![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
In this article I investigate the characteristics and determinants of out-migration in the context of the HIV/AIDS epidemic, using data from a panel designed to investigate the household impact of the epidemic. Departure models show that individual attributes, notably age and gender, play an important role in explaining out-migration from households that have not experienced morbidity or mortality. In affected households, a number of household-level variables, notably the sex of the household head, place of residence, family structure, the dependency ratio, human capital and household size, feature as important determinants of out-migration. Health shocks independently explain part of observed differences in out-migration from affected households, the out-migration of ill persons from affected households and the out-migration of orphaned children from affected households. Thus, migration represents an important strategy for poorer households having to cope with the HIV/AIDS epidemic, both as an economic survival strategy and as a social strategy aimed at accessing support from the extended family.
Keywords:
Notes
1. According to Lurie (Citation2000), the role of migration in the diffusion of the HIV/AIDS epidemic remains unclear, given the focus in most migration studies on ‘receiving’ areas only, thus precluding the study of migrants and their families concurrently in both the ‘receiving’ and ‘sending’ areas.
2. This research project was sponsored jointly by the UNDP and the foreign development agencies of Australia (AusAID), the United Kingdom (DfID) and the United States (USAID) and administered firstly by the Joint Center for Political and Economic Studies and more recently by the Joint Economics, AIDS and Poverty Programme (JEAPP) of the African Asian Society. Other research support includes a research grant from the National Research Foundation (NRF) and a Mellon Foundation grant from SALDRU, University of Cape Town.
3. An exception, though, to the above evidence of urban to rural migration of HIV/AIDS patients are two studies conducted in Canada (Hogg et al. Citation1997) and the United States (Buehler et al. Citation1995) respectively. Both studies found that a relatively small proportion of infected persons actually changed residence between AIDS diagnosis and death. However, this may reflect the limited mobility of people between AIDS diagnosis and death, given that the illness at this stage is quite severe and patients often are weak and unable to perform any daily tasks. Therefore, the latter evidence does not negate the above evidence that mobility is relatively high between HIV diagnosis and treatment or death.
4. Caution, however, is required insofar as not all households or migrating individuals were tracked in this survey; due to cost constraints a decision was made only to track households that moved from one residence to another if they remained in the same study site. As such, the results presented in this article provide an incomplete picture of migration patterns in this cohort of households. Nevertheless, the data provide some useful insights into aspects of out-migration in the context of the HIV/AIDS epidemic.
5. Similar data were collected since the third round of the panel for persons who had joined the households in the study since the previous round of interviews. These data, however, are not analysed here because there is no information about the characteristics of the households from which these persons originated. Given that the so-called departure models employed to investigate determinants of migration require such information, out-migration rather than in-migration is the focus of this article.
6. According to Massey (Citation1990), however, this predominant focus on the determinants rather than the consequences of migration has meant that our understanding of migration has remained partial. Greenwood (Citation1985) correspondingly highlighted the fact that most advances in migration studies in the 1980s pertained to our knowledge of the causes of migration.
7. According to Findlay (Citation1987) and Donato (Citation1993), migration runs in families and the probability of migration is higher for households with a previous experience of migration. However, our data provided no evidence to this effect and past migration experience was not a statistically significant determinant of out-migration in any of the departure models.
8. This paper reports and discusses the nature of these familial relations as if they mean the same thing to everyone. Caution, however, is required in such interpretation, given that these relationships may mean different things in different social and linguistic settings.
9. It should be noted that the definition of a household that Statistics South Africa employs in the October Household Survey probably exaggerates this pattern given that the household is defined with reference to living arrangements over a period of one week only. Other household surveys, for example, define the household with reference to living arrangements over a longer period of time, for example three months or longer.
10. Caution is required insofar as these findings are based on self-reported orphan status (based on whether the child's father and/or mother was alive at the time). This could result in the over-reporting of paternal orphanhood in particular insofar as the father may be reported as deceased where the mother does not know the father of her child and/or the father is estranged from the child's mother or family.