Fertility rate absent in data

Cape Town-111010- Census 2011 is in full swing, as officials go door to door collecting information at every house-hold big or small. Census official; Sherwin Latchman interviews Johan Pool from the Panarama area. Report:Aziz, Photo:Ross Jansen

Cape Town-111010- Census 2011 is in full swing, as officials go door to door collecting information at every house-hold big or small. Census official; Sherwin Latchman interviews Johan Pool from the Panarama area. Report:Aziz, Photo:Ross Jansen

Published Nov 4, 2012

Share

Johannesburg - A striking feature of the census data released last week is the absence of an estimate of the country’s fertility rate. This is unfortunate given that Census 2011 showed a sudden surge in population growth – a finding that has proved controversial.

Azar Jammine, the chief economist at Econometrix, observed that population growth had accelerated “markedly since 2007 to 1.63 percent a year, from 1.01 percent during the six year period 2001 to 2007”.

The surge can only partially be explained by a decline in Aids deaths and an increase in immigration, according to some experts. The rapid rise, therefore, could imply an increase in fertility – a reversal of a declining trend that has been in place for decades.

There are two reasons why data on fertility is crucial.

One: in a developing country a declining fertility rate demonstrates that women are being offered more choices; it is a measure of their education and economic freedom. And it is a sign that children will have better life chances because small families have more resources per child than large families.

Two: while updated population data tells us what has happened since the last census in 2001, the fertility rate would tell us what lies ahead: how many babies are likely to be born and what the government needs to do to provide for the next generation.

Statistics SA says the fertility rate has still to be calculated. But, when it comes, it may do little to clear the confusion. In an explanation of the census process, the agency admitted that male fieldworkers did not feel free to ask questions on fertility and the relevant part of the questionnaire was often left blank.

Eric Udjo, the head of demographic research at Unisa’s Bureau of Market Research, said South Africa’s fertility rate had fallen consistently from 4.9 children per woman in 1970 to 2.7 in 2007, based on Stats SA’s community survey that year. He said the latest census findings showed the decline had stalled and, therefore, his estimate of the fertility rate remained unchanged at 2.7.

He said experience in Europe showed it was not uncommon for declines to stall. And he said his estimate on fertility was not out of line with the latest findings on population growth. Census 2011 showed 51.8 million people are living in South Africa – 2.4 percent higher than last year’s mid-year estimate of 50.6 million.

However others disagree.

A high profile critic of the census is Rob Dorrington, a professor at the UCT Centre for Actuarial Research, a member of an expert evaluation team for the Statistics Council, who presented an adverse minority report. Dorrington argued that the population acceleration was out of line with previous estimates of fertility, which had been “fairly consistent” between the 1996 census, the 1998 demographic and health survey, the 2001 census and the community survey in 2007.

Dorrington said, prior to the release of the latest census data, population growth was estimated to be growing at a little over 1 percent a year, taking into account both a decline in the number of Aids deaths and immigration. “Although our estimates of survival of children from birth may not be perfectly accurate, it is very unlikely that any error would account for anywhere near the sort [of] excess in the estimates of the number of children.”

He said: “The Actuarial Society of South Africa estimated the ‘total of fertility rates’ to be 4.2 children per women in 1985/86 and this had dropped to 2.4 children per women by 2010/11.“

He pointed out that a total fertility rate of 2.7 in the year before the census would imply that the infant mortality rate would have to have been more than double our current estimate simply to reproduce the numbers in census 2011. He also argued that a steadily declining total fertility rate, to 2.7 by 2007, would imply that the census undercounted the children above the age of 5.

There is a related anomaly in the data. An important census finding was a rapid migration to Gauteng and the Western Cape where the country’s largest cities are situated.

Jammine pointed out that the increase in urbanisation runs counter to the finding of accelerated population growth. “Internationally increased urbanisation has been associated with lower population growth over many decades because of the recognition by newly urbanised families of the costs of having larger families.”

However, he believed the anomaly could be explained by the increased roll out of antiretroviral treatment for people suffering from Aids.

Another related issue is the population dent near the bottom of the population pyramid. The latest census shows a marked decrease in the number of children aged 5 to 19.

Griffith Feeney, a demographic consultant working with Stats SA on the census, said at a press conference last week that Dorrington had pointed out “a pattern in the age distribution that seemed very peculiar and unexpected”.

Feeney said investigation of other data sources, including registered births and deaths confirmed the age distribution was “real”. But Dorrington noted: “Births and deaths are under registered so there is scope for making assumption.”

Stats SA said an explanation of the apparent change in social behaviour was not part of its brief. And Howard Gabriels, the chairman of the Statistics Council, said an explanation would be up to “scientists”.

Business Report

Related Topics: