dominated the East Pakistan (now Bangladesh) economy before independence. In the view of these scholars, the garments industry emerged in Bangladesh because of favourable world market conditions and the support provided by government policy. While these factors were undoubtedly very important for the rise of the garment industry in Bangladesh, a deeper analysis reveals that the Bangladeshi garment entrepreneurs also had a role to play in this process. The literature tells us that a successful entrepreneur is one who, by using his skills and by taking risks, takes advantage of economic opportunities which are thrown up by prevailing economic conditions. In the case of the Bangladeshi garments manufacturer, the MFA quota system created the opportunity for his economic advancement. But this was a time during the country’s history, only a few years after its independence, when the physical and social infrastructure was highly inadequate and weak, where the economic policy regime was highly regulated and public sector oriented, and where the bureaucracy was inexperienced and far from honest. Thus, the RMG entrepreneur was taking a big risk in investing in this new, little-known, export-oriented manufacturing activity. And, he had to display various types of entrepreneurial skills like getting hold of foreign collaborators,

22

arranging finance, managing labour, getting around bureaucratic obstacles, etc. Thus, we may justifiably consider the garments manufacturer in Bangladesh as a successful entrepreneur who made at least some contribution to the rise and growth of this industry. By the late 1970s, the process of global relocation of the apparel industry had proceeded far and was marked by the remarkable garments export success stories in the NICs as well as in China, Thailand, Malaysia, Indonesia, the Philippines, Sri Lanka, India and Pakistan. Inspired by this, several Bangladeshi entrepreneurs decided to enter the business of manufacturing and exporting apparel. There was little encouragement from the government in this venture. In fact, in 1978 one industrialist had his application for a new garment factory rejected on grounds that the twelve existing factories already represented 40 an excess capacity . It was a difficult environment for exporting garments from Bangladesh; the entrepreneurs were not familiar with the technology, it was difficult and time-consuming to clear imported machines from the port, banks were unwilling to issue L/Cs for the import of fabrics without the importers putting up considerable margins, the bureaucrats were inexperienced to deal with an export activity that depended on use of imported inputs, and trained workers and supervisors were scarce. However, the pioneering Bangladeshi entrepreneurs successfully managed to overcome these impediments with their ingenuity and imaginative thinking. A very important step towards the anchoring of the RMG industry in Bangladesh was taken when Desh Garments, which was set up in 1979 by a retired civil servant, Mr. Noorul Quader Khan, formed the first joint-venture in the garment industry with the South Korean conglomerate and major garment exporter the Daewoo Corporation in 1979. It was a technical collaboration between two firms from two countries. The Bangladeshi partner provided the necessary equity capital, and the Korean partner provided training for the floor workers, supervisors and factory managers, supplied machinery and handled marketing. The formal training in Korea and the learning-by-doing that took place at Desh’s facilities in Chittagong have been viewed as the most critical element in the successful 41 expansion of the garment export industry in Bangladesh . Mr. Khan was the most senior civil servant before he left the government job. He commanded high respect of the bureaucrats who were at the helm of power and decisionmaking in the late 1970s. It was his personality and past position that helped him get the necessary approval of the government for the joint-venture proposal with Daewoo during a time when politicians and bureaucrats were not in favour of having joint-venture with South Korea which was fully committed to a capitalistic philosophy. An ordinary businessman would not have been able to achieve the joint-venture with Daewoo and the garments industry may never have taken roots in Bangladesh. Entrepreneurs who followed Mr. Khan obtained approval of their proposals of joint-ventures relatively easily by referring to his case. It appears that the initiative to form a joint-venture with Daewoo was taken by Mr. Khan, and not the other way round. During an official visit to France, Mr. Khan happened to visit Daewoo’s Paris branch, and soon afterwards he expressed a desire to collaborate with Daewoo in a new garment venture in Bangladesh (Rhee, 1990). It is said that General

23

Ziaur Rahman, then President of Bangladesh, took personal interest and successfully 42 persuaded the Chairman of Daewoo to sign this joint-venture agreement . While it cannot be proved definitively, it appears that Desh and Daewoo had a large hand in introducing efficient administrative arrangements for assuring free-trade status for 43 100 percent export-oriented garment factories . The two companies had realized that speedy access to imported inputs at world market prices was critical to Desh’s garment production for export. As we have seen earlier, the government introduced the Special Bonded Warehouse system for export-oriented garment factories. Even through there is no documented evidence, it appears that Daewoo’s intimate knowledge of the successful bonded warehouse system in South Korea, its ability to transmit that knowledge to Desh management, and the advice that Desh’s senior manager gave to government officials on the new system were instrumental in the design and implementation of the SBW system. Similarly, access to import financing was felt to be necessary by Desh and Daewoo to gain equal footing with foreign competitors. It seems that influence of the two companies on the public agencies was instrumental in arranging for exporters to have access to suppliers’ credits for inputs needed for garment export production under the back-to-back usance import L/C system. In general, it appears that garment manufacturers could overcome many of the problems which plagued Bangladesh during the 1980s because of their official connections traceable to their social economic and educational background. A survey of garment entrepreneurs carried out in 1993 shows that 23 percent of respondents were retired civil or military bureaucrats, 89 percent had at least a bachelor degree and 32 percent had studied abroad (Quddus and Rashid, 2000). Thus, a large number of educated, mature and experienced individuals have been attracted to the garment industry, and according to the above survey, healthy profits in this industry was a major attracting factor. The presence of such dynamic persons as entrepreneurs doubtlessly stimulated the growth of this industry in ways delineated above. The problem of shortage of skilled workers and inadequate technical knowledge of the entrepreneurs was gradually overcome through the mobility of skilled workers and managers within the industry. Many of the successful firstgeneration garment manufacturers had started their careers as employees of Desh Garments. The industry as a whole benefited from this knowledge spill-over. True, the opportunity was created by the MFA Quota system, but it is to the credit of the garment manufacturers that they had the courage and vision to respond positively to this opportunity. Sobhan (1993) has written the following about the RMG entrepreneurs, “The growth originated from the market niche created for low-cost garment exports under the Multi-Fibre Arrangement (MFA). However, the response from the small/medium entrepreneurs has been impressive and demonstrates that private entrepreneurship has great promise in areas where markets are secure, investment finance needs are modest, technology assimilation is easy and a system of back-to-back financing of letters of credit provides a ready supply of working capital free of hassles of import regulation”. (p.24,25).

24

VII. SUSTAINABILITY OF THE RMG INDUSTRY From 1st January 2005 the global apparel market has become fiercely competitive after the complete phase-out of the MFA Quota system. The implication is that Bangladesh would have to enhance its competitiveness in garment manufacture to retain its share in the world market. It is important to note here that competitiveness in the global apparel market no longer depends only on technical and economic standards (like cost competitiveness, product quality, on-time delivery, etc.) but also on social standards, especially labour standards. The labour standards which are accepted internationally are those that are included in the concept of ‘decent work’ introduced by the ILO in 1999, and it focuses on four objectives: the promotion of rights at work, employment, social protection, and social dialogue. The emphasis is ensuring decent and productive work, in conditions of freedom, equity security and human dignity. The implication is that employers would have to ensure, inter alia, payment of fair wages and acceptable working conditions. In recent times, foreign buyers and retailers have been insisting on the fulfillment of these labour standards as a pre-condition for importing garments from Bangldesh and other countries. Whatever the motive behind this move, if Bangladesh’s garment exporters fail to meet these labour standards, exports are likely to suffer. At the same time, workers of Bangladesh’s garment factories have been demanding for quite some time the payment of fair wages and other non-wage benefits permissible under national labour laws, together with improvement in general working conditions. Several violent strikes by these workers, including setting fire to several garment factories, have been taking placing in the country since May 2006. Thus, it is clear that unless the issue of compliance with acceptable level of labour standards is properly addressed by garment enterprises, the long-run sustainability of this industry may be jeopardised. It has been noted above that the wage rate in the garments industry in Bangladesh is not only lower compared to wages in competing countries even in South Asia, it is even lower than wages paid in most other industries in Bangladesh. Not only that, the government’s minimum wage for the garment industry has remained unchanged at Taka 930 for grade seven employees and Taka 4,500 for grade one employees over the last 10 years. Evidence from many studies show that the garment industry is characterized by a wide variety of deprivation of workers, 44 particularly women workers . These include, inter alia, lack of proper infrastructure facilities and safety at workplace, non-compliance with legal minimum wages, wage discrimination against women workers, lack of provision of essential service benefits to workers, lack of housing facilities, irregularities with granting maternity leave and benefits, insufficient number of toilets, congested workplace in unsafe buildings, non-issuance of any formal contract of employment, non-compliance with overtime work rules and allowances, etc. It is important to note that very little improvement has taken place in wage and non-wage benefits paid to workers and working conditions in the garment factories since the industry started in the late 1970s, despite many attempts made by researchers and civil society to convince garment factory owners to bring about improvements in these conditions.

25

The BGMEA argues that the level of social compliance has improved in recent 45 years . It is argued that even though the statutory minimum wage has remained unchanged at Taka 930 per month, garment factories have gradually increased actual wages paid (for example Taka 1300 in 2004 for grade – 7 workers). Most factories pay the same wages for the same job and do not discriminate against female workers. Generally, decent hours of work are maintained (including overtime and night-shift work for women workers) and overtime payments are given regularly and on time as per the provisions of the relevant national labour laws. Leave and holidays are given properly. Employers have started to implement the factory laws to ensure a safe work environment. To ensure security of workers, factory owners are providing compensation packages according to law in case a worker is fired, and some factories also have provident fund facilities after old age retirement. And so on. On the other hand, a recent survey (Repon and Ahmed, 2006) confirms earlier research findings that compliance with labour standards is poor in the garment factories of Bangladesh. Thus, an overwhelming majority of empirical studies clearly indicate the low level of social compliance in garment factories. These empirical findings are vindicated by the recent occurrence of violence in the garment sector which sometimes appeared to threaten the very existence of the industry. So, if labour standards are not quickly complied with, both foreign buyers and garment workers would be very irritated, and this could jeopardise the long-run sustainability of production and export of garments from Bangladesh. It is important to note at this juncture that this problem with social compliance appears to have had its roots in the evolutionary pattern of the RMG industry in Bangladesh. In the early days of the garment industry, the entrepreneurs had relied mainly on the use of women workers primarily because they could be hired at lower wages compared to male workers. Furthermore, the workers were deprived of their due rights to ‘decent work’. There is enough empirical evidence which points in this direction. How could this happen when all the different dimensions of decent work, from payment of a fair wage to good working conditions, were incorporated in various national labour laws, and when the 10 core ILO labour standards were ratified by the government? A probable answer would be the weak enforcement of national labour laws which gave the garment manufacturers a free hand to flout these laws. Also perhaps, through bribing corrupt labour ministry officials, the garment producers succeeded in persuading these officials to keep their eyes closed to the violation of labour laws in the garment factories. It may be noted in this context that the government has recently formed a high-powered ‘social compliance Forum for RMG’ with the objective of ensuring social compliance in the RMG industry. If the government was sincere in protecting workers and had a long-term vision regarding sustainable growth of the garment industry, then such a forum would have been created a long time ago. This would have saved the industry from the problems it is now faced with relating to labour standards.

26

VIII. CONCLUDING REMARKS The garment industry struck roots in Bangladesh during the late 1970s and then witnessed remarkable growth to emerge as the major export item of the country with a share of about 75 percent in total export earnings. The industry began to grow under very unfavourable initial conditions characterized particularly by strong government regulation of the economy, domination by the public sector, and an inward-looking industrialization strategy. While these unfavourable conditions had virtually stifled industrial growth in the country, the garment industry flourished. How was this possible? The analysis in this paper has identified several factors which have stimulated growth of this industry. The MFA Quota system created the opportunity for Bangladesh to step into the global apparel manufacturing chain. Prior to 1985, Bangladesh was not under MFA quotas, and this attracted manufacturers from quota-restrained countries like South Korea to move into Bangladesh with the objective of collaborative production and export. Low wages in Bangladesh was another important factor which attracted foreign garment producers to this country. MFA quotas were imposed on Bangladesh in 1985. Prior to this, Bangladesh’s garment manufacturers were making handsome profits due to the low capital investment requirements of the industry, good rate of return, and low wages. High profitability continued even after the imposition of MFA quotas due to the fact that Bangladesh’s exporters of garments were now earning considerable quota rents. It is this high profits that enabled the garment manufacturers to get around the various government controls and regulations that were constricting the growth of other industries. The garment producers could afford to pay ‘speed money’ to corrupt bureaucrats to circumvent the various restrictions imposed by wrong policies. No wonder it has been said in the literature that the garment industry had ‘helped’ to institutionalize corruption in Bangladesh. Subsequently, policy reforms undertaken by the government created a more congenial environment for the expansion of the garment industry. In particular, relaxation of investment sanctioning procedures allowed more firms to enter the garment industry, implementation of the SBW facility brought Bangladesh’s garment manufacturers on a level playing field with its competitors by providing them access to imported inputs at world prices, and institution of the back-to-back L/C system helped the garment producers to largely overcome the problem of investment finance. Thus, there is no doubt that supportive public policy had contributed substantially to the growth of the garment industry in Bangladesh. The ingenuity and vision of RMG entrepreneurs was also a contributory factor in the remarkable growth of this industry. While the opportunity was created by the MFA Quota system, the dynamic pioneering entrepreneurs of this industry had the vision and the courage to take the risk of setting up this industry in unfavourable times. This perhaps was

27

nothing short of the “animal spirit” that Schumpeter had written about. Equally important was the contribution that the pioneering entrepreneurs made in influencing the government to design and implement supportive policies like the SBW facility and the back-to-back L/C system. This was largely possible because many of the entrepreneurs were former highranking civil and military bureaucrats and politicians. These entrepreneurs used their influence with government officials, who often were former colleagues and friends, to overcome obstacles created by restrictive government policies. While good governance in the form of supportive government policy had been helpful, paradoxically, bad governance had also contributed to the growth of the garment industry. Thus, for example, the corrupt government officials colluded with the garment producers in the sale of fabric imported on a duty-free basis in the open market, an activity which fetched large illegal returns for the garment industry. Again, corrupt labour ministry officials ignored, on the payment of graft, the violation of national labour laws by the garment manufactures which resulted in exploitation of labour but larger profits for the factory owners. Such misgovernance eventually gave rise to the serious problems of noncompliance with labour standards which, if not addressed satisfactorily, may threaten the long-run sustainability of the garment industry in Bangladesh.

28

NOTES 1. 2. See World Bank (1990). It appears useful in this regard to point to the following conclusion of Rhee (1990) : “The lessons to be learned from Bangladesh’s success with garment exports, as they apply to the search for a workable development strategy in low-income developing countries with very unfavourable initial conditions, are profound enough to warrant further in –depth research designed to generalize the preliminary findings”. (page 333). See, for example, Siddiqi (2004), Razzaque (2005) and World Bank (2005). Notable exceptions are Rhee (1990) and Quddus and Rashid (2000). This becomes clear from their following comment: “It would be hard to argue that the exports of garments were developed primarily as a response by domestic entrepreneurs to opportunities created by favourable policy changes....... Both value added in production and the net foreign exchange content of exports are very low for readymade garments. To cite their growth as an example of export dynamism induced by policy improvements is perhaps a bit misleading”. (page 95). Commenting on the general state of the private entrepreurs in Bangladesh, sobhan (1993) had written, “All this suggests whateverbe the long-term potential of private enterprises in Bangladesh..... there are no readymade panaceas in the promotion of the private sector while the entrepreneurial tradition has remained weak”. (p.29). Unless otherwise mentioned specifically, ‘RMG industry of Bangladesh’ in this paper shall refer to export-oriented garment enterprises only and shall not include the domestic market-oriented segment of the industry. As reported in newspapers, the owners of RMG factories had threatened to permanently close down their establishments if the labour unrest and the concomitant vandalism was not immediately stopped. For details see Rashid (1993). According to B.B.S. estimates, the average share of manufacturing value-added in GDP during 1973-90 was 9.3 percent. On the other hand, Bakht (1991) estimated this share to be about 12 percent. See Ahmed and Sattar (2003). The analysis in this section draws heavily on Bhuyan and Rashid (1993). See Sobhan and Ahmed (1980). For details see Bhuyan and Rashid (1993). See Trade and Industrial Policy (TIP) Reform Programme (1987). For details see Bhuyan and Rashid (1993). See Rahman and Bakht (1997). Also see Bhuyan and Rashid (1993). See Siddiqi (2004). See Quddus and Rashid (2000). The US and Canadian quotas continued till 31st December 2004, but the European quotas were withdrawn before they became effective. See World Bank (2005). The source is BGMEA, as cited in Quddus and Rashid (2000). See Rashid and Begum (2004). See World Bank (2005). See Rahman and Bakht (1997).

3. 4. 5.

6.

7.

8.

9. 10.

11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

29

26. Under the MFA, quotas grew at a specified rate each year, the quota growth rate being determined separately for each exporting country by the importing country. 27. For a theoretical exposition see, for example, Annex B in World Bank (2005). 28. The quota premium is the difference between the prices in the restricted market and the unrestricted market, i.e. (PR-PU) where R and U denote the restricted and unrestricted markets, respectively. The quota price is the amount that a producer not holding a quota is willing to pay in order to acquire the quota and sell in the restricted market; the quota price will be equal to PR-PU. The quota price may of course be less than the quota premium mainly due to extraction of a portion of the quota premium by large importers or retailers using their market power. See World Bank ibid. 29. We make the reasonable assumption that total production of garments was exported. 30. See Khundker (2002). 31. See Islam (2001). 32. For example, a 1997 BIDS survey shows that 74 percent of female workers earned no wages before joining the garment industry. 33. For a list of these studies and a more detailed descussion on these issues see PaulMajumder (2002). 34. There are reports that some garment factories, especially the small ones engaged in subcontracting work, are still employing child labour. 35. See Paul- Majumder (2002) and Repon and Ahmed (2006). 36. This is the so-called ‘catalyst model of development’. See Rhee (1990). 37. The XPL scheme was subsequently replaced by the Export Performance Benefit (XPB) scheme. 38. See Siddiqi (2004). 39. Thus, what cost the apparel manufacturer $ 100 to import was actually worth $ 250 in the market. 40. See Quddus and Rashid (2000). 41. See Rhee (1990). 42. See Quddus and Rashid (2000). 43. See Rhee (1990). 44. See Paul- Majumder (2002) for a brief review of some of these studies. Also see CPD (2003). 45. See Jabbar (2006).

30

REFERENCE Ahmed, Sadiq and Zaidi Sattar, 2003, Trade Liberalization, Growth and Poverty Reduction : The Case of Bangladesh, The World Bank, Report No. IDP-190, Washington D.C. Bakht, Zahid, 1991, “Time Series Estimate of Manufacturing Value Added at 4-digit Industry Level : 1980/81 – 1989/90”, Bangladesh Institute of Development Studies, Dhaka. Mimeographed. Bhuyan, Ayubur R. and Mohammed A. Rashid, 1993, Trade Regimes and Industrial Growth : A Case Study of Bangladesh, Dhaka : International Center for Economic Growth (ICEG) and Bureau of Economic Research (BER), Dhaka University. Centre for Policy Dialogue (CPD), 2003, Phasing Out of the Apparel Quota : Addressing the Livelihood Concerns in Bangladesh. Dhaka : The University Press Limited. Islam, S., 2001, The Textile and Clothing Industry of Bangladesh in a Changing World Economy. The University Press Limited. Jabbar, S.K. Jenefar, 2006, “Ready Made Garment Sector in Bangladesh : A Study from the Employers’ Perspective”, in ILO (2006). Khandker, Bazlul Haque and Abdur Razzaque, 2006, “Exports, Employment and Working Conditions : Emerging Issues in the Post-MFA RMG Industry”, in ILO (2006). Khundker, Nasreen, 2002, “Globalisation, Competitiveness, and Job Quality in the Garment Industry in Bangladesh”, in M. Muqtada, A.M. Singh and M.A. Rashid (ed.), Bangladesh : Economic and Social Challenges of Globalisation. Dhaka : The University Press Limited. Khan, A.R. and Mahabub Hossain, 1989, The Strategy of Development in Bangladesh. Macmillan. Paul-Majumder, Pratima, 2002, “Organising Women Garment Workers : A Means to Address the Challenges of Integration of the Bangladesh Garment Industry in the Global Market”, in M.Muqtada, A.M. Singh and M.A. Rashid (ed.), Bangladesh : Economic and Social Challenges of Globalisation. Dhaka : The University Press Limited. Quddus, Munir and Salim Rashid, 2000, Entrepreneurs and Economic Development : The Remarkable Story of Garment Export from Bangladesh. Dhaka: The University Press Limited. Rashid, Mohammed A. and Salma Begum, 2004, “RMG Value Chain Analysis”. Report prepared for GTZ.

31

Rashid, Mohammed A., 1993 “Bangladesh : Output and Employment Effects of Recent Macroeconomic Policy Reforms”, in ILO-ARTEP, Social Dimensions of Economic Reforms in Bangladesh. New Delhi. Rahman, Mashiur and Zaid Bakht (1997), “Constraints to Industrial Development : Recent Reforms and Future Directions”, in M.G. Quibria (ed.), The Bangladesh Economy in Transition. Delhi : Oxford University Press. Razzaque, Abdur, 2005, Sustaining RMG Growth after MFA phase –out : An Analysis of Relevant Issues with Reference to Trade and Human Development. UNDP, Dhaka. Rhee, Yung W., 1990, “The Catalyst Model of Development : Lesson from Bangladesh’s Success with Garment Export”, World Development, vol. 18, No. 2. Siddiqi, Hafiz G.A., 2004, The Readymade Garment Industry of Bangladesh. Dhaka : The University Press Limited. Repon, A.R. Chowdhury and Syeed Ahamed, 2006, “Social and Economic Costs of PostMFA Phase Out : Workers’ Perspectives”, in ILO, RMG Industry Post-MFA Regime and Decent Work : The Bangladesh Perspective. Dhaka, International Labour Organization. Sobhan, Rehman and Muzaffer Ahmed, 1980, Public Enterprise in an Intermediate Regime : A Study in the Political Economy of Bangladesh. Dhaka : BIDS. Sobhan, Rehman, 1993, “From Theology to Empiricism : Reappraising Bangladesh’s Experience with Economic Reforms”, M.N. Huda Memorial Lecture, Bangladesh Economic Association, Dhaka, May 8. Trade and Industrial Policy (TIP) Reform Programme (Planning Commission), 1987, Overview of Industrial Assistance Policy in Bangladesh and Recommendations for Reform. Dhaka. World Bank, 1990, Garments : Global Subsector Study. World Bank, 2005, End of MFA Quotas : Key Issues and Strategic Options for Bangladesh Readymade Garment Industry. Bangladesh Development Series – Paper no. 2, The World Bank Office, Dhaka, December.

32