Bangladesh: Development Outcomes and Challenges in the Context of Globalization

by

Wahiduddin Mahmud University of Dhaka

Paper presented at the conference on The Future of Globalization: Explorations in Light of Recent Turbulence co-sponsored by the Yale Center for the Study of Globalization and the World Bank, October 10-11, 2003, Yale University.

1. Introduction Bangladesh faces the challenge of achieving accelerated economic growth and alleviating the massive poverty that afflicts nearly two-fifths of its 135 million population. Strategies for meeting this challenge have included a shift away from state-bureaucratic controls and industrial autarky towards economic liberalization and integration with the global economy. These policy reforms were initiated in the mid-1980s against the backdrop of serious macroeconomic imbalances, caused in part by the declining level of foreign aid and in part by a preceding episode of severe deterioration in the countrys terms of trade. The policy reforms in the 1980s included the withdrawal of food and agricultural subsidies, privatization of state-owned enterprises, financial liberalization, and withdrawal of quantitative import restrictions. The beginning of the 1990s saw the launching of a more comprehensive reform program, which coincided with a transition to parliamentary democracy from a semiautocratic rule. These later reforms were particularly aimed at moving towards an open economy such as making the currency convertible on the current account, reducing import duties generally to much lower levels, and removing virtually all controls on the movements of foreign private capital. Besides, fiscal reforms were undertaken including the introduction of the value-added tax. During the 1990s, notable progress was made in economic performance. Along with maintaining economic stabilization with a significantly reduced and declining dependence on foreign aid, the economy appeared to begin a transition from stabilization to growth. In the 1980s, per capita GDP had grown slowly at the rate of about 1.6 per cent per annum; the growth rate accelerated to 2.4 per cent in the first half of the 1990s, and further to 3.6 per cent in the second half of the decade. The acceleration in the growth of per capita income owed itself both to a slowdown in population growth and a sustained increase in the rate of GDP growth (which averaged 5.21 percent annually during the second half of the 1990s). During this time, the progress in the human development indicators has been even much more impressive. Bangladesh is in fact among the top performing countries in the 1990s in terms of the extent of improvement in the Human Development Index as estimated by the UNDP.1 However, there are signs that continued progress in this respect may prove increasingly difficult, particularly in the absence of strong international support and further consolidation of the economy's growth performance. The relatively strong growth of the Bangladesh economy in the 1990s was underpinned by even much stronger export growth. That performance has now been put to test by the global economic recession and the threat of post-MFA competition in the export of readymade garment - the country's flagship in the global market. Given the increased reliance on trade, the country's overall economic performance has come to depend to a large extent on how well it can cope with the risks and opportunities in the global market. Undoubtedly, the country will have to deal with a whole range of internal policy issues, from public finance and the financial system to governance and investment climate. But, much
The absolute increse in the value of HDI for Bangladesh between 1990 and 2001 is surpassed only by that for China (and Cape Verde) among all the countries for which such estimates are available; cf. UNDP (2003), see the table on "Human development Index trends", pp. 241-44.
1

will also depend on the changes in the global economic scenario and the way domestic policies respond to such changes. This paper is aimed at particularly highlighting this later aspect in the wider context of Bangladesh's development options and challenges. 2. Trends in Macroeconomic Indicators The stabilization program was primarily aimed at reducing the fiscal and external deficits to a sustainable level, consistent with the reduced and declining level of aid availability. By the end of the 1980s, the external current account deficit had been already reduced to about 5 percent of GDP from between 8 to 10 percent in the beginning of the decade, and there was a similar decline in the overall budgetary deficit. However, this was achieved by cutting back on investment, both public and private, rather than by mobilizing larger domestic savings. The ratio of investment to GDP steadily declined, and the government's development budget became almost entirely dependent on foreign financing. Throughout the 1980s, the contribution of the governments fiscal operations to domestic savings, in the form of public savings, continuously declined because of the rapid growth in current expenditures along with a stagnant and low revenue-GDP ratio. As a result, macroeconomic strains started to reappear towards the end of the decade (Mahmud 1995). Against this backdrop, the macroeconomic indicators showed a marked improvement in the 1990s. Although the net flow of foreign capital further declined to less than 2 percent of GDP, indicating a further decline in foreign aid, the investment GDP ratio steadily increased from about 17 percent in 1990/91 to 23 percent in 2000/01. Furthermore, this increase was almost entirely due to the dynamism in private investment, which increased from about 10 percent of GDP to about 16 percent in the above period, while the investment rate in the public sector remained unchanged at around 7 percent of GDP. The increase in the investment rate was backed by an even more marked improvement in the domestic (and national) saving rate. A significant increase in the tax/GDP ratio in the early 1990s, following the introduction of the value added tax, helped to increase public savings and largely reduce the dependence of the government's development spending on foreign aid. Although the later was also partly achieved by an increase in domestic borrowing, the overall budget deficit was mostly within the limit of fiscal sustainability. The average annual inflation rate, as measured by the official consumer price index, came down to about 6 percent in the 1990s from about 10 percent in the earlier decade - a further evidence of successful stabilization. Clearly, the transition to democracy was accompanied by improved macroeconomic management. However, there was some evidence of periodic lapses in the fiscal discipline related to the timing of the approaching national elections, thus producing the symptoms of the so-called "political business cycle".2 In spite of the overall soundness of the above macroeconomic trends, there are some disconcerting features. While the increase in the investment rate was entirely led by private investment, there were at times symptoms of a feeble investment response to policy reforms. In the early 1990s, for example, the marked increase in the saving rate was not matched by
2

For evidence, see Mahmud (2002), Chapter 19.

similarly strong response from private investment, thus leading to a situation of aggregate demand deficiency. This was evident from a sharp fall in the inflation rate to a near-zero level accompanied by a fall in private sector credit expansion and an unprecedented build-up of international reserves.3 It is difficult to ascertain whether the rapid reduction in industrial protection through import liberalization at that time had a role in this. The uncertainty created by these reforms, which had no pre-announced targets or time-table, could have been a contributing factor as well. There were also other periods of stagnation or decline in investment in the manufacturing sectors.4 Thus, apart from the resource constraint to investment growth, the "desire to invest" factors may have become important in the post-reform era, particularly because of the withdrawal of public investment from the directly productive sectors. This problem has perhaps much less to do with the structure of industrial incentives than with the overall investment climate, including what are known as the "costs of doing business". The factors adversely affecting investment incentives are well documented in Bangladesh's context: poor infrastructure, a weak financial system, corrupt and inefficient bureaucracy, collection of illegal protection money and an inadequate legal system, to name a few. These factors also inhibit the flow of foreign private investment, the volume of which remains insignificant despite liberal policies.5 The experience of macroeconomic management in the 1990s also shows the fragility of the country's balance of payments in spite of very impressive export growth. Whenever there was a marked dynamism in investment and industrial activity levels, it turned out to be short-lived as it came up against the balance of payments constraint leading to a depletion of foreign exchange reserves. This was true of the two episodes of mini-boom that took place around 1994-95 and 2000-01 respectively. In both cases, there was an upturn in large-scale manufacturing production associated with a rapid expansion of private sector credit and high growth in the import of capital goods and industrial inputs (Mahmud 2003a). In both cases, the external deficits increased and the foreign reserves sharply depleted, until stabilization measures led to the subsequent stifling of the growth of investment and manufacturing activities. The above phenomenon is reminiscent of a dominant trade gap in the erstwhile popular two-gap model of foreign aid.6 It reveals the highly import-intensive nature of investment and manufacturing activities in Bangladesh. Most capital goods are imported and
See Mahmud (2001), Chapter 4. This can be seen from the growth of machinery import, which is a good proxy for industrial investment; see Mahmud (2003a). 5 Foreign direct investment peaked at about US dollar 250 million in 1997-98 accounting for about 0.5 percent of GDP in that year; in 2001-02, it fell sharply to only US dollar 65 million.
3 4

The problem seems to be that of the external balance rather than of the aggregate resource balance, that is, the savings-investment balance. This was more evident in the later episode of economic upturn in 2000-01, in which year the domestic inflation rate in fact declined to less than 2 percent, but the foreign exchange reserves came down to a critically low level - equivalent of less than two months' import payment; see Mahmud (2003a).
6

domestic manufacturing is highly dependent on imported raw materials and intermediate goods, while finished consumer goods account for less than 10 percent of all imports. There is thus little room for adjusting to a lower import growth without subduing investment demand or creating capacity underutilization in the economy. Until hit by the global recession in 2001, there had been robust and sustained growth of export earnings, averaging about 15 percent per year in nominal US dollar terms in the 1990s. As a result, the ratio of export earnings to GDP doubled from 7 percent to nearly 14 percent. In 2001-02, export earnings declined in dollar terms for the first time since 1985-86. Although there was a recovery in the following year, the medium term outlook is that it will be difficult to regain the export momentum of the 1990s. A redeeming factor is the continued increase in the inflow of migrant workers' remittances, which grew from about 2.5 percent of GDP in the beginning of the 1990s to above 5 percent in 2001-02 (amounting to about 2.5 billion US dollars). 3. Sources of Growth Stimulus All three broad economic sectors namely, agriculture, industry and services contributed to the growth acceleration of the 1990s. The growth of agricultural GDP accelerated from 2.5 per cent in the 1980s to 3.2 per cent in the 1990s; industrial GDP accelerated from 5.8 to 7.0 per cent, and the service sector GDP from 3.7 to 4.5 per cent. 7 The growth in large-scale manufacturing has been led almost singularly by the ready-made garment industry. The national income estimates of manufacturing value-added by sector show that, during the 1990s, medium and large-scale manufacturing grew at about 7 percent annually; but at only about 4 percent excluding the garment industry.8 This implies that growth in medium and large-scale manufacturing has been mainly export-led. But it also means that the economy has moved away from, rather than moving towards having a more diversified manufacturing and export base. Fisheries, within agriculture, have been another high performing sector, with the annual growth rate increasing from about 2.5 percent in the 1980s to over 8 percent in the 1990s. It is no coincidence that, next to garments, frozen shrimp has been the only other fast growing major export item in the 1990s (Table 1). 9 It would thus appear that the strong export growth over the last decade or so did succeed in stimulating some parts of the economy, namely, manufacturing (through readymade garment export) and fisheries (through shrimp export).

The industrial sector includes construction, mining and utilities, besides manufacturing. Cf. Statistical Yearbook of Bangladesh 2000, p.452. 9 By the mid-1990s, Bangladesh accounted for more than 4 percent of commercial shrimp, although productivity per acre is very low by international standards.
7 8

Table 1 Growth of Exports in the 1990s (3-year average of annual export earnings in million US dollars)

Export item Readymade garments & knitwear Frozen foods (mainly frozen shrimp & fish) All other exports Total exports

1989/90 1991/92 858 137 751 1745

1998/99 200/01 4411 327 750 5488

Export-led growth is, however, only a part of the growth dynamics. In spite of the changing structure of the economy, only slightly over 10 percent of GDP currently originates from large-scale manufacturing (and about 6 percent from fisheries). As regards other sources of growth, it should be noted that the growth contribution of crop agriculture remains important, given its still significant share in total GDP - about 15 percent. Even a much larger share of GDP originates from the so-called informal sectors outside crop agriculture, comprising of small-scale processing and manufacturing activities and various informal services.10 The breakdown of GDP estimates shows that the growth rates of these activities have accelerated in the 1990s, thus contributing substantially to the acceleration of the overall GDP growth. Much of these informal sector activities, being extremely laborintensive and requiring very little capital investments, are largely demand-driven. Where has the demand stimulus come from? One can identify at least three major sources: increase in income from crop production, readymade garment export and workers' remittances, in that order of importance.11 Although the above informal sectors are largely driven through demand linkages with the leading productive sectors of the economy, they have their internal growth dynamics as well. There is evidence that the growth acceleration in these sectors in the 1990s was accompanied by a tilt towards relatively up-scaled activities which have higher labor productivity and can cater to more income-elastic demand (Mahmud 1996, Mahmud 2003; Osmani et al. 2003). Import liberalization is likely to have had a role in infusing this dynamism by providing improved access to imported inputs and technology. Within manufacturing, for example, small-scale manufacturing activities (excluding handlooms and cottage industries) have fared better than large-scale manufacturing in the post-liberalization period, growing at an average rate of more than 9 percent annually in the 1990s. 12 Small industries seem to have benefited from the liberalization of import of capital machinery and
According to the official 1999/00 Labor Force Survey, these informal sectors currently employ about three-fourths of the country's non-agricultural labor force. 11 For an estimation of the relative importance of these growth stimuli, see Osmani et al. (2003).
10 12

Cf. Statistical Yearbook of Bangladesh 2000, p. 452.

raw materials, while their products, being mostly remote substitutes of imported items, had an advantage over those of their large-scale counterpart facing stiffer competition from imports.13 The growth in crop agriculture, averaging at the rate of about 2 percent annually over the last two decades, has been so far almost entirely due to increased rice (and wheat) production. Profitability analyses show that, beyond attaining self-sufficiency in rice, there is a potential for accelerating agricultural growth through crop diversification. Interestingly, the country does not appear to have comparative advantage in products such as edible oils and sugar that currently enjoy high protection and there is scope for import substitution (Mahmud et al. 2000). On the other hand, the high-value crops such as fruits and vegetables can be profitably produced both for meeting domestic demand and for export; but a shift from rice to such crops will require technological dissemination, better integration with processing and marketing and provision of other support services. The R and D activities in agriculture will also need to be strengthened and will need more funding and donor assistance than is currently available. For these products to enter the export market, it will require even far more improved marketing infrastructure. 4. External Sector Policies Impact of Import Liberalization Starting from the withdrawal of quota restrictions in the late 1980s to tariff reductions in the first half of the 1990s, Bangladesh has had one of the most rapid episode of import liberalization compared to other countries during their similar period of trade reforms. Yet, at the end of this period, Bangladesh still remained relatively "closed" in terms of the extent or depth of liberalization (though ahead of many countries including India, Pakistan, Vietnam, Thailand, and China). The un-weighted average rate of protective import duty declined from 71 percent in 1991-92 to 26 percent in 1996-97, while that of all import taxes declined from 83 percent to 39 percent (Ahmed and Sattar 2003). Further liberalization since then has been rather slow. The extent and the speed of further import liberalization remain a contentious issue in the country's economic reform agenda. One important concern is the possible adverse effect of tariff reductions on government revenue. Although the introduction of the value added tax has reduced to some extent the dependence on import duties, more than half of total tax revenue still comes from such duties. So far, the reductions in the rates of protective duties (that is, customs duties) have been more or less compensated by the growth of import. In fact, revenue from import duties as a proportion of GDP slightly increased during the period of rapid import liberalization in the first half of the 1990s (Mahmud 1997). If there is now a slowdown in trade expansion because of global economic factors, the revenue concerns will become stronger in further trade liberalization.
Small industries are likely to have grown also at the cost of cottage industries, the value added from the latter growing at only 2.8 percent annually during the same period.
13

More generally, tariff reforms have to cope with many structural limitations of the existing tax system. For example, the taxation of imported intermediate and capital goods is an important source of revenue and is often the only means of taxing a large part of domestic production. There is thus a need for removing discrimination against exports, often by selective intervention. The existing schemes of duty-drawback, bonded warehouses and selective cash incentives for exports are essentially means for providing exporters access to duty-free imports of production inputs (or for compensating for the payment of such duties). The selectivity of such intervention depends not only on economic criteria, but also, of necessity, on their administrative feasibility and effectiveness. Also, the existing import tariff escalation creates room for making the incentive system discriminatory and generally biased in favor of producing finished consumer goods for the domestic market. Clearly, an active industrial policy has a key role to play, more so in the transitional reform period, as the impact of reforms is bound to vary considerably across industries and sectors. If there is no well-devised industrial policy, there will be one by default. This calls for adopting an analytical approach to trade policy reforms, the capability of which is largely lacking within the government. While the impact of external liberalization in Bangladesh has been an acceleration of overall industrial growth, that is almost entirely because of the success in garment export, and in spite of substantial contraction in many import-substituting industries. The official index of large and medium-scale manufacturing industries shows that almost half of 4-digit industries registered negative growth during the first half of the 1990s - the period of rapid tariff reductions. While there were some gains in the overall production efficiency of importsubstituting industries, that was more because of the exit of the relatively inefficient (and mostly state-owned) industrial units than any firm-level technological improvements (World Bank 1999). One exception is the pharmaceutical industry, which grew rapidly as an importsubstituting industry by trebling its output in the 1990s and is now well positioned to even enter the export market. (Incidentally, Bangladesh stands to gain from the WTO agreement regarding the waiver of patent rights for the domestic production of drugs in the LDCs until 2016 and for the production and trade in certain life-saving drugs among developing countries generally.) As noted earlier, small industries as well as many nontradable sectors also gained because of better access to imported inputs. Export Outlook The garment export from Bangladesh grew rapidly by taking advantage of export quotas and preferential access in the major markets (the United States and the European Union) along with the abundance of low-cost female labor. The impact of the prospective removal of MFA quotas in the US market could be significantly negative for Bangladesh, although the impact could be cushioned to some extent by the continued preferential access to the EU market. It is difficult to guess how far Bangladesh's garment industry will be able to cope with competition from other exporters.14 The dependence on imported fabrics and yarns puts Bangladeshi garment exporters at a disadvantage, by increasing the lead-time to meet orders
In recent years, there has been some growth in backward-linkage domestic production of fabrics, but only under high cash incentives (subsidies) that are now being phased out. Keeping stocks of gray cloth and establishing domestic capacity for only dying and finishing may be another way out.
14

form foreign buyers. Although the country has the advantage of the lowest wage rate (followed closely by China), it looses out in the marketing value chain (because of dependence on marketing intermediaries). Relatively larger firms with better marketing ability has a better chance of surviving the competition and even gaining in the quota-free export market (Nadvi 2003). A possible slowing down of garment-led industrial and export growth poses a threat to Bangladesh's prospects for achieving accelerated economic growth. Given the high importintensity of manufacturing and investment activities, adjusting to a lower import growth will not be easy. Also, in the event of a slowdown in the growth of the garment industry, the desirability of further import liberalization (that may hurt overall industrial growth) may be put to question, thus hindering the progress towards establishing a more efficient, competitive industrial base. The policy imperatives are thus clear, namely, (a) to make every effort to compete effectively in the post-MFA garment export, and (b) to move towards export diversification by at least preparing the ground for it in the next few years. Bangladesh has potential comparative advantage in a number of export items like horticultural products (fresh and processed), leather goods, light engineering products and certain chemical products. The country should also be able to take advantage of duty-free access of LDC exports to be provided by the EU, Canada, Australia and other industrialized countries. To take advantage of these opportunities, attention has to be given to a number of export facilitating factors, such as better infrastructure including efficient port facilities, standardization of product quality, technological improvements leading to higher productivity, and an improvement in the overall domestic investment climate. Foreign direct investment can also be an important factor not only by bringing in investment funds, but also by promoting export through product quality assurance (e.g. through brand names). Trends in the Real Exchange Rate A remarkable aspect of Bangladesh's reform experience is in respect of the movements in the real exchange rate (Mahmud 2001, Ahmed and Sattar 2003). In spite of substantial trade liberalization taking place at a time of a marked and steady decline in external deficits, there was hardly any real devaluation of taka (with only mild devaluation up to about 1997, followed by modest appreciation since then). The strength of taka was due to the rapid growth in export earnings along with the increasing flows of workers' remittances, which together more than offset the decline in aid inflows relative to GDP.15 During this time, many developing countries, including India and Pakistan, had massive real devaluation of their currencies. This has put Bangladesh at a disadvantage in export competition and in promoting an export market within these countries, especially neighboring India.16

15

Thus, the effect of tariff reduction was not generally compensated by indirect protection through real devaluation. Moreover, there was a steady increase in real wages in the organized part of the domestic import-substituting industries, further eroding their advantageous position. According to the IMF estimates, Pakistan, India and China have all maintained a lower REER compared to Bangladesh with 1990 as the base year; only Sri Lanka and Vietnam had higher REER.
16

The real value of the Bangladesh taka as against Indian rupee is estimated to have appreciated by as much as 25 percent or more between the late 1980s and the mid-1990s. This, combined with the fact that Bangladesh has gone for a much faster import liberalization compared to India, resulted in a dramatic increase in Bangladeshs trade deficit with India. By the late 1990s, Bangladeshs exports to India could pay for a meager 4 percent of its imports from that country compared to more than 15 percent a decade or so earlier. However, from the poverty perspective, there has been one benefit arising from these movements in bilateral real exchange rate. With the liberalization of rice import, commercial imports of rice from India can now mitigate the potential adverse effects of food price hikes at times of poor rice harvests in Bangladesh (Dorosh and Shahabuddin 1999). While the past experience with the success in garment export may have raised the specter of a Dutch disease, the prospect of any large real devaluation caused by an export debacle is of a much more serious concern. Most parts of the Bangladesh economy in effect represent semi- or non-tradables; these include not only services accounting for nearly a half of GDP, but also small-scale manufacturing producing poor substitutes for imports and many agricultural products including rice in which there is no commercial foreign trade in most years.17 Devaluation will adversely affect production incentives in these sectors by increasing the price of imported inputs and turning the domestic terms of trade against these sectors vis-vis the tradable sectors - not an encouraging prospect for achieving accelerated GDP growth, let alone pro-poor growth. This underlies the importance of international support to tide over the adverse balance of payment effect of an export slowdown. It also highlights the importance of promoting export growth and diversification through various non-price incentives and measures as mentioned above, rather than relying on an aggressive exchange rate policy as is often advocated. An added implication is that, in an economy that has yet to complete a transition from non-tradable-based to tradable-based growth, the use of exchange rate as a tool for export promotion needs to be treated with caution.18 5. Poverty Reduction and Social Development Trends in Poverty With the acceleration in the growth of per capita income, there has been considerable progress in poverty reduction. The head-count national incidence of poverty was reduced from about 50 percent to 40 percent during the decade of the 1990s, compared to a relatively much slower reduction in the previous decade (Table 1). The progress in the 1990s would have been more, had there not been a worsening of income distribution both in rural and urban areas. Even if poverty reduction continues in its present pace, it will fall short of the national target set in line with the Millenium Development Goal (Government of Bangladesh 2003). This underlies the importance of strengthening GDP growth and making such growth more pro-poor.

The nontradable nature of these agricultural products arises from the large spread between export and import parity prices caused by poor marketing infrastructure and limited access to the export markets (Mahmud et al. 2000). 18 On this, see, for example, Harberger (2001), p. 559.
17

10

The analysis of past trends in income growth and inequality shows that the relatively rapidly growing parts of the economy happen to be the ones with more unequal income, such as the urban/organized sector vis-a-vis rural/informal sector or the dynamic part of the rural non-farm sector vis--vis agriculture (Mahmud 2003, Osmani et al. 2003). For example, the rural economy showed some dynamism in the 1990s due to income growth in the relatively up-scaled non-farm activities that yielded higher labor productivity and could better respond to income-elastic demand. However, non-farm income was not only more unequally distributed compared to rural income as a whole, but also it became increasingly so throughout the 1990s; as a result, income distribution worsened.19 Bangladesh thus seems to be precariously positioned in the growth-inequality link as in the initial stage of the "Kuznets process".
Table 2 Trends in Poverty and Income Distribution in Rural and Urban Areas, 1983-84 to 2000 Year Percent of population under poverty line Rural 53.8 52.9 43.6 Urban National 40.9 52.3 33.6 26.4 49.7 39.8 Gini index (%) of consumption expenditure Rural Urban 24.6 29.8 25.5 29.7 31.9 37.9 Mean consumption expenditure as % of poverty line Rural Urban 106 131 109 129 153 197 Urban-rural ratio of mean expenditure 1.39 1.60 1.74

1983/84 1991/92 2000

Source: Estimated from data reported in Osmani et al. (2003) and Government of Bangladesh (2003).

The remedy seems to lie in the simultaneous growth of both low-productivity selfemployment, such as through provision of microcredit, and wage employment in relatively more productive jobs through the gradual scaling up of enterprises. A higher agricultural growth can also have an equalizing effect, while stronger growth of the rural economy as a whole can favorably affect the overall income distribution in the economy by reversing the observed increasing trends in the rural-urban per capita income gap (Table 1). Furthermore, with high, employment-oriented GDP growth, there is expected to be a tightening of the labor markets in the informal sectors including in agriculture, thus leading to an increase in the wage rates for the poorest sections of the labor force. So far, however, this trickle-down effect has not been realized and the wage rates in real terms in the informal labor markets have remained stubbornly stagnant (Mahmud 2003). A greater integration with the global economy seems to fit well with the above kind of poverty-alleviating growth. The export-oriented garment industry presently employs
In contrast, the growth of employment in the rural non-farm sector in the previous decade seems to have been in the extremely low-productivity activities, resulting in an overcrowding in these activities and overall low and declining average labor productivity in that sector. Although there was very little worsening of rural income distribution, there was little improvement in poverty incidence because of stagnation in income (as can be seen from the trends in the mean level of consumption as a proportion of poverty line consumption expenditure in Table 1).
19

11

around 1.8 million workers who are mostly female and are drawn from low-income, often rural, background. The industry is also highly labor-intensive with a significantly higher share of wage in value-added compared to the average of the organized manufacturing sector.20 The rapid employment growth in this industry has been a contributing factor to the observed trends towards feminization of the labor force, which is a welcome development so far as the gender dimensions of poverty are concerned. Shrimp farming, the second dominant export-oriented activity, is also very labor intensive, presently employing nearly half a million rural poor. More generally, external liberalization is likely to have contributed to the creation of productive employment for the poor through the strengthening of many smallscale and informal sector activities as discussed earlier. The other important external factor having a favorable impact on poverty is the rapid growth of workers' remittances. Most migrant workers come from relatively poor rural families and the remittances have been a means of upward mobility for these families. Besides, there is evidence that the spending out of remitted funds has helped the growth of the rural economy, particularly in localities with a concentration of families having migrant workers abroad (Mahmud 1989). Human Development Compared to progress in reducing income-poverty, there has been more significant progress in human development indicators, and this progress has been particularly marked in the 1990s (BIDS 2001, Government of Bangladesh 2003). Bangladesh is in fact among the few developing countries which are on target towards achieving the Millenium Development Goals in respect of most of the social development indicators (Table 3). The decline in infant and child mortality rates, for example, is among the fastest in the developing world. Bangladesh has already eliminated gender disparity in primary and secondary school enrolment and has made remarkable progress in providing universal basic education. The success in reducing the population growth rate through the adoption of birth control methods is also quite unique for countries at similar per capita income levels. Bangladesh belongs to a regional belt stretching across northern Africa, the Middle East, Pakistan and northern India which is particularly characterized by patriarchal family structures along with female seclusion and deprivation. That makes Bangladesh's achievements all the more noteworthy. There are, however, some doubts about whether these trends in the human development indicators can be sustained. The progress achieved so far is attributed to favorable budgetary allocations with donor support and a strong presence of nongovernmental organizations that characterizes the development scene in Bangladesh. These improvements in part represent "catching up", since two decades or so ago, Bangladesh was in fact a laggard in these indicators among countries with similar per capita income level (World Bank 2003). While the situation has now reversed, maintaining this rate of progress without a more rapid reduction in income-poverty may become increasingly difficult. 21 There
This can be seen from the results of the official Census of Manufacturing Industries. For most social development indicators, the actual values of the indicators achieved by the country are found to be far superior to the "predicted" values at the given level of per capita income; see Government of Bangladesh (2003), Table 1, p. 5.
20 21

12

is ample evidence that large disparities exist in health and educational achievements among households of different income groups and that poverty and deficiencies in human development perpetuate each other.
Table 3 Improvements in Some Human Development Indicators during the 1990s (Bangladesh compared to South Asia)

Net primary enrolment ratio (%) Ratio of girls to boys in primary and secondary education (%) Under-5 mortality rate (per 1,000 live births) Population with access to improved sanitation (%) GDP per capita (2001 PPP US$)

Bangladesh South Asia Bangladesh South Asia Bangladesh South Asia Bangladesh South Asia Bangladesh South Asia

1990* 64 73 72 69 144 126 41 22 1160 1957

2001* 89 79 103 78 77 96 48 36 1610 2730

* Or closest year with available data. Source: UNDP (2003) and Government of Bangladesh (2003).

It is also noteworthy that some of the spectacular achievements in health indicators could be achieved with the availability of low-cost technology (viz., oral rehydration technology for diarrhea treatment leading to a decrease in child mortality) and by creating more awareness (viz., immunization, contraceptive use). But, lowering maternal mortality rate, for example, requires provision of relatively costlier health services, so that the level of public health expenditure and the quality of health services will become an important factor beyond merely creating awareness. Similarly, remarkable progress has been made in school enrollment, especially for girls, but there are serious concerns now about the quality of education. Clearly, the challenge in these areas increasingly lies in mobilizing more resources and improving the quality of service delivery. The trends in the government's budget allocations show that the shares of both health and education sectors out of total budget expenditure have increased considerably as a result of fiscal adjustment. The increase has been particularly rapid for education, whose share has doubled from 8 percent in the first half of the 1980s to 16 per cent in the second half of the 1990s, while the share of health and family planning has increased from 5 percent to 7 percent during the same period.22 However, the absolute per capita expenditure in these sectors (say, in dollar terms) is far too low even by the standards of developing countries. Also, the structural shift in the budget towards larger social spending has come about from a
22

The figures are 5-year averages based on "revised budget" estimates.

13

redefining of the role of the government and is, therefore, of a once-and-for-all nature (Mahmud 2002a). In future, higher allocations to social sectors will require more difficult reforms for public resource mobilization. 6. Role of Foreign Aid In order to tide over the problem of export slowdown, Bangladesh has negotiated a multiyear loan program beginning from 2003-04 under the so-called PRSP-PRGF facility of the World Bank and the IMF. The continuation of the loan facility is subject to conditionalities involving reforms in areas ranging from budgetary deficits to privatization, financial system, public administration and other institutions of economic and political governance. There is no arguing that many of these reforms are quite essential for improving Bangladesh's economic performance, but most of them have no proximate relationship with the current balance of payment problem created by the global economic factors. One could thus legitimately question the wisdom of tagging such a broad reform agenda to what is essentially a balance of payment support. Bangladesh's macroeconomic reforms in the 1980s and the early 1990s were implemented under fairly rigid aid conditionality. 23 The articulation of the rationale for such reforms and their evaluation have come almost entirely from the donor side and very little from the side of the government. It is, therefore, difficult to assess the extent to which the reforms were actually "owned" by the government, and how much were imposed by the exercise of external aid leverage.24 It is a fair assumption that the later factor was the dominant one. The current Bank-Fund approach is supposed to encourage more local "ownership" and participation in the reform process; but it is difficult to see how this will happen in reality. While aid conditionalities are mainly regarding economic policy reforms, donors try to influence the budgetary allocations through project aid, which is the predominant form of foreign aid received by Bangladesh. The scope for such targeting of aid to priority areas is, however, limited by the possibility that local funds may be diverted to less-priority areas in response to the availability of foreign funds (so that foreign funds substitute rather than add to local funds - the so-called fungibility problem). Even then, the allocations for public development spending do represent donor priorities to a considerable extent, since foreign aid is spread over a large number of projects to meet only the foreign exchange component of those projects, thus requiring the government to commit local resources. Over the years, donor priorities have changed with the changes in aid ideas. For example, the previously high donor support for public investments in industry and power has drastically fallen, reflecting the policy shift in favor of private sector (Mahmud 2002a). On the other hand, health and population control seems to get the highest donor priority. The
Besides IMF's Structural Adjustment Facility (1986-89) and Extended SAF (1990-93), there were several credit facilities from the World Bank during this time involving conditionality regarding economic reforms. 24 On this, see Mahmud (2002a).
23

14

most recent changes in these aid ideas are towards more emphasis on the quality of aid utilization and on poverty alleviation. There is a legitimate concern of the taxpayers in donor countries to see that the benefit of foreign aid goes to the poor within the poor countries. While this shift of emphasis is welcome, a narrow interpretation of this approach may lead to a kind of "aid populism". This new poverty concern has sometimes led donors to chase the same projects, namely, those directly targeted to the poor. Surely, the government needs assistance to be able to provide safety nets and essential social services to the poor. But, in a pro-poor growth strategy, there are many legitimate areas of development spending besides directly povertyalleviating activities. Adaptive research for increasing crop yields, for example, has the potential to benefit the poor much more than many directly poverty alleviating programs possibly can. Incidentally, agricultural research is one of the most under-funded budgetary heads in Bangladesh.25 Even more importantly, the withdrawal of foreign aid from such areas as ports, power or natural resource exploration did not lead to the envisaged growth in private investment, including foreign direct investment, to fill in the gap. This has proved to be a serious constraint to industrial and export growth. Yet another new aid idea is to improve the quality of public service delivery and project implementation by incorporating mechanisms for accountability, beneficiary participation and people's empowerment. The incorporation of these ideas can definitely improve the design of development projects. However, institutional innovations of these kinds have to be based on intimate knowledge of ground realities; these can hardly be transplanted from outside, less so through the leverage of aid conditionality. Donors could perhaps make more use of local expertise in determining what works and what does not; and in the process, they could also reduce their own excessive delivery costs. Bangladesh's Health and Population Sector Project (HPSP) for 1998-2003 is a good example of what can happen when an otherwise well-intentioned initiative lacks ownership and knowledge of ground realities. The project was designed to provide integrated healthcare including population services along the guidelines adopted at the UNs International Conference on Population and Development (ICPD) held in Cairo in 1994. The program proved to be over-designed in terms of both resource needs and wide-ranging institutional changes; one result is that the family planning visitors have been withdrawn before the alternative system of reproductive healthcare could be put in place. Also, in order to overcome the so-called fungibility problem, the program stipulated for a five-year period the amounts of both local and foreign funding as well as the minimum levels of allocations to the priority areas like primary health care. But the arrangement proved unworkable because of its lack of flexibility.26 There seems to be little alternative for donors but to encourage the government, through concerted efforts, to devise its own plans and strategies for efficient public spending.
Bangladesh spends an extremely tiny proportion of its agricultural GDP on agricultural research, which was 0.38 percent during 1995-2000. This is far below the recommended 2 percent target for developing countries. 26 Such a fate for HPSP was predicted by many well in advance; see, for example, Mahmud and Mahmud (2000).
25

15

7. Concluding Remarks The discussion in this paper highlights the importance of many global economic issues in Bangladesh's development challenges. While most low-income countries depend largely on the export of primary commodities, Bangladesh has made the transition from being primarily a jute-exporting country to a garment-exporting one. This transition has been dictated by the country's resource endowment characterized by extreme land scarcity and a very high population density, making economic growth dependent on the export of labor-intensive manufactures. The issue of access of labor-intensive manufacturing exports from developing countries to the major global markets is, therefore, of utmost importance for Bangladesh. It is not easy, however, for a Least Developed Country (LDC) like Bangladesh, to specialize in manufactured exports. Having low wage costs can hardly compensate for its relative disadvantage in marketing skills and infrastructure (including transport, ports, standards and certification facilities). Moreover, the high degree of dependence of domestic industries on imported raw materials and intermediate goods makes it difficult for Bangladesh to satisfy the so-called "rules of origin" in getting preferential access for its exports in the markets of the developed countries. Thus, most of Bangladesh's garment exports are not eligible for GSP in the EU market, since its garment industry is largely dependent on imported fabrics.27 This problem has not got adequate attention, since the other major players in textile trade among developing countries are hardly affected by it. Bangladesh can potentially benefit from the European Union's decision to allow dutyfree import of "everything but arms" from the LDCs, and it would like to see the replication of such trade concessions in other industrialized countries. 28 But, since the same rules of origin as under GSP apply here as well, Bangladesh has not so far been able to take advantage of this facility. Bangladesh has also to worry about non-tariff barriers such as those relating to the environmental or labor standards. Anti-dumping actions are already under way against exports from Bangladesh and they are an important latent threat when the MFA is dismantled. The tough sanitary and phytosanitary regulations are also an impediment for diversifying into the prospective agro-processed export items. Clearly, the so-called Trade Related Technical Assistance sought by LDCs like Bangladesh cannot avoid looking at these issues in a comprehensive manner. Bangladesh's development experience also brings into focus the issues surrounding the role of foreign aid, such as the funding needs for achieving the Millenium Development Goals, poverty-targeting of aid, supporting the low-income countries to absorb economic shocks and the effectiveness of aid conditionality. Bangladesh stands to gain from an increase in the global flow of aid and its distribution according to the criteria of poverty prevalence and the presence of conditions that make aid effective in reducing poverty
Within the textile category, a relaxation in the rules of origin for knitted garments since 1999 has led to a surge of exports of knitwear from Bangladesh to the EU market. 28 Incidentally, the fact that "everything but arms" excludes rice, sugar and banana is not of much concern to Bangladesh, as it is to some LDCs.
27

16

(Williamson 2001). But, since the country is not included among the so-called highly indebted poor countries (HIPC), it will not benefit form the current debt relief initiatives. If, therefore, the debt relief were to be financed out of existing aid funds, it would be at the expense of countries like Bangladesh. Lastly, Bangladesh is also greatly concerned with the issue of freer movement of temporary workers across borders, given the important role of workers' remittances in its economy.

17

References Ahmed, S. and Sattar, Z., "Trade liberalization and poverty reduction: the case of Bangladesh", Paper presented at the World Bank's Annual Bank Conference on Development Economics, Bangalore, May 2003. BIDS (2001), Fighting Human Poverty: Bangladesh Human Development Report 2000 , Dhaka: Bangladesh Institute of Development Studies. Dorosh, P. and Q. Shahabuddin (1999), "Price stabilization and public foodgrain distribution: policy options to enhance national food security", Working Paper No.12 of Food Management and Research Support Project, Dhaka: Ministry of Food and International Food Policy Research Institute. Government of Bangladesh (2003), A National Strategy for Economic Growth, Poverty Reduction and Social Development (the I-PRSP document prepared for the World Bank and the IMF), Dhaka: Economic Relations Division. Harberger, A. C. (2001), "The view from the trenches: development processes and policies as seen by a working professional", in Meier, G. M. and J. E. Stiglitz, Frontiers of Development Economics: The Future in Perspective, Washington, DC: World Bank and New York: Oxford University Press. Mahmud, W. (1995) Recent macroeconomic developments in Centre for Policy Dialogue, Experiences with Economic Reform: A Review of Bangladeshs Development 1995 , Dhaka: University Press Limited. Mahmud, W. (1996), "Employment patterns and income formation in rural Bangladesh: the role of rural non-farm sector", The Bangladesh Development Studies, Vol. XXIV, Nos. 3&4, Sept.-Dec. Mahmud, W. (1997), "Macroeconomic update", in Centre for Policy Dialogue, Growth or Stagnation: A Review of Bangladesh's Development 1996, Dhaka: University Press Limited. Mahmud, W. (1989), "The impact of overseas labour migration on the Bangladesh economy", in Amjad, R., To the Gulf and Back; Studies on the Economic Impact of Asian Labour Migration, New Delhi: United Nations Development Programme and ILO - Asian Employment Programme. Mahmud, W. (2001), Adjustment and Beyond: the Reform Experience in South Asia , Basingstoke, U.K. and New York: Palgrave-Macmillan in association with International Economic Association. Mahmud, W. (2002), Popular Economics: Unpopular Essays, Dhaka: University Press Ltd. Mahmud, W. (2002a), "National budgets, social spending and public choice: the case of

18

Bangladesh", IDS Working Paper 162, Brighton, UK: Institute of Development Studies, University of Sussex. Mahmud, W. (2003), "Strategy for pro-poor growth in Bangladesh", paper presented at the seminar on Accelerating Growth and Poverty Reduction in Bangladesh, jointly organized by World Bank and the Bureau of Economic Research, University of Dhaka, Dhaka, June 26-27. Mahmud, W. (2003a), "Macroeconomics of poverty reduction: the case of Bangladesh", Dhaka: United Nations Development Programme (UNDP), processed. Mahmud, W., S. H. Rahman and S.Zohir (2000) Agricultural diversification: a strategic factor for growth, in Ahmed, R., S. Haggblade and T. Chowdhury (eds.), Out of the Shadow of Famine: Evolving Food Markets and Food Policy in Bangladesh , Baltimore: The Johns Hopkins University Press. Mahmud, S. and W. Mahmud (2000), Chapter on Bangladesh, in Forman, S. and R. Ghosh (eds.), Promoting Reproductive Health: Investing in Health for Development , Boulder: Lyne Rienner Publishers. Nadvi, Khalid (2003), "Globalization and the challenges to Bangladesh's garment industry", Institute of Development Studies, University of Sussex (memeo.) Osmani, S.R., W. Mahmud, B. Sen, H. Dagdeviren and A. Seth, The Macroeconomics of Poverty Reduction: The Case of Bangladesh, Dhaka and Kathmandu: UNDP Asia-Pacific Regional Programme of Macroeconomics of Poverty Reduction, 2003. UNDP (2003), Human Development Report 2003, New York and Oxford: Oxford University Press for the United Nations Development Programme. Williamson. J. (2001), "Financing for development: the implications of the Zedillo Report for South Asia", BIDS-SANEI Public Lecture 2001, Dhaka: Bangladesh Institute of Development Studies and Washington, DC: Institute for International Economics, processed. World Bank (1999), Bangladesh Trade Liberalization: Its Pace and Impacts , Washington, DC: World Bank (South Asia Region), Report No. 19591 BD. World Bank (2003), Bangladesh Development Policy Review, Washington, DC: World Bank (South Asia Region), Report No. 26154-BD (Draft).

19