The decline in fertility by half in high-income countries during the 20th century brought
population growth to a halt in many of these countries. The decline in fertility by more than half
in low-income countries in 40 years (1965-2005) is not yet comprehensively accounted for,
although demographers are agreed that these trends in fertility are irreversible and the size of the
world’s population will stabilize later in the 21st century. How much does each of these
conceptually distinct factors economists have described explain of this remarkable decline in
fertility? I do not yet find a consensus on how to weight these factors in explaining cohort
fertility. What fraction is due to an exogenous decline in mortality, the decline in the relative
value of child labour, the increase in the value of women’s time used in child care and the related
increase in their empowerment, the increase in returns to schooling children, the greater income
elasticities of demand for child quality than for quantity, and finally the improvements in birth
control technology?
Identifying the effect of fertility on the welfare of families and society
The policy-relevant externalities of fertility could arise at the aggregate level or in terms of
substitution effects within families. Malthus assumed that fertility added to subsequent
generations of workers, which reduced their wages and also changed the age composition of the
population. But empirical evidence for these aggregate effects of fertility has not led to a
consensus on their importance for today’s low-income countries (National Research Council,
1986). At the microeconomic level of the family, fertility is found to be closely associated with
other life-cycle choices by parents, including the share of time women allocate to the market
economy, the investments parents make in the human capital of each of their children, and
perhaps the savings out of income they accumulate in physical capital, possibly for old age
support or precautionary insurance. But to assess the magnitude of these cross-effects of fertility,
researchers must first specify an exogenous factor (not a choice variable within the orbit of the
family) that affects fertility but leaves other constraints on the family life-cycle choices and
outcomes unaffected and is unrelated to parent preferences (Schultz, 2005). In other words, an
exclusion restriction or a valid instrumental variable is needed to account for some part of the
variation in fertility that is independent of parent preferences and family life-cycle economic
constraints. Otherwise, these cross-effects observed at the family level may not be causal and
cannot be expected to occur when population policies reduce (or increase) fertility.