Well, after my last post I finally got around to calculating another measure of the birth rate – births per capita. Total births are a useful aggregate but the measure does not adjust for population size. There are other fertility measures out there such as births per woman aged 15 to 44 years but per capita births is a nice intuitive measure that adjusts for population size and can be readily compared to other numbers – such as per capita income. Moreover, it was easy to calculate.
When per capita births are considered for Canada (Figure 1), the Depression era sees a drop in births and is then followed by a rising birth rate - a baby boom. (The data sources are described in my previous post). However, when per capita births are used, the birth rate starts to rise in 1937 and peaks in the mid-1940s and then starts to decline. Interestingly enough, even this series shows a slight rebound in the birth rate since about the year 2000.
Figure 2 plots per capita births against real per capita GNP/GDP (in 1981 dollars) for the period 1921 to 2010 and then takes a LOWESS smooth of the relationship. There is a bit of an inverted J-curve here. As per capita income increases, per capita births at first increase and then begin to drop sharply. The overall relationship is negative and babies eventually become an inferior good if income rises high enough. Babies in Canada became an inferior good right about the point when real per capita GNP in 1981 dollars hit about $5,000 annually which happened in the 1940s.
Indeed, a log-log linear regression of per capita births on real per capita GNP gets you a coefficient estimate of -0.45, which is negative income elasticity. A 1 percent increase in real per capita income causes per capita births to drop by almost half of one percent. The linear regression estimate (t-statistics in brackets) was:
Log of Per Capita Births = 0.133(0.50) - 0.453(-15.65)*Log of Real Per Capita GNP
and with an adjusted r-squared of 0.73. Given that real per capita income in this regression explains almost three quarters of the variation in per capita births and that the post war baby boom is usually seen as starting in the late 1940s - about the same time having children becomes an inferior good - I think that the post-war baby boom is the residual in this regression equation. Or, we may need to rethink when the baby boom actually occurred. On a per capita birth basis, the birth rate satarts rising in 1937 so the baby boom may have been underway a decade earlier than usually ascribed. This first phase was a response to the front end of the J-curve - a positive relationship between income and children. Rising income after the abyss of the Depression led to people having more children. The second phase of the boom after 1947 actually coincided with a negative income elasticity of demand for children and therefore can be viewed as driven by extraordinary factors that countered the effect of a rising income.