TO ADVANCE the programme of social and economic development highlighted in Finance Minister Nhlanhla Nene’s budget speech, accuracy of data on SA’s population is needed.
Some commentators suggest comparing national census data with other sources. Such a comparison offers insights into the accuracy and interpretation of national data.
Using long-term population data, we do this for rural Mpumalanga, where a population registration system to record information on births, deaths and migrations has been in place for about 20 years.
Mpumalanga and Limpopo cover the former Bantustans of Kangwane, Gazankulu, Lebowa and Venda, home to thousands of rural families. Not all households are poor, but a sizeable number depend on sustenance from the natural environment. Employment and education opportunities are limited, so people often migrate to towns or cities for work.
Between 1996 and 2011 there have been rapid changes in the age and sex structure of the population, including a fall and, more recently, an increase in the number of children under the age of five compared to older children, as reported in Census 2011.
As fertility has been relatively stable in SA in recent times, this increase could be accounted for by increasing numbers of people — women especially — entering the reproductive age group and a large group of people born 20-24 years ago who are now entering the labour market.
Demographic transition theory suggests that mortality rates decline as healthcare, personal hygiene, food supply, urbanisation, education and overall wellbeing improve. A decline in fertility follows when women choose to have fewer children given improved child survival, educational opportunity, changing gender norms and contact with urban centres.
In Mpumalanga, fertility levels declined until about 2002, rose to a higher level from 2005 to 2008, and declined again after 2008. The fertility estimates for rural areas at the national level showed a similar decline and levelling off, but with fewer data points and higher rates, possibly due to these data representing a more diverse population.
Mortality among younger, working-age adults of 20-34 years old showed a sad increase between 1994 and 2006, rising from below 40 deaths per 1,000 persons in 1994 to about 150 per 1,000 in 2006. The rates have levelled off and are now steadily declining, reaching 110 per 1,000 in 2010 after sustained efforts by the Department of Health and related sectors to boost HIV/AIDS treatment and prevention.
Improvements in female survival are likely to influence fertility rates, contributing to the increase in the under-five population. Despite increasing mortality from HIV/AIDS, comprehensive cause-of-death data highlight the rapid rise in chronic medical conditions including cardiovascular and respiratory disorders, diabetes and cancers.
The national census gives an impression that some provinces are losing people while the provinces with metropolises are gaining. However, the census is a snapshot of population changes every 10 years, with well-known limitations regarding the dynamics.
Both shorter-distance migrations caused by households forming or splitting and longer-distance temporary migration for purposes of work and education occur. Our evidence suggests more than 70% of those who migrate retain links with their original families and communities and return frequently. This temporary migration increasingly involves women, many of whom join the labour force.
Using information on remittances, we see that most children in households left behind by migrant workers tend to have better health outcomes. Similarly, data suggest an overall improvement in the socioeconomic wellbeing of rural families.
When migrant labourers working in Gauteng state their income in a census questionnaire, the amount is treated as income generated within Gauteng. But this is not all expended in Gauteng and origin households benefit from the remittance income. The benefits for a rural household can translate into a heavy cost if a migrant becomes seriously ill and returns home for family care and support. Such insights can only be learnt from complementary sources of data.
Ownership of assets — appliances, building materials, access to electricity, communication and transport — can be assessed quite accurately in surveys and used as a measure of household socioeconomic status.
High asset scores suggest more resources and better opportunities for members of a household, while lower scores reflect fewer belongings and limited opportunities.
In Mpumalanga, asset ownership has increased between 2001 and 2011 for rural households, reflecting improvements in housing, electricity access and ownership of modern goods. A similar trend is seen in Census 2011, which reported 51% of households in Mpumalanga using electricity for lighting in 1996, 69% in 2001 and 86% in 2011.
The poorest households tend to improve their socioeconomic status through government grants and female labour migration, and are more resilient to shocks and stresses if they use available natural resources for fuel and dietary supplements. Better-off households improve socioeconomic status via local employment and male labour migration.
The findings for northeast Mpumalanga illustrate trends in former Bantustan areas and, potentially, other parts of rural SA.
Migration is a complex process and care is needed when interpreting urban-rural dynamics from the national census as most rural-to-urban movement takes the form of labour migration that links urban and rural households.
Population-based information systems are a resource for Statistics SA and can enhance the interpretation of data available at provincial and national level. Continuous follow-up of vital events (fertility, mortality and migration) and their determinants can support health and social development of local communities through improved planning that is based on robust data.
• The authors are researchers at the Medical Research Council/Wits Rural Public Health and Health Transitions Research Unit.
Post a comment