17 February 2013

Uncounted Babies and Total Fertility Rate's Incomplete Picture of Procreation Statistics

Singapore is a rather unique country in many ways.

One unique aspect is the significant levels of cross-border population inflow.  It grants of tens of thousands of permanent residency a year.  Significant numbers of residents marry non-residents.

In the light of these, does the total fertility rate truly reflect the country's fertility rate?

Total Fertility Rate Defined
Total fertility rate is defined as the average number of live births each woman would have during her reproductive years if she were to experience the age-specific fertility rates prevailing during the period.[1]  It is expressed as number of children per woman.

TFR takes the number of children born to women in a given year to represent the number of children they will give birth to by the end of their reproductive years.[2]

In other words, all the women in a society (whether single, married or divorced) are considered by the statisticians to go through their child-bearing years and give birth according to the composite age-specific fertility rates in a given year.  Whether or not every such woman actually gives birth in that year, or have given birth in earlier years, or will give birth in later years, or will never give birth at all is immaterial.

Despite its synthetic nature, TFR is used worldwide and there is no better way to measure a society's fertility rate at any given point in time.  Statisticians can (and do) determine the number of children that the average woman has at the end of her reproductive years, but policy planners can't wait.

For a more detailed discussion of TFR, see my earlier article [here].

In this article, TFR refers to resident TFR, the TFR pertaining to resident women, which is the official definition.

New residents
TFR is based on the number of babies born during a specific year to resident mothers.

If a woman gives birth to children before she is granted residency in, for example, 2012, these children are not, and cannot, be included in TFR for 2012.

It is possible, though unlikely, for statisticians to retroactively adjust the TFR for the years that the immigrant children were born in, but this would require adjusting the number of both babies and resident mothers (the numerator and denominator, respectively, in the TFR formula), and perhaps ancillary data too, or else they cannot be reconciled.

The impact on TFR may not be significant if the number of people granted residency yearly is small vis-à-vis the total population, but Singapore granted residency to about 270,000 individuals between 2007 and 2011 (excluding double counting the cases in which citizenship was granted to permanent residents).[3]  In comparison, the resident population below the age of 50 years was about 2.6 million in June 2012.[1]

Overseas Singaporeans
Singapore uses the de jure concept for population statistics based on a person's usual residence.  A de jure population comprises all persons present at their place of usual residence, as well as those who may be temporarily absent from their place of usual residence.  The place of usual residence is where a person usually resides, and it may or may not be the person's place of domicile or permanent residence.  Residents are excluded if they have been away from Singapore for a continuous period of 12 months or longer as at the reference period.[1]

Singaporeans living overseas are not part of Singapore's de jure population, and babies born to Singaporeans living overseas are not included in TFR.

There were about 200,000 overseas Singaporeans as at June 2012.  Of these, 9,200 were below the age of 5 years.  This number had increased steadily from 7,600 in 2008.  It is not clear, however, how many of young citizens were born overseas (and therefore not included in the TFR of the respective years in which they were born).[3]

Separately, about 2,000 children were granted citizenship upon registration by their parents last year.[4]  Presumably, each of these children was born outside Singapore and at least one of his/her parents is a citizen.

New Residents and Returning Singaporeans Born Overseas (Combined)
Let's estimate the impact of new residents bringing in already-born children and Singaporeans returning from overseas.

In 2010, 18,358 residents below the age of 5 years and 29,844 residents aged between 5 years and 9 years were born outside Singapore.[5]

Consider the residents aged between 5 years and 9 years in 2010.  There were 215,700 of them in 2010.  Five years earlier, in 2005, when they were aged less than 5 years, there were 199,500 of them.[1]  The size of this cohort increased 16,200 despite mortality and emigration.

Non-Resident Mothers
TFR is based on the number of babies born during a specific year to resident mothers.

A citizen or permanent resident baby is not included in TFR if the mother is not a resident.

There were 25,344 resident marriages (i.e., at least one spouse is a citizen or permanent resident) in 2011.  Of these, 5,490 were between a non-resident bride with a citizen groom and 1,066 were between a non-resident bride with a permanent resident groom.[3]  That is, 25.9 per cent of resident marriages were between non-resident brides and resident grooms.

If any of these non-resident brides are not granted residency by the time their children are born, their babies are not included in TFR (and neither would the mothers).

In the light of the significant number of individuals granted permanent residency annually, the significant number of overseas Singaporeans and the significant percentage of marriages between residents and non-residents, Singapore's TFR possibly under-estimates the country's "true" fertility rate.

While Singapore cannot unilaterally redefine TFR, Department of Statistics can, and should, develop additional measures of fertility, taking into account the country's significant levels of cross-border people movement.

For example, the US Department of Labor publishes six measures of unemployment, with U-3 being the official (and generally internationally recognised) measure and U-6 being the broadest.[6]

The broadest measure of Singapore's fertility rate will still be below 2.1, but at least it will not be as dismal as 1.2.  It will also be more realistic.  And, it will help policy planners make better decisions.


1. DEPARTMENT OF STATISTICS Population Trends 2012.

2. TFR = (B15-19/W15-19 + B20-24/W20-24 + B25-29/W25-29 + . . . + B45-49/W45-49) x 5

where, for example, B15-19 is the number of babies born to mothers aged between 15 years and 19 years and W15-19 is the number of women in the general population aged between 15 years and 19 years in the year for which TFR is computed.  Babies born to mothers younger than 15 years or older than 49 years are added to the nearest age group.  The total is multiplied by 5 because each sub-cohort spans five years.

3. NATIONAL POPULATION AND TALENT DEPARTMENT Population in Brief 2012 and previous years 2008, 2009, 2010 and 2011.

4. 4 in 10 S’poreans Married Foreigners in 2012 TODAY 30 Jan 2013.

5. DEPARTMENT OF STATISTICS Census of Population 2010.  Information about residents born outside Singapore was found in this publication only.

6. U-6 is defined as total unemployed, plus all persons marginally attached to the labour force, plus total employed part time for economic reasons, as a percent of the civilian labour force plus all persons marginally attached to the labour force.  See BUREAU OF LABOR STATISTICS The Employment Situation 2013.

1 comment:

  1. I doubt that even if the figs showed the TFR is higher than the govt claims that the govt will change its mind on the 7million.

    They have set their hearts on AT LEAST that number by 2030. Anything else they say should be ignored, simply because we've heard the promises about control on the number of foreigners entering the country, that the popn figures voiced are for way down the road - and not one of them has been kept.