There were more than 7 billion people on earth in 2013. While this is the highest number ever, the population growth rate has been steadily declining, in part due to declining fertility rates. Tomorrow, Friday, July 11, is World Population Day, and in this spirit, I'd like to talk about a key component of population growth: fertility rates.
Total fertility rate (TFR): the average number of children per woman
If you're looking for fertility data, a good place to start is total fertility rate (TFR), which is the average number of children born to each woman. In 1960, women worldwide had an average of 5 children. The rate has since halved, and in 2012, women had an average of 2.5 children across all regions.
As the interactive visualization above illustrates, TFR is declining worldwide, but different regions and countries have varying degrees of change. Sub-Saharan Africa, for example, has had a slower decline in fertility rates than other regions. A woman in this area of the world had an average of 5 children in 2012. This is the highest TFR in the world.
Calculating total fertility rates from age specific fertility rates
In most developing countries, TFR is calculated using data from household surveys such as the Demographic and Health Survey (DHS). The survey produces indicators on age-specific fertility rates (ASFR), which show the number of children born per 1,000 women in 5-year age groups. For example, in the table below, a country like Niger sees 287 births per 1,000 women aged 30-34. We can take the "total" of age-specific fertility rates (ASFR) to calculate the "total" fertility rate. To do this, we sum the ASFRs, and because the ASFRs are by 5-year age groups per 1,000 women, we multiply the sum by 5 and divide it by 1,000 to get the TFR.
The table below shows ASFRs and TFRs from Niger's Demographic and Health Survey (DHS). There were 206 births per 1,000 women of ages 15-19, 338 births per 1,000 women of ages 20-24 (and so on) in a given year. To calculate the TFR from these ASFR: (206+338+326+287+221+100+49)*5/1000=7.6, so we can say that the TFR in Niger is 7.6 per woman.
TFR is a period measure, not a cohort measure
A few months ago, I wrote a blog on life expectancy and explained that the "life expectancy at birth" we use is not a mortality pattern that the actual birth cohort born in the year is going to experience, as it is calculated by using age-specific death rates in a given year. In fact, this is the case for TFR, too. TFR is calculated from the ASFRs of a given population (e.g. country) for a given time (e.g. year), and the "population" consists of women in different age groups. Therefore TFR is a period measure and not the fertility pattern that a woman is actually going to experience during her entire reproductive period of 15-49. It is not a cohort measure. We can also say that TFR is an age-standardized measure because it is calculated from the fertility rate by age groups.
But the mean number of children born to women aged 40-49 is a cohort measure
While we're talking about cohort and period measures, DHS counts the number of children ever born to women ages 40-49 as well. This measure is a cohort measure, and is basically the fertility pattern that the actual cohort of women has experienced during their entire (or nearly entire) reproductive age. In countries where the fertility level has been declining, it is higher than the TFR from the same survey. This is because it measures the retrospective fertility level of women ages 40-49 who entered the reproductive age 25-34 years ago, whereas TFR measures the current fertility level of women at ages 15-49 at the time of the survey.
The chart above shows TFR and the mean number of children ever born to women ages 40-49 from DHS for select countries. In countries where the fertility rate decline has been substantial, such as Bangladesh and Nepal, the difference between the cohort measure (mean number of children ever born to women ages 40-49) and period measure (TFR) is large (1.9 and 1.7 respectively). The difference is small in Burundi and Niger (0.2 and 0.3 respectively), where the fertility rate decline has been minimal.
So the next time you think of fertility rates, you'll know that TFR is a total of ASFR, an age-standardized period measure. And if you're interested in finding out more about why TFRs have changed, have a look at the video below:
And don't forget to follow us on Twitter at @worldbankdata for news and updates on the most widely followed measures of the global economy and our Open Data initiative.
Indicators and codes used in this post:
- Fertility rate, total (births per woman) SP.DYN.TFRT.IN
- Adolescent fertility rate (births per 1,000 women ages 15-19) SP.ADO.TFRT
- Population growth (annual %) SP.POP.GROW
Hi – I've seen before on Hans Rosling's twitter account that there was a drop in fertility following reduction of infant mortality, but I hadn't realise how sharp & pointed that was. Is there an academic paper on this?