Rural Poverty in Bangladesh, India and Pakistan: profiles and policies.

Khan, Mahmood Hasan

I. INTRODUCTION

The Bangladesh--India--Pakistan subcontinent has about one billion people or one-fifth of the world's population. A majority of the subcontinent's people live in rural areas and depend directly or indirectly on agriculture and related pursuits. The highly differentiated rural population includes a large proportion of the poor in the throe countries. Their differentiated structure is based on (a) control of land through right of ownership and usufruct, and (b) employment opportunities for the labour power and its wages. Absolute poverty afflicts nearly 60 percent of the rural population in Bangladesh, 40 percent in India and 35 percent in Pakistan. The major causes of rural poverty are (a) landlessness and (b) lack of adequate employment (including low real wages) in or outside agriculture. Both these factors are responsible for keeping the poor consigned to low "entitlements" to income and commodities. The growth of the economy in general, and of agriculture in particular, has not produced the "trickle-down" effects large enough to counter the basic causes of persistent and high levels of rural poverty.

One of the serious problems in comparing the three countries is that generalizations or statements about each of them do not have the same meaning. There is regional diversity in rural India, and the same is true, but to a lesser extent, of Pakistan. Bangladesh has far less disparity between its rural areas. The major economic indicators for the three countries are given in Table 1. India's economy is the largest, the most diversified and industrially advanced. Pakistan has, however, the highest income per capita. As compared with Bangladesh, Pakistan is more urbanized and less dependent on agriculture. Bangladesh is most heavily dependent on agriculture; and is the least urbanized of the three countries. Pakistan's economic growth has been the most impressive, including the growth of its agriculture. All three economies have done better in the mid-1970s to mid-1980s than they did during the 60s. Bangladesh has not kept pace with the other two in expanding per capita output of agriculture in general and of food in particular. But all three countries have apparently become "self-sufficient" in staple grains of wheat and rice. A word of warning is in order here. The state of agriculture in all three of them remains fragile. Technical advance has not fully overcome the major constraints on agricultural growth imposed by Nature. What is even more important is that higher supplies of food, does not mean that it gets distributed proportionately across regions and income groups or classes. Distribution of food depends on "entitlements", which in turn are determined by the endowments of income earning assets, wage employment, and prices.

External economic relations, given in Table 2, have been less important to India's growth than to either Bangladesh or Pakistan. Pakistan's economy is the most "open" of the three to trade, and its current account deficit has also been the largest. The exports of the three countries have grown more rapidly in the second decade than in the first; and imports have also grown faster in the same period. Capital inflow through workers remittances has been as high as the export earnings in Pakistan, but it accounts for less than one-half in Bangladesh and just over one-quarter in India. India has the lowest burden of external debt. Bangladesh has the highest burden of total debt which constitutes nearly 40 percent of its GNP. Pakistan's external debt liability is also very large: its total debt is 30 percent of the GNP and its service payments take up over 27 percent of its export earnings.

The differences in per capita income and growth of the economy are not altogether consistent with the differences in the "quality of life" in the three countries (see Table 3). An average Pakistani is better off than his counterparts in India and Bangladesh: his income is higher and he consumes more calories and proteins every day. However, an average Indian, male or female, expects to live a little longer. Fewer infants and children die in India than in Pakistan. India also has a greater supply of physicians and nurses per person. Bangladesh is not far behind Pakistan on life expectancy, infant and child death rates. It has, however, fewer physicians and nurses per person than Pakistan. It is interesting to note that Pakistan has the highest population growth rate than India and Bangladesh, mainly due to higher birth rates. The latter countries have shown a distinct slowdown in birth rates, which is not as clearly seen in Pakistan. Pakistan's performance on literacy and education has been the least impressive while both India and Bangladesh have higher literacy rates as well as enrolments in primary and secondary schools.

II. RURAL POVERTY PROFILES

Poverty may not always be transparent, but it blights the lives of millions in most underdeveloped countries. No matter how poverty is defined, it always implies a state of deprivation and dispossession. The individual's judgement on poverty depends on (a) how deprived or dispossessed he feels in some absolute measure and (b) his own predicament relative to others in the society. From a society's point of view, the concept of absolute poverty implies existence of some socially acceptable norm for subsistence or survival. But most societies have also made commitments to alleviate relative poverty by focusing on income'(or asset) redistribution. The alleviation of poverty is the dominant rhetoric in the Third World.

While normative judgement of individuals and households may differ widely, poverty is best described in terms of the following interrelated aspects:

(a) lack of income-earning assets and flow of income;

(b) vulnerability to contingencies of sickness, debt, crop failures, etc.; and

(c) powerlessness to influence the economic and social environment.

Viewed in this way, a large proportion of rural households and individuals in most underdeveloped countries are indeed poor.

A comparative study of rural poverty in Bangladesh, India and Pakistan raises many issues of what is meant by poverty and how it can (or should) be measured. At one level the concept of poverty is simple enough: it is a state of deprivation and dispossession. However, it does not fully answer the main questions. For one thing, the relative and absolute concepts are different measures of poverty, even though they are closely associated with inequality of-income distribution. Relative poverty arises directly from the inequality in income distribution, irrespective of the level of income or the state of deprivation of those at the bottom end of the income scale. Absolute poverty, on the other hand, represents a societal view of deprivation through some physical manifestation of consumption, nutrition, etc. Absolute poverty, however measured, will decline if, and only if, income inequality falls in a way in which (a) the value of the Gini coefficient is lower and (b) the new distribution (Lorenz) curve lies entirely above the old one. If the mean income of distribution rises, without any change in the inequality measure, absolute poverty falls; which means there is an inverse relationship between the two. However, as the measure of inequality (Gini coefficient) is independent of the mean income, a change in the latter does not necessarily reflect a change in relative poverty. It is, therefore, possible that changes in relative poverty may not have a significant impact on absolute poverty. But it is also possible that absolute poverty may be greatly alleviated by adjustments in the existing inequalities. The focus of this paper is on absolute poverty in rural areas of Bangladesh, India and Pakistan.

What is the best measure of absolute poverty? There is none known so far. Most of the measures are related to the level of income (purchasing power) necessary to provide a "minimum" consumption of (a) calories from a basket of food commodities, or (b) "basic needs", including food, shelter and clothing. The first measure has been used more extensively than the second in establishing a "poverty line" by carrying out a headcount of the poor in rural and urban areas. Social scientists, particularly the economists, have developed a substantial body of literature on both the methodology and measurement of absolute poverty in underdeveloped countries. Most studies of absolute poverty in India, as in other underdeveloped countries, have so far used the method of headcount below the poverty line which is based on consumption of calories or, in other words nutritional intake.

The headcount approach has serious limitations for the comparative study of poverty between countries and over time in the same country. For one thing, nutritional intake on the average means little in light of the variations required for different age and working groups. Also, various baskets, each perhaps requiring a different purchasing power, can produce the same level of nutrition. For this reason alone, poverty lines within the same country can vary at the same time and over time. It makes comparisons very difficult. Then there is the problem of distribution of the poor below the poverty line. How close are they on each side will determine the impact of incremental incomes or consumption on the alleviation of poverty. These distributions are not easy to estimate even in cross-section studies; they are even more difficult to study through time-series data. In the end what matters is the identification of the poor and the severity of their poverty for designing public policy. As long as these limitations are kept in mind in interpreting the measured variables, the poverty line studies can yield meaningful results for research and policy making.

The headcount (given in Table 4) of the rural poor in Bangladesh, India and Pakistan shows that (a) a greater proportion of the rural population is poor m Bangladesh than in India and Pakistan, and that this proportion has been rising; (b) the proportion of the rural poor in India remained around 45 to 50 percent during the 1960s and 1970s, but this proportion has fallen to around 40 percent in the early 1980s; and (c) rural poverty first rose in Pakistan in the 1960s, but had fallen since to less than 40 percent in 1979. Assuming that the proportion of the poor in the rural populations of Bangladesh, India and Pakistan have fallen to 50, 35 and 30 percent, and using population estimates of 1985, there were 255 million absolutely poor people in the rural areas of the three countries, making up 36 percent of their rural populations and 27 percent of their total populations. It is still more difficult to identify who these poor are. But the rural poor in these countries may be identified better by looking closely at the agrarian structures in Bangladesh, India and Pakistan.

As the data in Table 5 show, small owners--hose owning 1 hectare in Bangladesh and 2 hectare in India and Pakistan--form the overwhelming majority (three-quarters in Bangladesh and India and two-thirds in Pakistan), but they own one-third in Bangladesh, one-quarter in India and one-fifth in Pakistan. Their numbers and areas have both increased. The very large owners (4 hectare for Bangladesh, 6 hectare for India and 10 hectare for Pakistan) are 3 to 6 percent of all owners, but claim one-quarter in Bangladesh, 40 percent in India and 44 percent in Pakistan. Their proportion has fallen in numbers and area. Most of the farms operated are also small in the three countries (Table 6): one-half are less than 1 hectare in Bangladesh, three-quarters in India, but one-third in Pakistan. They have one-fifth area in Bangladesh, one-quarter in India, but just over one-twentieth in Pakistan. Large farms-owning 3 hectare in Bangladesh, 10 hectare in India and Pakistan--are less than one-tenth in Bangladesh, 2 percent in India and one-tenth in Pakistan. They possess one-third area in Bangladesh, one-quarter in India, and nearly one-half in Pakistan.

Landownership is far more concentrated than access to land through use. Landownership is highly concentrated in both India and Pakistan, but not as much in Bangladesh. Land use concentration (shown in Table 7) is also the lowest in Bangladesh, but it is the highest in India. It has gone down substantially in Bangladesh, slightly in India, and has somewhat gone up in Pakistan. Rural income is the least concentrated in Pakistan; it went down in the 1960s, but went up in the 1970s. The reverse seems to have happened in India. In Bangladesh, it did not change much in the 1960s, but went up in the 1970s. It appears that income concentration goes down in the period of high growth and remains stationary or goes up in the stagnant or slow growth period!

Landlessness is difficult to define because of conceptual and data problems. Three alternative definitions of landlessness are given below:

(a) who own no land,

(b) who operate no land, and

(c) whose major source of income is wage employment, even when they own or operate some land.

There is considerable overlapping between these categories, which creates confusion about the incidence of landlessness. The best way would be to construct the following four-way classification:

 Owning Land Not Owning Land

Operating Land A: O-O C: Pure Tenants
Not Operating Land B: Landlords D: Pure Landless

There are some estimates of the landless, based on landownership and land use (Table 8). In terms of land ownership, landless households constitute 62 percent in Bangladesh, 45 percent in India, and 58 percent in Pakistan. They include even those households which have less than 0.4 hectare of land. The pure landless households were 29 percent in Bangladesh and 26 percent each in India and Pakistan. By land use, or on the basis of access to the use of land, landless households constituted 29 percent in Bangladesh, 31 percent in India, and 22 percent in Pakistan. The pure landless (with no access to land) were 21 percent in Bangladesh, 16 percent in India and 20 percent in Pakistan. There is a large number of the marginal and small cultivators which has less than 2.0 hectare to cultivate; they are 30 percent in Bangladesh, 35 percent in India, and only 12 percent in Pakistan. Pakistan has a much higher percentage (66 percent) of households with access to land of over 2 hectare.

Poverty in rural areas may not be affected much by ensuring access to land if (a) land is plentiful or distributed equitably through communal or tribal rights, and (b) reliance on non-farming occupations is high and rising. Poverty is also not reduced by holding bad quality land, for instance in semi-arid regions. A little land can avoid poverty if and only if land has a high return per hectare. Land reduces the risk of poverty because it can be mortgaged in emergency. However, landlessness alone is not important for analysing poverty. It should be seen together with (a) the conditions of cultivation of land by those who operate it (size of holding, productivity of land, and tenure), and (b) condition of wage employment for those who depend on wages alone for household incomes. A majority of the rural poor come from households which are completely landless (agricultural labourers) and which have some access to land as small owner-operators and tenants. They all have to sell a part of the household labour to earn their incomes.

Labour-force data are even scarce and of poor quality, particularly in Pakistan. But they show (Table 9) that the proportion and absolute number of households dependent primarily on wages are increasing, or of those dependent mainly on income from land are declining in numbers. Nearly one-quarter of the rural populations of Bangladesh and Pakistan are in the labour force. Mainly because of the higher female participation rate, a much higher proportion of India's rural population participates in work than in Pakistan or Bangladesh. This is shown below.

 Bangladesh India Pakistan
 (%) (%) (%)

working on their own farms 50 25 33
helping on farms 25 23 33
agricultural labourers 25 39 10
outside agriculture - 13 25

Wage employment is becoming important in all three countries. But the total dependence on wages is not increasing. One-tied of the rural households in India depend on wage employment (of which 50 percent are agricultural labourers and 50 percent are marginal and small farmers). Around 40 percent of the rural households in Bangladesh are labourers, of which three-quarters depend mainly on wage labour. Pakistani data do not permit this breakdown.

Rural (real) wages are an indicator of the incomes of the poor rural households. They depend on (a) agricultural growth, (b) employment prospects within and outside agriculture, (c) rural-urban migration, and (d) public-sector policy on minimum wages, technology, etc. The real rural (agricultural) wage has remained stationary and has even fallen in Bangladesh, has risen somewhat in India, and has increased impressively in Pakistan (Table 10). The condition of the rural poor households also depend on (a) period of gainful employment, and (b) demographic composition of the household.

Rural poverty is persistent among certain groups--namely, among the landless agricultural labourers (including the low-caste and tribals), and the marginal (small) owner-operators and tenants. They depend increasingly on wages and employment. Rural populations are still increasing in absolute numbers, but cultivated areas are not likely to increase. Agricultural growth has not been stable or evenly distributed by regions or income groups. The prospects of finding jobs outside agriculture have not become brighter either.

III. POLICIES FOR ALLEVIATION OF RURAL POVERTY

Poverty alleviation means raising the entitlements of the rural poor. A weak asset base, with increasing or even stationary state of landlessness (high concentration of land, population growth, and limited area), makes the poor vulnerable in the labour and commodity market, Landlessness can be countered by:

(a) improving terms of trade;

(b) increasing land productivity;

(c) increasing employment opportunities and high real wage rates; and

(d) the growth of non-farm activities in the rural and urban areas.

However, these forces, left to themselves, do not work well. In none of these countries has there been a long and sustained period of rapid economic growth, particularly in agriculture. There is also evidence that the rural poor have not been the major beneficiaries of agricultural growth, because they were unable to participate in this process. Growth policies have not necessarily been consistent with the goal of alleviating poverty. National resources given to the development of agriculture and social sectors in the five-year plans have not been focused on the target groups (Table 11). The growth experience in Bangladesh, India and Pakistan shows that poverty has to be attacked directly by (a) distributing more equally the income generating assets (land and non-land assets) and (b) expanding wage employment and creating human capital.

These strategies for alleviating rural poverty can be classified as follows:

 Group I Group II

Income Generation Through Creation Income Generation Through Guarantee
or Redistribution of Assets of Employment, Wages, Consumption

1. Land Reform Measures 1. Employment Guarantee
 (a) land redistribution (a) food for work
 (b) new tenancy contracts (b) minimum wage/work guarantee
 (c) land consolidation

2. Non-land Assets: Capital 2. Food Distribution
 (a) livestock/other enterprises (a) rationing system
 (b) labour training for skills (b) school/farm meals

3. Public Infrastructure for Small 3. Rural Industrialization
 Farmers/Poor Regions (a) non-farm employment
 (a) water/irrigation system (b) labour-intensive technology
 (b) subsidy on credit/inputs
 (c) research/extension
 (d) price support/procurement.

Land Reform Measures

The land reforms in all three countries have been generally disappointing. Two basic reasons stand out for this; one is the inadequate political support for land redistribution and new tenancy contracts, and the second is the poor implementation through the existing bureaucracy without the participation of the prospective beneficiaries. The states of Kerala and West Bengal in India have been the only exceptions, because the rural poor there were allowed to organize themselves. In addition, political and infrastructural support was provided to help them to gain their claims.

Land redistribution in Bangladesh has been the least impressive, although the first tenancy laws were enacted in the early 1950s. In fact, land ceilings have been changed twice since then: they were raised in 1961, and lowered again in 1972. The total amount of land resumed in 35 years is 300,000 hectare, of which only 137,000 hectare have been distributed. The government toyed with the idea of redistributing land again in 1983, but nothing has been done so far. However, it has legislated long-term tenancy rights to sharecroppers after 5 years of tenancy and has introduced a minimum rural wage. The two most important impediments to land redistribution in Bangladesh are; first, there is not enough land, and second the data on rights of landownership and tenancy are very poor. No serious attempt has been made to register the legal claims. This has added to the problems of tenant eviction and costly litigation.

In India, the land ceiling laws were first enacted in the 1950s, and revised in the early 1970s. A majority of the Indian states now have laws about ceilings on agricultural land holdings. However, the surplus area declared to date amounts to 2.97 million hectares or less than 2 percent of the total cultivated area. Of this area, only 1.8 million hectares have been redistributed. The total number of beneficiaries are reported to be 3.3 million. It has been estimated that even with the present land ceilings, there should be nearly 6 million hectares available for redistribution. Tenancy contracts have not been revised in most states, with the consequence that tenants remain at the mercy of the landlords. The states of Kerala and West Bengal have pursued better the goals of land redistribution and revision of tenancy contracts. The land reform measures there have made a considerable impact on the lives of millions of the landless and near-landless. The follow-up support system to new landowners has been an important part of the land reform programme there.

The land reform measures in Pakistan have so far resulted in the redistribution of about 1.4 million hectare to about 256,000 beneficiaries. The land ceilings in Pakistan are the most generous. Hence the potential for land redistribution to the landless. A 25 hectare ceiling on land holdings can release 3 million hectare for redistribution. While the tenancy laws have removed the most blatant excesses against the sharecroppers, their enforcement is greatly hampered by the lack of political support given to the tenants.

There are several common features of the existing policies on land redistribution and revision of tenancy contracts in the three countries:

(a) The data base for titles of ownership and tenancy rights is at best inadequate to establish individual claims.

(b) Land ceilings have remained high because of poor legislative action and evasion, with the exception of Bangladesh where there is perhaps little land to redistribute.

(c) The poorest quality of land has often been surrendered and transferred.

(d) The beneficiaries, especially the marginal owners and tenants have been small and the landless labourers have been excluded.

(e) Tenancy rights have not been enforced and evictions of tenants remains a serious problem.

(f) Minimum wage laws exist only in name, except in the Food for Work Programme (FFWP) in some states of India and in Bangladesh.

(g) The follow-up or support for the new landowners in terms of specific inputs and credit remains weak and uncertain.

(h) Political support for peasant organizations is not forthcoming, and the existing bureaucracy is not suited to deliver the needed support.

Distribution of Non-land Assets and Creation of Skills

The idea of extending loans to the landless poor for non-land assets to generate new income was first experimented in the state of Rajasthan in India. The purpose of the antyodaya programme in Rajasthan was to lift the poor above the poverty line by providing them loans to buy income-generating assets other than land. This programme had been preceded by various rural-development programmes in the sub-continent, starting from the community development and village-aid programmes in the 1950s. However, they did not aim at alleviating the problem of poverty for specific groups or regions. The Integrated Rural Development Programme (IRDP) in the early to mid-1970s was in the tradition of providing social and physical infrastructure to rural areas with emphasis on agricultural development through new technology, etc. The experience in India and Pakistan was that the IRDPs there strengthened the economic and social positions of the rural elite and transferred public resources to a bureaucracy which was not capable of delivering inputs and services to the majority of the rural population. Inadequate accountability and low participation of the poor in IRDP made the programme further wasteful through mismanagement and corruption.

The IRDP in India was remodelled in 1979 in the light of the antyodaya experiment of Rajasthan. It was initiated in less than one-half of the blocks, and was then extended to the entire country. The IRDP is funded by the central and state governments on a 50:50 basis. It uses a mixture of subsidy (25 to 50 percent) from the state and low-interest loans from the banking system. The aim of the programme is to raise the income of the rural poor above the poverty line on a permanent basis. It is basically a scheme to provide incremental income through selfemployment. The Training of Youth for Self-Employment (TRYSEM), which is an integral part of IRDP, is designed to give specific training to the youths (18 to 35 years) from the landless poor households to be used for self-employment in or around the village. The village panchayats, Block Development Officers (BDOs), and the District Rural Development Agency (DRDA) are expected to play an important role in formulating and executing each IRDP project.

The IRDP experience in India has several interesting features. The programme was allocated Rs 45 billion during the Sixth Plan. Several studies have been done on its impact and the lessons learnt about its implementation. The estimates are that 16.5 million families benefited from the programme; investment per household went up from Rs 1,642 to Rs 3,344 per household; 17 percent of the target families crossed the poverty line and 51 percent improved their income levels. The TRYSEM programme trained about 1 million youth. These studies have also revealed some of the major weaknesses of the programme: only about 3 percent of the poor rural households were able to cross the poverty line and that too temporarily, because a long-term impact on their income would have required at least Rs 9,000 per household; there were significant "leakages" from the system: about 20 percent of the beneficiaries were wrongly identified as poor; projects and activities were not always well chosen and the infrastructural support to make them viable was poor, etc. Funds were equally distributed to the blocks but without any concern about their actual needs. The TRYSEM programme has also been criticised because a high percentage of the trained youth did not go for self-employment; but looked for jobs in or out of their villages. One of the most serious problems has been the implementation of these programmes within the existing bureaucracy, particularly in those areas in which the representative panchayats are either non-existent or weak.

The IRDP in Bangladesh is focused on the development of village infrastructure for agricultural development, based on a rural works programme within the existing two-tier system of cooperatives in the Thanas. Its foundation dates back to the Rural Works Programme of the 1960s. The size of the programme has shrunk significantly since the mid-1970s due to inadequate resources. Some of these resources have been chanelled into the Food for Work Programme (FWP) to generate employment for the landless agricultural workers. As an anti-poverty programme, the Grameen (rural) bank scheme was introduced in 1979 in a small number of Thanas to extend low-interest loans to the landless poor for buying non-land assets (livestock, etc.). The borrowers are required to repay these loans in weekly instalments and contribute to a common fund to be used in emergency for the poor in the village. The landless are defined as those with less than 0.25 hectare. Over 50 percent of the borrowers are females. The average size of loan was Tk. 3,300. The total number of beneficiaries in 1980-85 was 100,000, and the project covered less than 3 percent of the villages. It has been estimated that the incomes of beneficiaries increased by 30 percent and their saving levels rose to 10 percent.

The experience of the Grameen banks has already brought to light some of their weaknesses and strengths. For example, the landless agricultural workers were excluded from this project, with benefits flowing to those with some land. Many of the activities or products could disappear without replacement. A cow or a flock of poultry could die or be sold in emergency. Similarly, a sewing machine could be sold. In view of the small numbers and limited investment, it is not clear if the impact on the individual household income was significant or lasting. Serious doubts have also been expressed about the impact of these ventures on labour productivity. One of the advantages of the project has been its ability to attract the poor at the village level and introduce them to the idea of self-employment. This is particularly important for females in many households. The problem of leakages has not been very severe because the project relies on decentralized decision-making, in which villagers can play an important role in the local setting.

In Pakistan, there has been virtually no such programme to enable the landless poor in rural areas to own non-land assets for generating new or supplemental incomes. As part of its "Islamization" policy, the government has introduced two levies on Muslims: zakat (since 1980) as 2.5 percent of the wealth during a year, and ushr (since 1983) as 5 to 10 percent of the value of output of agriculture (crops and livestock products) during each season. The revenue from zakat and ushr has been transferred to the "eligible" poor. (It cannot be spent for any activity other than the welfare of the poor.) Zakat councils have been set up at each administrative level, with representation from the elected council members at the local level in rural and urban areas, to distribute the revenues from zakat and ushr levies. The decisions about the amounts to be transferred to individual families or to welfare institutions are usually left to the local councils in each province and district.

Zakat collections during 1980-86 were Rs 6.2 billion, out of which Rs 3.9 billion were distributed to 7.9 million persons, representing a transfer of Rs 498 per beneficiary. These payments have been given as cash grants to the destitute families; as educational scholarships to the children of poor families; as grants to religious schools and to welfare trusts for the health care of the poor. There are no details given about the breakdown of the beneficiaries by the urban and rural poor. A rough estimate is that about 50 percent of the rural poor have benefited at least once from one of these grants. In view of the nature and size of the transfer payment, it is doubtful if it can make a big impact on the resource availability of the rural poor, except in those families in which a scholarship grant is provided for the education or training of a number of family members or over a number of years. The subsistence grants on a regular basis can only alleviate the problem of destitution for a large number of families. The structure of zakat councils is reasonably decentralized, which helps in reaching the poor without excessive bureaucratic red-tape and corruption. However, the Central and Provincial Zakat Councils retain a significant portion of the total revenue from zakat and ushr.

A "Five Point" programme was launched in Pakistan for four years from 1986. One of the major goals of this programme is to improve the conditions of life for the poor in rural areas. Under it, nearly Rs 118 billion have been allocated to activities ranging from rural social and physical infrastructure (electrification, health centres, schools, adult literacy, water supply, and roads) to land for housing (a 7 marla plot for each family without shelter). In the first year, 1986-87, it has spent Rs 23 billion or nearly one-half of the annual development plan. One of the problems in assessing the impact of this programme is that there is no independent evaluation of its achievements so far. The only source of information is the Planning Commission. Given the existing administrative and political structure in the country, it is hard to say that the Five Point Programme would achieve most or any of its lofty goals by 1990. Similar programmes in other countries have brought great disappointments.

Income Generation through Employment/Food Guarantee Programmes

The landless poor have become increasingly dependent on wage income, derived from work in or outside agriculture. This has also become an important source of income for sharecropping tenants and marginal owners, because of their low income levels and the available slack time during the crop season. In all three countries, a rural works programme was used in the 1960s to create employment for agricultural labourers and poor cultivators and build rural infrastructure at the village level. Wages were paid in cash or kind under the PL-480 food aid programme from the U.S. The experience in some Thanas of what East Pakistan (now Bangladesh) was particularly impressive, although it did not extend to many parts of the country.

In both Bangladesh and India the rural works programme was converted into a Food-for-Work Programme (FFWP) in the mid-1970s. The rural-works programme in Pakistan was part of the IRDP in the early 1970s, but it was abandoned in the late 1970s. The FFWP in Bangladesh has been a major source of using the food aid (particularly wheat) for providing jobs to the landless poor in rural areas. However, this is not an employment guarantee programme. It provides only part-time employment. It created employment for 110 million mandays. Not more than 3 percent of the unemployed labour time was used in 1984-1985. Almost three-quarters of the labourers were from the landless or near-landless households. The minimum wage, paid mostly in kind, was 3 kgs. of wheat per day.

Several issues have been raised about the impact of FFWP in Bangladesh. First, it does not guarantee job or food to anyone who is landless and poor. Second, the minimum wage has not always been paid to workers. Third, a large proportion of the workers have sold their food for cash, so their consumption of food may not have gone up by much. A fourth problem is the disproportionate representation of workers from the landowning households, who give up their own farm work to earn the minimum wage in kind and then convert it into cash. Finally, the quality of infrastructure such as water channels, culverts, roads, etc., built by these workers has not been well controlled both in terms of design and life.

India's FFWP was restructured in 1980 as National Rural Employment Programme (NREP). The central and state governments pool their resources (Rs 18 billion during 1980-1985) to provide employment to landless workers. Wages are paid in cash and kind (grain). Workers build rural infrastructure, e.g. social forestry, village tanks, roads, schools, land reclamation, and watercourse improvement. The NREP was supplemented in 1983 by a new rural employment programme, called the Rural Landless Employment Guarantee Programme (RLEGP). It was designed to guarantee work for 100 days in the year to at least one member in each household of the landless labourers. The RLEGP was started along the lines of the Employment Guarantee Scheme (EGS), which had been functioning in the state of Maharashtra since 1974.

The estimates are that the NREP and RLEGP in India created 2,035 million mandays of employment during 1980-1985; they provided an average daily employment of 1.1 million mandays or 7 percent of the estimated unemployed mandays in rural areas. The planners had estimated that there were 15.36 million daily workers looking for work in 1980. The daily wage rate increased from Rs 5.63 to Rs 8.92 during 1981-1985. The NREP has provided short-term employment to landless workers, on average 50-55 days in the year. The RLEGP did not get the required resources--it was given Rs 4 billion for two years--to fulfil its mandate. The Seventh Five-Year Plan proposed to increase funding to reach the goals of the NREP and the RLEGP. One of the serious problems in designing and implementing these programmes has been the lack of an effective political and bureaucratic framework which would guarantee the participation of the prospective beneficiaries and ensure the delivery of quality infrastructure to rural areas.

IV. CONCLUSIONS

The problem of rural poverty in the three countries remains intractable mainly because it has not been confronted by policies with the biggest impact on the target groups; and because they have relied on "soft" policy options. The agricultural sectors in these countries have not been dynamic enough to sustain the growth process. What is even more disturbing is that agricultural growth has not provided new jobs and incomes to the landless poor. This sorry state of affairs is partly reflection of the institutional impediments and partly a result of policies that have been contradictory. Industrial growth has not touched many of the rural communities; and the relative abundance of labour in these countries has not been used to advantage in selecting the industries which have the biggest impact on growth in both rural and urban areas. A direct attack on poverty requires that income earning opportunities are provided to the rural poor, by making available assets like land and human capital and through providing productive employment in and out of agriculture. The efforts made so far in the three countries do not represent an effective strategy to alleviate rural poverty.

Comments on "Rural Poverty in Bangladesh, India and Pakistan: Profiles and Policies"

Professor Mahmood Hasan Khan has conditioned us through the years to expect from his writings penetrating and statistically well-documented analyses of rural poverty. He has once again delivered on our expectations--and he has delivered his lecture with great verve and least statistical indiscretion. In an area where Professor Khan has not ventured before, the comparative analysis of the three countries of the sub-continent, he has collected a huge amount of data in comparing poverty, landlessness, external economic relations, the openness of the economy, the quality of life and so on. A number of stories emerge from studying the differences among the three countries on various performance indicators. These have been discussed by the author and some are quite convincing. It is the feeling of this reviewer, however, that the big story is not in the differences in individual performance indicators but in the similarities of the basic development outcomes. If this is correct, it becomes a puzzle which is left unanswered by the paper.

Absolute poverty is difficult to define and to measure on a comparable basis. Professor Khan makes a valiant effort by putting disparate sources together. He finds poverty as a percent of rural population at 60, 40 and 35--or alternatively 50, 35 and 30--for Bangladesh, India and Pakistan, respectively. This represents about 255 million people, or 27 percent of the total population. With the possible exception Of Bangladesh, the outcome is rather close for the two countries. In a paper, that I presented in another session of this meeting, I have found poverty levels even closer together, at 22, 27 and 23 percent of the total population, for Bangladesh, India and Pakistan. The real difference on the poverty score is somewhere else: there has been no famine in Pakistan. Still, when it comes to broader guages of the quality of life, Pakistan is behind India (e.g., life expectency, child mortality, number of physicians and nurses) and occasionally behind Bangladesh (education, both male and especially female).

Landlessness and near-landlessness (less than 0.4 hectares), whether defined in terms of ownership or operational holding, is roughly comparable in the three countries--between 25 and 30 percent for operation and double that for ownership. Where the three contestants seem to differentiate themselves is the rate of increase in rural real wages. With an index of 100 in 1975, real agricultural wages increased by 0, 33 and 86 percent by 1984, respectively, for Bangladesh, India and Pakistan. The substantial wage increase in Pakistan may be misleading as the market wage rate applies to members of the labour force, that is, those employed or looking for work. The labour force in Pakistan is a highly selective concept. According to the Labour Force Survey (1985-86), the participation rate for males is 50 percent but that of females is 6 percent (as opposed to about 30 percent for other LDCs). That women do not participate in the labour force does not necessarily mean that they do not work. The Agricultural Census (1980), for example, reported that women were 42 percent of the 22.8 million economically active persons in agriculture. This implies a female participation rate of 74 percent. Women are omitted from the labour force reporting simply because most of them (96 percent) work on their own holdings. Moreover, since these working women could not accept paid employment outside their own farms, they have hardly benefited from the real wages increase. In such a tightly segmented labour market, the only way to increase the real incomes of those in hidden unemployment is to increase the productivity of smallholders in agriculture.

In my opinion, the big story of the tripartite comparison lies in the similarity of development outcomes which, if normalized by initial conditions, indicate vastly dissimilar economic performances. Initial endowments, especially in agricultural infrastructure, since the time of the Partition and more recently, have been unequal in the three countries. Population pressure on resources and ecological systems have been quite different. Economic orientation and socio-political systems have also differed. The similarity of development outcomes in the light of these differences in initial conditions is counterintuitive. Moreover, it brings into question the economic performance of Pakistan first, and of India next, in comparison to that of Bangladesh.

It seems to me that this is the real story which is hidden in Professor Khan's comparisons.

Pan A. Yotopoulos

Stanford University, U.S.A.

Comments on "Rural Poverty in Bangladesh, India and Pakistan: Profiles and Policies"

Professor Mahmood Hasan Khan is well aware of the methodological issues involved in the measurement of poverty and in the comparison of such estimates across regions and overtime. However, at the expense of incurring his wrath, I would like to reiterate a note of caution. The problem that Professor Khan has tackled is indeed extremely complex and difficult. It is as difficult to conceptualize as it is to measure. The concept is multi-dimensional. It depends in part on the absolute level of income and consumption and in part on the relative distribution of income, wealth, power and social status. It depends, in part, on the ease with which it is possible to move from one occupation or social category to another. The concept has stock and flow elements, absolute and relative dimensions and static and dynamic characteristics. Not only is the concept difficult to capture in terms of simple indicators but it is almost impossible to expect the conditions that affect it to be similar across regions or across time. The sensitivity of the estimates of rural poverty especially in Pakistan's case, has been well documented.

Many studies have found poverty increases even where there have been quite rapid increases in incomes. When the poor starve it is not mainly because there is no food but because they do not have the wherewithall to acquire food. It is widely recognized that the problem of world hunger cannot be solved merely by attempting to increase production. The solution requires better distribution and more productive employment both to increase incentives to expand output and to create effective demand for greater food output. In most countries where poverty is widespread the resources needed for development are available. They are, however, unutilized or poorly utilized. Underutilization of labour and land is often accompanied by the underutilization of capital as well. Factor prices are generally distorted --either as direct conscious policy or as a spill-off of other policies. This leads to a sub-optimal choice of techniques and distortions in the distribution of the product.

A lot of work has been done on the measurement of rural poverty and on the policies of rural development designed to alleviate it. The main issue still remains: what can be done to. help alleviate this poverty? In particular, the question that should concern us directly is; what is the role that the research community can and should play in alleviating this scourge?

I have recently reviewed some work done by the International Agricultural Research Centres supported by the consultative group on International Agricultural Research. Their work, which incorporates the objective of improved human nutrition in agriculture research, is a great step forward in maximizing the contribution of the research community to the alleviation of absolute poverty and related ills such as insufficient food intake and malnutrition. These centres have set up commodity priorities that reflect nutritional considerations. The desirability of changing the nutritional characteristics of certain commodities is, however, considered in relation to the effects of changes in the composition of the diet. In some instances, dietary deficiencies can best be alleviated through changes in individual commodities. In others, diversifying a diet and increasing consumption of commodities already available offer the best solution.

Research is being done on improving the nutritional characteristics of a given commodity, without in any way jeopardising other characteristics such as yield, consumer acceptance as reflected in prices and resistance to pests etc., which are important considerations in the farmer's adoption of these commodities. This research is taking explicit consideration of the technology characteristics of new and existing commodities and of the production systems prevailing in the areas for which this research is being conducted.

This is the type of work designed to alleviate some of the associated ills of poverty to which we as professional researchers can contribute.

Sohail J. Malik

Pakistan Institute of Development Economics, Islamabad

Appendix I

ANTI-POVERTY PROGRAMMES IN BANGLADESH

1. In the 1950s: Land Reforms were undertaken to abolish intermediary rights of landlords and other rentiers; land ceilings were lowered and land was redistributed; tenancy laws were passed to secure rights for the sharecropping tenants.

2. In the 1950s: Village-Aid Programmes were introduced for community development at the village level through the existing government line departments which involved the locally constituted or elected village councils.

3. In the 1960s: Rural Works Programmes (RWP) were undertaken to create employment for agricultural labourers and poor cultivators and to build rural infrastructure at the village level. They involved the "Basic Democracy" councils at the Thana (sub-district) level. Wages were paid both in cash and kind under the food-aid programmes.

4. From 1971 to-date: Integrated Rural Development Programme (IRDP) was initiated in order to deliver inputs and give advice to small farmers (owners and sharecroppers) to increase productivity. The programme excludes the landless.

5. From 1974 to-date: Food-for-Work Programme (FFWP), replaced RWP. The FFWP implies a minimum wage, paid mainly in grains, but it does not guarantee employment to poor landless workers or farmers.

6. From 1979 to-date: Grameen (Rural) Banks were established as a pilot programme in a small number of Thanas to extend low-interest loans to the landless poor to buy non-land assets (livestock, etc.) for new incomes. The borrowers are required to repay these loans in weekly instalments and contribute to a common fund to be used in emergency for the poor in a village.

7. In 1982-83: Land Reform measures were debated and acts were passed to lower land ceiling and rearrange the tenancy contracts, but they have not been undertaken so far.

Appendix II

ANTI-POVERTY PROGRAMMES IN INDIA

1. In 1948-1950: Land Reforms were introduced in several states to (a) abolish the intermediary fights in land, (b) lower the land ceiling, and (c) change the tenancy laws to give more secure rights to tenants and sharecroppers.

2. In the 1950s: Community Development Programme (CDP) was undertaken in every state. The country was divided into Blocks (with 80 to 100 villages) and the focus was on (a) building local infrastructure, (b) providing agricultural extension, (c) encouraging rural/cottage industry, and (d) establishing community centres. At the village level, the village Panchayats (elected assemblies) were to work with government departments. CDP was premised on the idea of building a harmonious and undifferentiated community in rural areas.

3. In the 1960s: Intensive Agricultural District Programme (IADP) was launched mainly to promote rapid agricultural growth. Its focus was on the distribution of seeds, particularly of high-yielding varieties, fertilizer and the extension of other services to farmers. It was "technocratic" in its approach to rural development without involving the landless.

4. In the 1970s: The realization, that agricultural growth was not adequate, and it was creating inter-personal and inter-regional disparities without making a dent in the level of absolute poverty, led to several new (but some ad hoc) programmes targeted at small farmers and the landless in rural areas:

(a) 1971-1980: Small Farm Development Agency (SFDA); Marginal Farmers and Agricultural Labourers Programme (MFALP).

(b) 1971-1974: Crash Scheme for Rural Employment (CSRE); Pilot Intensive Rural Employment Programme (PIREP).

(c) 1977-1980: Food-for-Work Programme (FFWP). It was restructured in 1980 as National Rural Employment Programme (NREP). The central and state governments pool resources on a 50:50 basis to provide employment to landless workers, and wages are paid in cash and kind (grain). Workers build rural infrastructure, e.g. social forestry, village tanks, rural roads, school buildings, land reclamation, and watercourse improvement.

(d) 1983 to-date: Rural Landless Employment Guarantee Programme (RLEGP) was started based on the model of Employment Guarantee Scheme (EGS) which was functioning in Maharashtra state since 1974. The RLEGP is funded entirely by the central government to guarantee employment for 100 days in a year to at least one member of a landless household or family. This labour is used to build the rural infrastructure as in NREP. There is provision for a minimum wage in cash or kind.

(e) 1979 to-date: The Integrated Rural Development Programme (IRDP) was started in 2,300 Blocks initially and then expanded to over 5,000 Blocks (almost the entire country). It is funded by the central and state governments on a 50:50 basis. IRDP included the idea of Antyodaya, which had been used since 1977 in Rajasthan to lift the poor above the poverty line by providing them loans to buy income-generating assets other than land. IRDP also merged the SFDA and MFALP for small and marginal farmers to focus on their problems through subsidies and loans. IRDP uses a mixture of subsidy (25 to 50 percent) from the state and low-interest loans from the banking system. The aim of IRDP for those below the poverty line is to raise their income above the line on a permanent basis: it is basically a scheme that can provide incremental income through self-employment. The village Panchayats, Block Development Officers (BDOs) and the District Rural Development Agency (DRDA) are expected to play their roles in formulating and executing each programme.

(f) 1979 t0-date: The Training of Youth for Self-Employment (TRYSEM) is an integral part of IRDP, but it focuses on the youth (18 to 35 years old) from the landless poor households. The aim is to create new skills through specific training programmes to be used for self-employment in or around the village.

(g) 1983 to-date: The Development of Women and Children in Rural Areas (DWCRA) was added to IRDP as a sub-programme to let more poor rural women and their children participate in IRDP. The initial experience of IRDP was that no more than 6 percent of its beneficiaries were women, although they constituted 50 percent of the rural labour force.

(h) 1971 to-date: The Minimum Needs Programme (MNP) is simply a name given to allocations in the public sector for education, health care, electrification, water supply, housing assistance, and nutrition in rural areas.

(i) The Drought Prone Area Programme (DPAP) since 1971 and the Desert Development Programme (DDP) since 1978 focus on regional handicaps that rural populations, both cultivators and the landless, face due to deserts and droughts in arid and semi-arid regions of several northern states.

5. In the late 1960s and early 1970s: Land reform measures were introduced in some states (e.g. West Bengal and Kerala) to readjust land ceilings and tenancy contracts. However, in many states no measures have been undertaken despite the recurring rhetoric in the Sixth and Seventh Five Year Plan documents.

Appendix III

ANTI-POVERTY PROGRAMMES IN PAKISTAN

1. In the early 1950s: Tenancy Laws were enacted in the provinces (except Baluchistan) to give security to tenants and redefine tenancy contracts.

2. In the late 1950s: Land reform measures were undertaken to redistribute land and lower the land ceiling on individual holdings; revenue-free estates were resumed by the state.

3. In the 1950s: Village AID programmes were undertaken for community development at the village level through the existing line departments which involved the locally constituted, but not elected, village councils.

4. In the 1960s: Rural Works Programme (RWP) was introduced in 1962 after the pilot programmes in the then East Pakistan were implemented. The "Basic Democracy" provided the support of Union Councils to government departments for the employment of rural labour to build infrastructure.

5. In the early 1970s: Land reforms were introduced to lower land ceiling, redistribute land, and readjust the tenancy contracts. However, the land reforms announced in March 1977 were shelved after the Martial Law in July was imposed.

6. 1972-77: Integrated Rural Development Programme (IRDP) was undertaken to build markaz (centre) around which villages were to be strengthened by using public departments and village representatives through either the cooperative societies or the unelected constituted councils. The aim here was to "develop" rural areas by taking a holistic approach in providing inputs and infrastructure through the public sector, but without adequate representation of the target groups.

7. 1972-77: Peoples Works Programme (PWP) was a new name given to RWP, which was used to provide some employment in rural areas to the landless labour and build physical infrastructure.

8. In the 1980s: Since 1978, the elected (rural Union Councils have been given some resources to continue the RWP, but IRDP has been dismantled. The provincial line departments are engaged as before in providing services and inputs to rural areas without focus on the landless or the rural poor.

9. As part of its "Islamization" policy, the government has introduced two levies on Muslims: zakat (since 1980) as 2.5 percent of wealth during a year, and ushr (since 1983) as 5 to 10 percent of the vlaue of output of agriculture (crops and livestock products) during each season. Zakat is collected on a voluntary basis from those whose wealth position is above a specified level. Ushr is likewise collected from agriculturists whose production level is above a specified value in the year. Non-Muslims and Muslims belonging to the Shia sect are exempted from the payment of zakat and ushr to the state. The revenue from zakat has to be spent as transfer payment to the "eligible" poor and for no other purpose. The zakat councils have been set up at each administrative level, with representation from the elected council members at the local level in rural and urban areas, to distribute the revenue from zakat. The ushr revenue can be spent on transfer payments and welfare projects for the rural poor. These decisions are left to local councils in each province and district. It is important to note that payment of zakat does not reduce one's tax liability, although an agriculturist is exempted from land tax if he is paying the ushr levy.

10. A "Five Point" Programme was launched by the Prime Minister in mid-1986 for four years. One of its major aims is to improve the conditions of life for the poor in rural areas by focusing resources on building physical and social infrastructure to which the poor would have direct access and on distribution of small plots for those without shelter. This programme was given one-half of the budget in the annual development plan for 1986-87. It is expected to get a higher share in the next three years.

REFERENCES

Ahluwalia, M. S. (1985). "Rural Poverty, Agricultural Production and Prices: A Reexamination". In J. W. Mellor and G. M. Desai (eds.), Agricultural Change and Rural Poverty. Baltimore: The Johns Hopkins University Press.

Abroad, Q. K., and Mahabub Hossain (1984). "An Evaluation of Selected Policies and Programmes for the Alleviation of Rural Poverty in Bangladesh". In Rizwanul Islam (ed.), Strategies for Alleviating Poverty in Rural Asia. Bangkok: I.L.O./A.R.T.E.P.

Bandyopadhyay, D. (1986). A Study of Poverty Alleviation in Rural India through Special Employment Creation Programmes. Asian Employment Programme Working Paper. New Delhi: ILO/ARTEP.

Bardhan, P. K. (1984). The Political Economy of Development in India. Oxford: Basil and Blackwell.

Bhatty, I. Z. (1974). "Inequality and Poverty in Rural India". In T. N. Srinivasan and P. K. Bardhan (eds.), Poverty and Income Distribution in India. Calcutta: Statistical Publishing Society.

Cain, Mead (1983). "Landlessness in India and Bangladesh: A Critical Review of National Data Sources". Economic Development and Cultural Change. Vol. 32, No. 1.

Dandekar, V. M. (1986). "Agriculture, Employment and Poverty". Economic and Political Weekly. Vol. 21, Nos. 38/39. pp. A90-A100.

Food and Agricultural Organization (1985). Production Year Book. Rome.

International Labour Organisation (1985). Yearbook of Labour Statistics. Geneva.

Irfan, Mohammad, and Rashid Amjad (1984). "Poverty in Rural Pakistan". In A. R. Khan and Eddy Lee (eds.), Poverty in Asia. Bangkok: ILO/ARTEP.

Islam, Rizwanul (ed.) (1985). Strategies for Alleviating Rural Poverty in Rural Asia. Bangkok: ILO/ARTEP'

Khan, Mahmood H. (1987). "Pakistan's Agriculture Sector: Performance and Policies, 1971-1985". Research Paper for the World Bank. (Unpublished)

Mohammad, Faiz, and Ghulam Badar (1985). "Structure of Rural Income in Pakistan: Some Preliminary Estimates". Pakistan Development Review. Vol. XXIV, Nos. 3&4. Autumn-Winter.

Osmani, S. R., and R. Rahman (1981). A Study of Income Distribution in Bangladesh. Dhaka: Institute of Development Studies.

Singh, Inderjit (1983). "The Landless Poor in South Asia". In Growth and Equity in Agricultural Development. Proceedings of the 18th International Conference of Agricultural Economists. London: Gower.

UNESCAP (1984). Statistical Year Book for Asia and the Pacific. Bangkok: ESCAP.

World Bank (1986). World Development Report. New York: Oxford University Press.

MAHMOOD HASAN KHAN, Dr Khan is Professor of Economics at Simon Fraser University, Burnaby, B. C. (Canada). His participation in the PSDE meeting was financed by the Friedrich Ebert Foundation of West Germany.

Table 1
Basic Data on Economies of Bangladesh, India and Pakistan

Economic Indicator Years Bangladesh

1. GNP per Person (US$) 1985 140
2. Population 1965 58
 1985 101
3. Percent Rural Population 1965 94
 1984 82
4. Percent Agricultural Population 1965 84
 1984 71
5. Percent Growth of Population 1965-73 2.6
 (Annual Rate) 1973-84 2.5
6. Percent Growth of GDP 1965-73 --
 (Annual Rate) 1973-84 5.0
7. Percent Growth of Agricultural 1965-73 0.4
 Output (Annual Rate) 1973-84 3.1
8. Percent Share of Agriculture 1965 53
 in GDP 1984 48
9. Percent of Labour Working in 1965 84
 Agriculture 1984 75
10 Index of Agricultural Output 1974 98
 per Person (1979-81 = 100) 1985 96
11. Index of Food Production 1974 98
 per Person (1979-81 = 100) 1985 96
12. Cultivated Area per Person 1985 0.13
 in Agriculture (Hectare)

Economic Indicator India Pakistan

1. GNP per Person (US$) 280 400
2. Population 495 57
 759 95
3. Percent Rural Population 81 76
 75 71
4. Percent Agricultural Population 73 60
 69 52
5. Percent Growth of Population 2.3 3.1
 (Annual Rate) 2.3 2.9
6. Percent Growth of GDP 3.9 5.4
 (Annual Rate) 4.1 5.6
7. Percent Growth of Agricultural 3.7 4.7
 Output (Annual Rate) 2.3 3.0
8. Percent Share of Agriculture 47 40
 in GDP 35 24
9. Percent of Labour Working in 73 60
 Agriculture 70 55
10 Index of Agricultural Output 91 98
 per Person (1979-81 = 100) 109 102
11. Index of Food Production 90 97
 per Person (1979-81 = 100) 109 100
12. Cultivated Area per Person 0.32 0.39
 in Agriculture (Hectare)

Note: Data are compiled from World Bank (1986); F.A.0. (1985) and
I.L.O. (1985).

Table 2
Basic Data on External Economic Conditions of Bangladesh, India and
Pakistan

Economic Indicator Years Bangladesh

1. Exports as Percent of GDP 1984 7.6
2. Imports as Percent of GDP 1984 16.6
3. Exports as Percent of Imports 1984 45.7
4. Current Account Deficit as
 Percent of GDP 1984 4.2
5. Terms of Trade (1980 = 100) 1984 106.0
6. Percent Growth of Exports 1965-73 -6.6
 (Annual Rate) 1973-84 2.9
7. Percent Growth of Imports 1965-73 -8.3
 (Annual Rate) 1973-84 4.2
8. Workers Remittances:
 As Percent of GDP 1984 3.5
 As Percent of Exports 1984 46.7
9. Total External Liability
 As Percent of GDP 1984 45.8
10. Total Long-term Debt
 As Percent of GNP 1984 40.0
11. Debt Service Payments:
 As Percent of GNP 1984 1.3
 As Percent of Exports 1984 14.2

Economic Indicator India Pakistan

1. Exports as Percent of GDP 5.8 9.3
2. Imports as Percent of GDP 9.2 21.1
3. Exports as Percent of Imports 62.9 44.1
4. Current Account Deficit as
 Percent of GDP 1.5 4.0
5. Terms of Trade (1980 = 100) 107.0 88.0
6. Percent Growth of Exports 2.4 3.7
 (Annual Rate) 3.3 7.4
7. Percent Growth of Imports -5.7 -2.9
 (Annual Rate) 5.4 7.5
8. Workers Remittances:
 As Percent of GDP 1.6 9.3
 As Percent of Exports 28.2 99.0
9. Total External Liability
 As Percent of GDP 18.9 42.0
10. Total Long-term Debt
 As Percent of GNP 13.6 29.7
11. Debt Service Payments:
 As Percent of GNP 1.1 2.8
 As Percent of Exports 13.8 27.1

Note: Data are compiled from World Bank (1986).

Table 3
Social Indicators for Bangladesh, India and Pakistan

 Social Indicator Years Bangladesh India Pakistan

 1. Life Expectancy Male 1965 45 46 46
 at Birth Female 1965 44 44 44
 Male 1984 50 56 52
 Female 1984 51 55 50
 2. Infant Mortality 1965 153 151 150
 Rate 1984 124 90 116
 3. Child Death Rate 1965 24 23 23
 1984 18 11 16
 4. Population per 9010 4860 3160
 Physician 1965 1984 6000 2610 2320
 5. Population per Nurse 1965 -- 6500 9900
 1984 19400 4670 5870
 6. Adult Literacy Rate Male 1970 36 47 30
 Female 1970 12 20 11
 Male 1983 40 55 36
 Female 1983 18 26 15
 7. Primary School Male 1965 67 89 59
 Enrolment Female 1965 31 57 20
 Male 1983 67 100 63
 Female 1983 55 68 33
 8. Secondary School 1965 13 27 12
 Enrolment 1983 19 34 16
 9. Daily Calorie Intake 1964-66 2001 1967 1757
 per Person 1981-83 1878 2088 2236
10. Daily Protein Intake 1964-66 43.6 49.1 49.8
 per Person 1981-83 40.3 50.8 58.6
 (in Grams)

Note: Data are compiled from F.A.O. (1985); World Bank (1986) and
UNESCAP (1984).

Table 4
Rural Poverty in Bangladesh, India and Pakistan

Countries Percent of Rural Population Below Poverty Line

 1963-64 1972-73 1977-78 1978-79 1983-84

Bangladesh 52.0 55.7 67.9 -- --

India
 a. 53.8 50.5 51.2 53.3 40.4
 b. 44.5 43.1 39.1 -- --
 c. -- 46.0 51.9 -- 48.9

Pakistan 40.5 51.5 -- 39.8 --

Notes: (1.) Data for Bangladesh are from Q.K. Ahmad and Mahabub
Hossain (1984) and Rizwanul Islam (1985).

(2.) Data for India are from (a) D. Bandyopadhyay (1986), (b) M. S.
Ahluwalia (1985) and (c) V. M. Dandekar (1986).

(3.) Data for Pakistan are from M. Irfan and R. Amjad (1984).

Table 5
Distribution of Landownership in Bangladesh, India and Pakistan

Size of Landholdings Percent of
 (in hectare)
 Households Population Area

1. Bangladesh: 1978.
 Up to 0.4 46.6 39.6 8.4
 Over 0.4 to 1.2 31.3 31.2 25.2
 Over 1.2 to 2.0 11.1 13.1 18.6
 Over 2.0 to 4.0 7.7 10.3 22.7
 Over 4.0 to 6.0 1.7 3.1 10.1
 Over 6.0 1.6 2.7 15.0

2. India: 1961-62, 1971-72

 Percent of Households Percent of Area

 1961-62 1971-72 1961-62 1971-72

 Less than 0.4 36.8 44.0 1.6 1.6
 0.4 to Less than 2.0 35.1 33.8 18.4 22.9
 2.0 to Less than 6.0 20.3 16.7 34.5 36.1
 6.0 to Less than 20.0 7.1 5.0 34.4 31.1
 20.0 and Over 0.7 0.4 11.1 8.3

3. Pakistan: 1976

 Percent of Population Percent of Area

 Up to 0.4 30.3 3.3
 Over 0.4 to 2.0 36.2 14.3
 Over 2.0 to 5.0 17.9 17.9
 Over 5.0 to 10.0 10.1 21.1
 Over 10.0 to 20.0 3.2 14.5
 Over 20.0 2.3 28.9

Notes: (1.) Bangladesh: The data exclude the landless. Their
proportion was (a) 28.8 percent in the households and (b) 23.4
percent in the population. The data are from Bangladesh Land
Occupancy Survey, 1978.

(2.) India: The data exclude the landless. Their proportion in the
households was (a) 27.5 percent in 1961-62 and (b) 25.6 percent in
1971-72. The data are from National Sample Surveys, Rounds 17 and 26.

(3.) Pakistan: These data exclude the landless. The data for
landowners and their area are from Federal Land Commission of
Pakistan. They are based on land records of 1976.

Table 6
Distribution of Farms and Farm Area in Bangladesh, India and Pakistan

Size of Farm (in hectare) Percent of Farms Percent of Farm Area

1. Bangladesh 1960 1977 1960 1977

 Up to 0.4 24.3 15.9 3.2 2.7
 Over 0.4 to 1.0 27.3 33.8 12.9 16.3
 Over 1.0 to 3.0 37.7 40.9 45.7 49.4
 Over 3.0 10.7 9.4 38.1 32.7

2. India 1971 1981 1971 1981

 Less than 1.0 50.0 56.5 9.0 12.2
 1.0 to Less than 2.0 18.9 18.0 11.9 14.1
 2.0 to Less than 4.0 15.0 14.0 18.5 21.2
 4.0 to Less than 10.0 11.2 9.1 29.7 29.7
 10.0 and Over 3.9 2.4 30.9 22.8

3. Pakistan 1972 1980 1972 1980

 Less than 1.0 14.0 17.0 1.0 1.0
 1.0 to Less than 2.0 14.0 17.0 4.0 5.0
 2.0 to Less than 5.0 39.0 40.0 25.0 28.0
 5.0 to Less than 10.0 21.0 17.0 27.0 25.0
 10.0 and Over 11.0 8.0 43.0 42.0

Notes: (1.) Bangladesh: Average size of farm (a) 1.43 hectare in
1960 and (b) 1.42 hectare in 1977. Data are from Census of
Agriculture 1960 and 1977.

(2.) India: Average size of farm (a) 2.28 hectare in 1971 and (b)
1.82 hectare in 1981. Data are from Census of Agriculture 1971 and 1981.

(3.) Pakistan: Average size of farm (a) 5.3 hectare in 1972 and (b)
4.7 hectare in 1980. Data are from Census of Agriculture 1972 and 1980.

Table 7
Land and Rural Income Concentration in Bangladesh, India and Pakistan

 Concentration of Land use

Countries 1960 1971 1972 1977 1980 1981

1. Land Concentration
 Bangladesh 0.47 -- -- 0.41 -- --
 India -- 0.62 -- -- -- 0.60
 Pakistan -- -- 0.50 -- 0.54 --

2. Rural Income
 Concentration

 1960 1962 1964 1969 1972 1974

 Bangladesh -- -- 0.34 -- -- 0.34
 India -- 0.41 -- 0.43 -- --
 Pakistan 0.35 -- -- -- 0.29 --

 Concentration of
 Ownership

Countries 1972 1976 1978

1. Land Concentration
 Bangladesh -- -- 0.58
 India 0.68 -- --
 Pakistan -- 0.68 --

2. Rural Income
 Concentration

 1976 1977 1980

 Bangladesh -- 0.40 --
 India 0.39 -- --
 Pakistan -- -- 0.36

Notes: (1.) Land concentration ratios are the Gini coefficients
estimated from data on land distribution by ownership and use given
in earlier tables.

(2.) Data on rural income concentration ratios (Gini coefficients)
are from: (a) S.R. Osmani and A. Rahman (1981), (b) I. Z. Bhatty
(1974), (c) P. K. Bardhan (1984), and (d) Faiz Mohammad and Ghulam
Badar (1985).

Table 8
Distribution of Landless Households in Bangladesh, India and Pakistan

 Percent of Rural Households

Landlessness Based on Ownership/
 Operation or use Bangladesh India Pakistan

1. Ownership Status 1978 1972 1980

 No Land Owned 28.8 25.6 26.2
 Owned Less than 0.4 Hectare 33.2 19.2 32.0
 Total Landless 62.0 44.8 58.2

2. Operational Status 1980 1980 1980

 Landless Rural Labour 20.9 16.0 19.8
 Operating 0.4 Hectare (NL) 7.7 15.0 2.0
 Operating 0.4 to 1.0 Hectare (MF) 16.1 18.1 5.0
 Operating 1.0 to 2.0 Hectare (SF) 14.0 17.0 6.9
 Other Rural Households 42.3 33.9 66.3
 All Rural Households 100.0 100.0 100.0
 (14.3) (93.6) (10.1)

Notes: (1.) Data on landlessness based on landownership are from Mead
Cain (1983) and Mahmood Hasan Khan (1987).

(2.) Data on landlessness based on operation or use of land are from
Inderjit Singh (1983).

(3.) Figures in parentheses are numbers of rural households in
millions.

Table 9
Rural Labour Force in Bangladesh, India and Pakistan
(in millions)

 Bangladesh India

 1972 1978 1972 1978

Rural Population 65.0 78.0 455.0 493.0

Rural Labour Force 16.0 24.0 200.0 218.0

Proportion of Rural Labour:
 a. Working on Own Farm 7.7 11.5 50.0 54.5
 b. Working as Helpers 4.3 6.5 46.0 50.1
 c. Working as Labourers 4.0 6.0 78.0 85.0
 d. Working outside Farms -- 26.0 28.4

 Pakistan

 1972 1978

Rural Population 48.0 59.0

Rural Labour Force 11.5 16.0

Proportion of Rural Labour:
 a. Working on Own Farm 3.8 5.3
 b. Working as Helpers 3.5 4.8
 c. Working as Labourers 1.0 1.4
 d. Working outside Farms 3.2 4.5

Notes: (1.) Data for labour working outside agriculture in Bangladesh
were not available.

(2.) All data are compiled from Inderjit Singh, op. cit., (Table 8).

Table 10
Index Numbers of Real Agricultural Wage in Bangladesh, India and
Pakistan

Years Bangladesh India Pakistan

1973 115.1 118.1 138.5
1974 123.2 99.2 144.3
1975 100.0 100.0 108.8
1976 108.1 128.1 100.0
1977 110.1 118.9 105.2
1978 119.5 123.1 108.3
1979 130.7 126.9 115.0
1980 122.5 131.1 121.8
1981 125.1 119.7 158.8
1982 124.0 126.7 147.7
1983 112.0 127.5 176.7
1984 100.3 133.1 186.2

Note: The indices of real wage rates have been constructed by
deflating the average daily wage rates for agricultural labourers
by the consumer price indices. Data on nominal wage rates and
consumer price indices are from I.L.O. (1985), UNESCAP (1984).

Table 11
Sectoral Shares of Plan Allocations in Bangladesh, India and Pakistan

 Percent Share of Sector
 in the Plan

 India
 Bangladesh
 Second Sixth Seventh
 Sector Plan Plan Plan
 1980-85 1980-85 1985-90

1. Agriculture, Irrigation
 and Rural Devel-
 opment 35 24 22

2. Industry and
 Mining 13 16 13

3. Energy (Power and
 Fuels) 19 28 30

4. Transport and
 Communications 13 13 13

5. Social Services 17 14 16

6. Others 4 5 6

 Percent Share of Sector in the Plan

 Pakistan

 Fifth Sixth
 Sector Plan Plan
 1978-83 1983-88

1. Agriculture, Irrigation
 and Rural Devel-
 opment 20 16

2. Industry and
 Mining 17 9

3. Energy (Power and
 Fuels) 25 38

4. Transport and
 Communications 23 19

5. Social Services 13 17

6. Others 2 2

Notes: (1.) Social Services include physical planning and housing,
health care, education, manpower, population planning, and social
welfare.

(2.) Data are from the Five Year Plan documents of the three
countries.