Print this page
Themes > Current Issues
6.12.2005

The Employment-Poverty Link in Bangladesh

Jayati Ghosh

For nearly three decades now, the economy of Bangladesh has been growing at slightly more than 4 per cent, and per capita income growth even accelerated in the 1990s compared to the previous decades. In the 1980s, per capita GDP had grown slowly at the rate of about 1.6 per cent per annum. In the first half of the 1990s, the growth rate accelerated to 2.4 per cent and further to 3.6 per cent in the second half of the decade.

Of course this increase in the per capita growth was mainly because of the demographic transition involving declines in rates of population growth (from 2.1 per cent in 1990 to only 1.6 per cent in 2000), since there was no apparent break in the trend rate of growth of around 4 per cent over the entire period. However, there was clearly greater macroeconomic stability in terms of reduced rates of inflation (from an average of 9.9. per cent per annum in the early 1980s to an average of 5.6 per cent per annum by the end of the 1990s).

However, it is notable that this aggregate economic expansion has had less apparent direct impact on poverty reduction. It is certainly the case that the long-term trends in poverty show notable progress since Independence, from 71 per cent in 1973-74 to around 45 per cent in 2000.

Chart 1 provides estimates of poverty according to two different methods: in terms of a poverty line derived from the cost of basic needs, and in terms of direct calorie intake. While some reduction in the incidence of poverty is evident, this is actually less rapid than occurred during the 1980s. Clearly, the reduction in poverty has not been commensurate with the expectations generated by the macroeconomic pattern of relatively stable and non-inflationary growth.

Some of the reason for this is probably the substantial increase in inequality over this period, as evident from Chart 2. The main source of increasing inequality was the increasingly unequal distribution of both non-farm income and remittance income.

The outward-looking macroeconomic policy pursued by Bangladesh in the recent past did succeed in stimulating some parts of the economy, especially readymade garments and fisheries which were the most rapidly growing activities in the 1990s. But these activities still have a relatively low weight in the economy, and most (at least two-thirds) of the incremental growth in the 1990s originated from the non-tradable sectors - mainly, services, construction and small-scale industry. The demand stimulus for this came from three major sources - the increase in crop production in the late 1980s, accelerated flow of workers' remittance from abroad and incomes generated by the readymade garments industry.

However, these sectoral contributions of changing GDP were not exactly matched by changes in employment patterns. Chart 3 show the extent of growth of labour force in different sectors over the two halves of the decade of the 1990s. The most rapid growth has been in financial services, but these still constitute a very small part of total employment. Manufacturing employment has grown only marginally, after falling in the previous decade. While the new export sector of ready made garments has provided an important source of new employment (especially for women) total employment in aggregate manufacturing has actually declined, in both relative and absolute terms. There has been significant de-industrialisation, particularly in the traditional sectors, which have suffered from import penetration.

However, agriculture has shown substantial increase in employment generation to around 4 per cent per annum in the second half of the decade, reflecting the impact of various policy measure offering more incentives to cultivators from the mid-1980s onwards. While construction increased its share of GDP rapidly, the rate of employment generation decelerated in this sector. Other services sectors also showed decelerating employment growth in the second half of the 1990s.

In the early 1990s there was a marked improvement in the government's budgetary position along with an equally marked increase in the domestic saving rate. However, the increase in the saving rate was not matched by a commensurate response from private investment, at least in the early 1990s. However, in the second half of the 1990s, while public investment rates remained broadly the same, private investment increased causing the aggregate rate to increase to more than 21 per cent.

The sectoral allocation pattern of development spending has undergone some significant changes in the last two decades, reflecting the changing role of the government under the economic reforms. Allocations have fallen appreciably for a number of directly productive sectors - most notably, manufacturing industry, water resources, and energy, and agriculture, and increased for transport and communication, rural development, education and health.

Open unemployment rose risen from 1.8 per cent of the labour force to as much as 4.9 per cent, and was as high for women as for men, while underemployment in 2000 was estimated to be very high at around 31 per cent. What is also of concern is the dramatic increase in the ratio of self-employed to total workers, as indicated in Chart 4. In general the shift to self-employment in non-agriculture tends to be less rewarding in income terms for the poor, than the shift to regular work. However, there has been only a very slight, almost negligible increase in the share of regular employment.

So it is apparent that one crucial link to ensure more rapid poverty reduction - the generation of productive employment - has simply not been operating in a way that would show more effective results. rather, employment elasticities of output growth have been low or falling in most sectors, and the persistence of large-scale underemployment implies the continued proliferation of low productivity jobs, most typically now in the services sector.

There is a common perception that the high presence of micro-credit delivery systems in Bangladesh has operated to provide a cushion for poor households in case of shocks such as crop failures, floods and other natural disasters, etc. It has also helped to improve the relative position of women. However, the basic features of micro-credit (short-term, relatively small amounts, groups lending pressure for prompt repayment) mean that it has not contributed much to asset creation among the poor, or to sustained employment generation.

In fact, the poverty reduction that has occurred may be related more to other forms of public expenditure. The expansion of public transport infrastructure, especially roads networks in the 1980s, may have contributed to subsequent rural development which in turn assisted some of the reduction of poverty in that later period.

However, trade liberalisation had the counter effect of reducing the viability of many small producers, so the net effect of all the policy changes over the period is not clear. It is likely that some of the effects of openness were positive (as in the garments industry) others were adverse for livelihood and therefore poverty, and these were to some extent mitigated by the spread of public transport networks and the availability of micro-credit.

 

© MACROSCAN 2005