What can land reform achieve in the 2020s? Historical and comparative reflections on the potential of land redistribution to contribute to transformation and poverty alleviation in South Africa

Abstract:This article assesses the likelihood that land reform will reduce poverty and inequality by highlighting important historical processes and relevant comparative experiences. These lead us to a different perspective than the one driving calls for a radicalisation of land reform. Conquest and colonial rule reserved most of the land for whites, and processes of land alienation and discrimination were constantly exacerbated until South Africa became a democracy. At the same time, while a system of private property and individual title was secured for whites, land rights were systematically diminished within black societies. Through most of the twentieth century an entrenched system of migrant labour ensured a massive flow of people, initially mostly men, between rural 'reserves' and white-owned mines, offices, suburban households and the factories of the emerging manufacturing sector. Influx control prevented black families from moving to, or acquiring property rights in, urban areas. During the apartheid era, these processes destroyed the ability of most Africans to generate any income from agriculture, while also leading eventually to mass unemployment and a dependence on state grants for those who continued to live in rural areas. Meanwhile, rather than benefitting from state attempts to boost their productivity on the land, weaker white farmers mostly exited the rural areas and moved to urban centres, where their standards of living rose substantially. Combining these historical insights with the reality that the successful land reform programmes of Asia were based on providing poor farmers with access to land titles rather than additional amounts of land, leads us seriously to question the notion that transferring huge amounts of land to black families will substantially reduce poverty or inequality over the next ten years.