Drivers of the rural economy

Since 1988, I have been generating and maintaining a longitudinal data set on the operation of the rural economy in Bangladesh through repeated household surveys in 62 randomly selected villages. The samples were selected using a multi-stage random sampling method at the union, village and household levels. The latest survey was conducted in 2013 covering all households in the villages. I would like to share the insights gained on the transformation of the rural economy over the last 25 years from the analysis of this data.

The most commendable progress is a dramatic demographic transition that has been a major factor behind the socio-economic development in rural areas. The number of children below five years of age relative to women of reproductive age, which is a measure of recent fertility, declined from 67 per thousand women in 1988 to 49 in 2000, and further to 36 in 2013.  The number of children living with parents declined from 2.9 in 1988 to 2.5 in 2000, and further to 1.8 in 2013. As a result, the size of the household went down from 5.9 to 4.2 members over the last 25 years. This means that the per capita income has increased much faster than household incomes in rural areas.

The data also shows that the number of households increased by 1.5 percent every year due to the formation of new households as married sons separate and become parents themselves. Also, about 1.1 percent of the people migrate every year to towns and cities in search of better economic opportunities. Due to the reduction in household size and the out-migration of family members and households, the rural population grew by only 0.6 percent annually between 1988 and 2000, and by 0.3 percent between 2000 and 2013.

As a result of the demographic transition, the dependency ratio (number of consumers per earning member) has declined substantially. The proportion of children up to 15 years of age declined from 48 percent to 36 percent between 1988 and 2013, while the proportion of population of working age (16 to 59 years of age) increased from 48 percent to 58 percent. Women's participation in income-earning activities has also increased, particularly between 2000 and 2013; thus, the growth of the rural labour force has been faster than that of the population. The proportion of earning members in the total population increased from 30 percent in 1988 to 37 percent in 2013. Fewer children and more earning members imply that the households can generate savings in food and healthcare expenses for investment in economic activities, education of children, improved housing, and so on. 

Human resource is the most abundant resource in Bangladesh. However, the quality of human resource is very poor. In 1988, 46 percent of adult men and 72 percent of adult women did not have any formal schooling. If we include those who did not pass primary schools, the number was 62 percent and 83 percent respectively for men and women.

The school participation rate of children has, however, improved substantially. The proportion of children of primary school age (6 to 10 years of age) attending schools increased from 66 percent in 1988 to 97 percent for boys and 55 percent to 98 percent for girls in 2013.  For the secondary age group (11 to 15 years of age), tremendous progress has been achieved in the school participation for girls – from 47 percent participation in 1988 to 91 percent in 2013. The progress has been relatively slow for boys due to early dropout from schools – from 58 percent participation in 1988 to 80 percent in 2013. The number also shows that more girls than boys now attend secondary schools. The dropout starts from age 11 for boys (to support the family, to augment income) and age 13 for girls (for marriage). The rate of dropout for the adolescents between the ages of 12 to 15 years was 58 percent in 1988, 24 percent in 2000, and only 9 percent in 2013.

The impact of the progress made in school participation will take some time to have full impact on the quality of the labour force since those who did not have schooling earlier are still in the labour force. In 2013, the proportion of the adult population (aged above 15 years) that did not attend school or have below primary education (basically illiterate) was 62 percent for men and 64 percent for women.

The educational attainment of the workers shows that the better educated are employed in services and business while the less educated are employed as wage-labourers and farmers (only if they own land). The average years of schooling are lower for workers providing agricultural wage-labour than for those who are engaged in rural transport, construction and processing activities. Households earning their livelihood from services exhibit working members with the highest level of education, followed by those dependent on trade, business and farming.

Agriculture was found to be the main source of livelihood for rural inhabitants. Land and water is the natural resource base for agriculture. The data obtained from the surveys shows that the amount of land owned per household has eroded substantially over time, despite rapid rural to urban migration of the population. The average size of land owned per household has declined from 1.50 acres in 1988 to 1.31 acres in 2000, and further to 0.75 acre in 2013. Percentage of households that are completely landless or have only homestead land has declined from 35 percent in 2008 to 41 percent in 2013.

In Bangladesh, we define “functionally landless” households as those who own less than 0.5 acre, in the sense that this size of land cannot be a significant source of income for the household. The proportion of these households was 47 percent in 1988 and the number increased to 52 percent in 2000, and further to 67 percent in 2013. On the other end, the proportion of households who own more than five acres of land declined substantially from 8.2 percent in 1988 to 5.1 percent in 2000, and further to only 2.2 percent in 2013.

Since land is such a precious asset, hardly any land transactions take place through purchases and sales, unless the seller is in distress. Even households that migrate to urban areas do not sell their land. Substantial land transactions, however, take place through the operation of the tenancy market. The sharecropping system, under which the harvest and certain input costs are shared between the landowner and the tenant, was the predominant tenancy arrangement till the 1980s.  The data obtained from the surveys however shows that lease arrangements (rights of cultivation for a number of years against advanced cash payments) and fixed-rent tenancy have gained importance with the spread of the cultivation of high-yielding varieties of rice, maize, potato and vegetables. The change in tenancy arrangements has provided incentive to tenants to use the rented land more intensively during the year and to maximise the productivity of land in individual crops.

The tenancy market transfers land from the absentee and relatively inefficient large landowners to the landless and marginal land-owning households. Thus, it has had a positive effect on the redistribution of land and an increase in land productivity. The data shows that the tenancy market is widespread and has grown over time. The proportion of farm households that rent land to make their holdings larger and more viable increased from 44 percent in 1988 to 54 percent in 2000, and further to 58 percent in 2013. The proportion of cultivated holdings that are rented increased from only 23 percent in 1988 to 33 percent in 2000, and further to 42 percent in 2013. Most of the tenants are landless or marginal landowners who find it economical to rent land to increase the capacity utilisation of the farm establishment. The tenancy arrangement is preferred over labour arrangements because it is socially prestigious to self-employ family labour on rented holdings than to work as wage-labourers on another's farm.

Despite larger availability of land from the tenancy market, the dominance of marginal and small farms in the agrarian structure has been growing. The proportion of non-farming households in rural areas has grown from 34 percent in 1988 to 46 percent in 2013. Amongst farming households, 36 percent operated land in sizes of less than 1 acre in 1988. This number was 49 percent in 2000 and 62 percent in 2013. At the other end, only 10 percent of farming households operated land in sizes of more than 5 acres in 1988. The proportion of these relatively large farms declined to 3.6 percent in 2000 and further to 1.9 percent in 2013. Thus, marginal and small farmers now dominate the agrarian structure. There are hardly any large farms in Bangladesh today. The average size of farm holdings has almost halved, from 2.2 acres in 1988 to 1.2 acres in 2013.

Although the size of cultivated holdings continues to decline, the land productivity has improved due to the rapid growth of investment in irrigation and the adoption of farm mechanisation. With the government adopting the policy on import liberalisation of agricultural machines in the late 1980s, private investment in small-scale irrigation equipment (i.e. shallow tube wells and power pumps) has vastly expanded irrigation infrastructure. Cultivated land with access to irrigation increased from 24 percent in 1988 to 62 percent in 2000, and further to 84 percent in 2013. The expansion of irrigation infrastructure has facilitated the growth of a number of crops on the same parcel of land over the year as well as the adoption of higher yielding crop varieties. The small size of the farms and the low education of farmers did not constrain the adoption and diffusion of improved technology in agriculture.

The improvement of human capital has facilitated occupational mobility of workers from lower-productive agriculture to higher-productive non-farming jobs. In 1988, 69 percent of earning members were employed in agriculture – 45 percent working as farmers, 3 percent engaged in fisheries and animal farming fulltime (these were secondary and tertiary occupations of many more households), and 21 percent providing wage-labour on others' farms. By 2013, the share of agricultural work in overall employment came down to 47 percent – 31 percent in farming and 14 percent in agricultural wage-labour. Non-farming activities such as business, service, agricultural processing, transport operation, construction labour, etc., have become the major source of rural employment. The proportion of workers employed in business was only 9.4 percent in 1988; it increased to 15 percent in 2000. The expansion of this activity has been relatively slow since then, with people relocating businesses to towns and cities and using the investible surplus for financing overseas migration. In 2013, 19 percent of rural workers were engaged in business activities. The largest increase in employment has been in industrial labour (i.e. working in brick fields and in agro-processing), rickshaw and van transport, and road and house construction. The proportion of workers employed in these activities has increased dramatically – from 9.7 percent in 1988 to 26.4 percent in 2013.

The drivers of this mobility from relatively low-earning to higher-earning jobs include the development in rural infrastructure such as roads, bridges and electricity; the increase in marketable surplus from the adoption of improved agricultural technologies; the rapid expansion of microcredit provided by NGOs; and the improvement in the quality of human capital. The mobility in rural occupations has been most pronounced for the land-poor households whose members were initially employed as agricultural wage-labourers. With the availability of jobs in the non-farming economy, the agricultural labour market has become tight, leading to the faster increase in wages than in food prices, particularly since 2000. Thus, it has had positive effects on reduction of poverty.

The analysis of the data shows that the average household income increased from USD 939 in 1988 to USD 1,169 in 2000, indicating a rate of growth of about 1.9 percent per year. The household income further grew to USD 1,518 by 2013, indicating a growth of 2.2 percent per year. The per capita income grew at a faster rate (at 3.0 percent per year between 1988 and 2000, and 3.7 percent between 2000 and 2013) due to the reduction in the number of members per household. The growth has been fastest during the recent period.

Agriculture's contribution to household income was estimated at 58 percent in 1988. This decreased to 38 percent by 2013. The income from rice production and agricultural wage-labour remained almost at the same level over the last 25-year period. The total productivity in agriculture has, however, increased. Maintaining the same level of agricultural income with substantial reductions in the size of farms and the number of workers indicates progress. The credit goes to the technological progress in agriculture.

Most of the growth in rural income came from the expansion in the rural non-farming economy.  The share of non-agricultural labour in total household income has grown from 42 percent in 1988 to 62 percent in 2013.  The fastest growing source of income was remittances received from members working abroad (service) - from 5 percent in 1988 to 21 percent in 2013. The share of business in total income increased sharply between 1988 and 2000, but it has receded during the recent period. The share of non-agricultural labour in total income also increased dramatically - from 5 percent in 1988 to 16 percent in 2013.

What has happened to the distribution of rural income? Has the expansion of the non-farming economy moderated the concentration of income from land that is highly unequally distributed in Bangladesh? The analysis of the data shows that income inequality has worsened marginally over the observed period. The bottom share of 40 percent of the households in the rank of per capita income has remained almost at the same level (16 percent), but the income share of the top 10 percent of the households has increased from 33 percent to 35 percent. The coefficient of the concentration of income has grown from 0.41 to 0.44. The income from business and remittances are more unequally distributed than the income from farming. The worsening of income inequality is due to the increase in the share of the income sources, which are more unequally distributed.

According to the self-perception of the respondents, their economic condition has improved over time. In the 2000 survey, 46 percent of the households reported improvement over the period of 1988 to 2000, while 29 percent reported deterioration. The economic condition has reportedly improved further in the most recent period. In the 2013 survey, 56 percent of the households reported improvement in economic conditions during the period of 2000 to 2013, while deterioration was reported by only 13 percent of the households. The main factors behind the improvement were reportedly the increase in earning members relative to dependents, access to land from the tenancy market, migration to cities and abroad, accumulation of capital and its investment in non-farming activities, and accessibility to developed infrastructure. The factors behind deterioration have been reported as the increase in the number of dependents, health shocks such as death and disability of earning members, loss of assets from natural disasters, and expenses for litigation.

In conclusion, commendable progress has been achieved in the rural economy of Bangladesh over the last 25 years. The fertility rate has almost halved, thereby reducing the proportion of children (up to 15 years of age) from 48 percent to 36 percent, and the proportion of the working age population from 48 percent to 58 percent. Almost 97 percent of the primary school age group and about 86 percent of the secondary school age group now attend school. The proportion of workers with no formal schooling has reduced from 46 percent to 34 percent for men, and 72 percent to 38 percent for women. Although the availability of land for agricultural production has declined, the base of growth in land productivity has expanded because of the rapid expansion in the coverage of irrigation and the diffusion of improved agricultural technologies.  Employment in the non-farming economy has increased to 53 percent of the total rural employment, creating a new “class” of people who depend more on physical and human capital than on land and manual labour. About 62 percent of the rural income is now generated in the non-farming economy compared to 41 percent in 1988.

The drivers of the progress have been the rapid technological advancement in agriculture and the generation of marketed surplus for growing urban residents; expansion of access to electricity and development of rural roads, bridges and markets; access of the poor to credit from NGOs for engagement in informal business; improvement in education to facilitate occupational mobility from low-paid to relatively high-paid jobs; increased earnings from remittance sent by members working in cities and abroad; and expansion of the service sector stimulated by the growing economic prosperity.

The writer is Advisor to BRAC and former Director General, BIDS.