Commentary on the paper ‘Striving against adversity: the dynamics of migration, health and poverty in rural South Africa’ by Mark A. Collinson

Across the developing world, rural-to-urban migration (i.e. internal as opposed to international migration) is probably the most important demographic phenomenon at this time. In the past year the global population has become 50% urbanized, and Africa is well on the way, and will attain that milestone in two more decades, although with considerable country variation. Asia will attain it five years sooner than Africa, and Latin America has done long ago (UN Population Division, World Urbanization Prospects: The 2009 Revision).

This trend towards urbanization should not be interpreted as a sign of economic progress, as it was in the West. Rather, in many poor countries it is a reflection of decades of high rates of population growth associated with large family sizes and population momentum, in a context of a declining capacity of agriculture in rural areas to absorb the ever-increasing numbers of young people entering the working ages.

A number of developing countries control rural-to-urban migration through the use of population identity cards and registration of movement (e.g. China, Indonesia), although they do not necessarily have the database systems required to track numbers accurately. One of the features of the studies carried out by Dr. Collinson is that the Demographic Surveillance System (DSS) in the Agincourt field site allows accurate monitoring of numbers migrating out of the rural area, to know something of their reasons, and their destination, and also how many return (1). These data are rather unique to DSS and will increasingly justify the maintenance of the DSS approach to monitoring population change within its social and economic context. Aspects of the studies in this paper also highlight what a DSS cannot do well, and that is to know what happens to out-migrants when they are away from the DSS area.

As we read in this paper, South Africa under Apartheid was not typical of other poor countries. Rural-to-urban migration was tightly controlled by the State to feed the labor demands of the industrial and mineral exploitation economy. As Dr. Collinson points out, it was expected that after the Apartheid system was abolished, it was assumed that the population would revert to the usual voluntary migration patterns where families often move as a unit to the cities. This did not happen, although there were other changes with younger men, and more women moving to the cities, though not as family units. It is important to understand why rural families continue to maintain their rural homes and lands while sending individual family members to the city. Unlike in the past, these migrants have little guarantee of employment in the present circumstances, thus making social networks more important to urban survival and gaining productive employment. More research is needed on this topic in Agincourt and elsewhere because it is a model that reflects a reduction in the overall rate of urbanization, compared to when entire families move permanently to the cities.

The most significant aspect of the migration issue is whether or not it is a successful strategy which brings economic benefits to the sending family. And do these economic benefits outweigh the social and other costs to the families left behind. The study identifies the selection question whereby there are costs associated with migration, so only better off families can actually afford to send a migrant to the city. And thus only those better off families can potentially benefit from the remittance income. This selection bias may influence the apparent economic benefits of the remittance system.

The paper documents well the important issue of what happens to those family members left behind when an adult family member migrates to the city in search of work. It should not be assumed that rural life carries on with ‘business as usual,’ as the decision-making role and authority of the head of the household may continue to reside with that person when they migrate out. This can affect and delay urgent decisions on seeking emergency health care, to education of children, and even marriage of children. It has implications for the economic productivity of the family, and for the security of family members remaining behind, as reflected in the HIV-infection patterns described in the paper.

It is evident that the absence of a mother through out-migration can be almost as catastrophic for the children as death of the mother, and certainly a much greater negative impact than fathers leaving. In addition, this study is set in a context where out-migration of family members may be combined with death to family members due to the pervasive dark cloud of HIV. This is important because it appears that these split families may be evolving as the norm for many future communities, thus it is important to document these trends and impacts, and attempt to mitigate the negative consequences, especially on the spouses and children left behind.

The component of the study on ‘returning home to die’ among HIV sufferers describes a pattern that may reasonably be expected, and presumably applies to sufferers of other potentially terminal diseases like cancer. The significance of this pattern is that, simultaneous with the family's loss of remittance income from the migrant family member a new economic burden of health care for the ill-returning family member. But it is not clear what alternative options can be put in place, other than provision of ART HIV treatment to reduce the incapacity of the HIV patient.

Overall, the paper highlights a number of important socio-economic and demographic changes that are both unique to post-Apartheid South Africa and at the same time, typical of much of the developing world. The power of the Agincourt DSS to generate comprehensive data on individuals in their rural community context is impressive, but raises the question of how much more powerful would a DSS be for migration studies if it could be linked to similar surveillance at the receiving end of the process. At this time, the In-Depth Network of DSS sites, of which Agincourt is a member, only has one fully functional urban DSS (APHRC in Nairobi, Kenya). Surely one of the conclusions of this study is that urban populations need to be much better understood in these times of rapidly urbanizing societies, and DSS is one way to do that. Nevertheless, the various components of this study have produced very useful and important insights into South African rural life.