Demography in the driving seat

An important element of analysing livelihood system in any society is its demographic characteristics. A society might be poor in terms of endowments of natural resources but given appropriate policies and their proper implementations, a developed human resource base could show the paths to overcome poverty. Taiwan sets an example where its richness sprang out of poor natural resource base only due to strong human capital. In this article, we shall dwell on demographic dynamics in rural areas of Bangladesh drawing upon a census of roughly 22,000 households spreading over 62 villages in the country. Of course, to make a comparison with the past, we shall use the data from a sample of these households in 1988.
Other things remaining the same, it is assumed that the welfare of a household depends on the number of members it has to feed, and a reduction in the household size signals to a rise in living standard of the members. Although the number of households in rural Bangladesh has increased over time, because of family split and migration, members per household (household size) have drastically declined from 6.15 in 1988 to 4.24 in 2013. Without going into hair-scratching calculus, we can possibly argue that such a reduction in household size would mean a reduction of rice consumption by 1.0 kg per day per household.   At household level, the money so saved on account of rice could be put to education or health. At national level, policy makers could breath a sigh of relief as far as rice is concerned and possibly ponder over emphasising non-rice crops. Field-level information further reveals an increase in the proportion of spouses (a result of selective migration of household heads) and a reduction in the proportion of non-relatives living in the household. The latter could be adduced to a reduction in permanent labour used previously by household to supervise farm work which is now rarely observed in rural households.
We observe that the proportion of infants (aged 0-5) has been gradually declining over time, and so is the case with children aged 6-10. In other words, the proportion of population aged up to 10 years has significantly fallen from about 40 to about 27 per cent between 1988 and 2013.  This implies that the base of the population pyramid has substantially shrunk over time. Suffice it to say that from policy point of view we should now be less concerned with construction of more primary schools as the primary-school-age population has been going down. Rather, it would be effective to divert the resources for the improvement of the quality of primary education.
However, there has been a rise in the ratio of 61 plus population - thanks to improved life expectancy following improved health care. The most important news relates to a fall in total dependency, thus, seemingly enabling households to divert resources to productive pursuits. Finally, the rise in working age population from 49 to 58 per cent during the comparable periods points to the much-talked-about demographic dividend that Bangladesh is faced with. The working age group has been constantly entering into the labour market to strengthen the economic base. Thus, the fall in the proportion of people at the lower end of the population pyramid provides us with a sigh of relief as the dependency rate has been reducing. But, at the same time, increased working-age population carries a tension for policy makers as jobs need to be created for the expanding labour force.
The child-woman ratio - a reflection of the current fertility level - has gone down from 67 per 1000 in 1988 to 36 which shows further success in population control.  However, there is very little room for complacency as fertility rate is still high in absolute sense. Counter-intuitively, the rate of reduction in fertility has been faster for the small land-owning groups. The large difference between the fertility level of large and smaller ones that existed in 1988 narrowed down in 2013 quite substantially although the inverse relation still holds. This could be due to the roles played by NGOs (non-governmental organisations) and government agencies which target the poor segment in terms of education and extension. Interestingly, education of husband and wife does not seem to have much effect on fertility control. For example, female education has improved from 1988 to 2013 but with no impact on fertility reduction. The rate is still higher for those with secondary education. The unexpected outcome could be due to the unfortunate fact that in the patriarchal rural society, the decision about babies generally comes from the males and women's views are rarely respected. It could also be due to the reason that fertility is mainly related to participation of women in income-earning opportunities rather than education per se  that raises opportunity costs of rearing children.
With economic growth and transformation, societies tend to witness break-up of families. Engagements of household members in different economic occupations apparently work behind the break-up. Bangladesh is no exception to this historic truth, not even its rural areas. The incidence of joint families has been going down with the migration of young adults to urban areas and formation of separate families. Large families comprising 6-7 members or more are now seen in TV serials -- possibly for reminiscence of the past. However, with the break-up of families, the proportion of single and two-member families has increased. Disconcertingly, the data show that about 3 in 100 households are now single households compared to less than 1 in 1988. Perhaps needless to mention that most of the single-member families are extreme poor (36 per cent single and 41 per cent two-member) with very low land and non-land asset base. This has immense ramifications for social safety nets.
There are two more important transformations that need to be mentioned. First, 62 village survey shows that roughly one-fifths of the girls aged 15-17 get married and there is no deviation in the trend over time. There are regional variations with, say, Kurigram, Bogra, Thakurgaon etc. witnessing child marriage to the tune of 30-40 per cent. Secondly, the incidence of female-headed households in rural areas has increased substantially from 6.0 to 15 per cent between 1988 and 2013. In Bangladesh context - and possibly everywhere too - female-headed households are generally considered as fragile in socio-economic indicators, and more so in terms of security. But a rise in the share of such households in Bangladesh warrants a different explanation.  This could be the result of migration of male members to urban areas or overseas.  We can also possibly presume that growing 'feminisation' of agriculture in Bangladesh has its roots in widespread migration of male members. Thus, it would be erroneous to conclude that female headed households are necessarily fragile in terms of socio-economic indicators.
Modern technology and other contributory factors apart, this rural transformation owes immensely to the changing demographic dynamic. Hindsight data show that demography has been at the driving seat of rural transformation in Bangladesh, and reinforcing the observation of Michael Lipton that fertility is one of the three drivers of rural transformation in Bangladesh.
Abdul Bayes is Professor of Economics, Jahnagirnagar University, Mahabub Hossain is Distinguished Professor of BRAC University and ANM Mahfuzur Rahman is Programme Coordinator, Data Management (RED, BRAC).
 [email protected]