Migration and the changing rural-urban interface in South Africa: What can we learn from census and survey data?

Literature › Works

Servaas van der Berg; Rulof Burger; Murray Leibbrandt; Cecil Mlatsheni (2002). Migration and the changing rural-urban interface in South Africa: What can we learn from census and survey data?. Draft 1.

Type	working paper
Scope	Post-1994 South African economy
Method	econometric
Period studied	1995–1996
Regions	Transkei, Eastern Cape, Western Cape, Cape Town, Northern Province, Gauteng, South Africa
Themes	demography urbanisation, labour markets wages, inequality poverty unemployment, bantustans removals influx, methods data sources, transition post apartheid
Citation key	berg2002

Argument

The paper uses the 1996 census 10% sample and the 1995 October Household Survey to study black rural-urban migration after influx control was abolished. Logit models show migrants from Transkei and Northern Province are predominantly young adults with secondary education, and gender differences in migration are smaller than under apartheid. Heckman-selection probit models of participation and employment are used to predict urban employment chances for rural participants. Most rural participants would not be near the front of the urban job queue, but urban prospects are still better than rural ones for many, particularly women. Unemployed rural men are uncompetitive in both markets.

Evidence and method

10% sample of the 1996 census (movement in the nine months before October 1996); district-level mean income and poverty headcount from Alderman et al. (2000); 1995 October Household Survey for black men and women aged 16-65, with survey design effects accounted for.

Findings

• Most movers settle in urban areas, contradicting the view that migration is mainly rural-to-rural. (p. 5)

  Of all persons who moved residence in 1996, 78% now reside in urban areas.3 Of all black people who moved, 71% are now urban

• The Eastern Cape supplied the large majority of black in-migrants to the Western Cape in 1996. (p. 3)

  Table 1 confirms that by far the majority (90.2%) of black migration to the Western Cape came from the Eastern Cape.

• Age is the dominant influence on migration; moving from school age into young adulthood matters most. (p. 7)

  The single most important factor influencing the propensity to migrate is moving out of the school-age into young adulthood (the age group 19 to 30).

• Tertiary-educated Transkei residents do not migrate more, probably because of local public-sector jobs. (p. 15)

  the very few with tertiary education in Transkei are likely to remain there, because many of them are there precisely because they fill the few public sector jobs available

• The gender gap in migration is small, which suggests a change since influx control ended. (p. 14)

  the relatively small magnitude of the difference in migration propensities between males and females comes as somewhat of a surprise.

• The Cape Metropolitan Area attracts less-skilled migrants, while the educated prefer Gauteng or Port Elizabeth. (p. 12)

  These results point to the Cape Metropolitan Area being a more attractive area for less skilled black workers; more highly educated people prefer Gauteng or even Port Elizabeth

• Rural participants have lower predicted urban participation and employment than urban residents, and the typical rural participant is poorly placed for urban jobs. (p. 24)

  The majority of the rural sample do not look as though they have a decent chance at urban employment.

• For most rural participants the urban labour market still offers better prospects than the rural one. (p. 27)

  the situation in rural areas is so desperate that the urban labour market appears to offer them a better option.

• Many rural unemployed men are uncompetitive in both urban and rural labour markets. (p. 32)

  It is a sobering fact that most unemployed males are not competitive in either urban or rural labour markets.

• Rural income sources are dominated by wages, with transfers and remittances next and agriculture marginal. (p. 2)

  wages (often earned in urban areas)(52%), social transfers (18%), and remittances (14%)