Jonathan Crush (ed)
with Charles Mather, Freddie Mathebula, David Lincoln, Claude Maririke and Theresa Ulicki

Southern African Migration Project

Migration Policy Series No. 16

PLEASE NOTE: Readers are welcome to reproduce and reference
this article as long as appropriate acknowledgments are given.


INTRODUCTION: Making Hay with Foreign Farmworkers -- Jonathan Crush

CHAPTER 1: "The farmer prefers us": Mozambican Farmworkers in the Mpumalanga lowveld –Mpumalanga lowveld --Charles Mather and Freddie Mathebula

CHAPTER 2: Southward Migrants in the Far North: Zimbabwean Farmworkers in Northern Province -- David Lincoln with Claude Mararike

CHAPTER 3: Poverty and Women's Migrancy: Lesotho Farmworkers in the Eastern Free State --Theresa Ulicki and Jonathan Crush



"South Africa will not have farm workers with a love and knowledge of their work and a willingness to labour, till the farmworker is ensured a living wage, adequate food, housing, education, leisure and security -- above all to enjoy the fullness of life" (Ruth First, 1949)

International migrants have been employed on South African farms for many years.1 Why, then, has their presence in South Africa produced such controversy since 1994?

First, the new South African concern with labour and human rights has focused official attention on employment sectors characterized more by the violation than the observance of basic rights. Commercial agriculture is often deemed to fall into this category. There is widespread suspicion -- fuelled by sensational media reports and independent investigators such as the Human Rights Commission, the International Labour Organization (ILO) and the Labour Market Commission -- that foreign workers are living and working on South African farms in particularly dire and exploitative conditions.2 These conditions clearly violate new labour laws and basic employment standards.

Second, the South African government is committed to reducing mass domestic unemployment. Sectors employing non-South Africans have inevitably been pressured to explain why. The government’s stated policy is that foreign workers should only be employed when no South Africans are available. But there are no objective mechanisms for proving or denying that foreign workers are really needed on the farms. Farmers say that foreign labour is essential; South Africans do not like farm work and the farms would go under without it. Critics respond that if the farmers paid a living wage and improved conditions, South Africans would work on the farms.

Third, white farmers were a privileged elite in the old South Africa; politically powerful and able to secure special privileges denied to other employers. Without massive infusions of state support and subsidies over many decades, commercial agriculture would never have developed to its current levels.3 A critical component was cheap local and foreign labour organized for agriculture by the state. Why, it is now asked, should they continue to enjoy the fruits of past political connections: state subsidies, side-deals on foreign labour, and immunity from employer sanctions?

Fourth, it is widely but erroneously assumed that South Africa has been swamped since 1994 by millions of foreigners. Research shows that anti-foreign sentiment in South Africa is all-pervasive and xenophobic.4 Any region or sector that encourages in-migration through employment is viewed as responsible for the supposed "flood" of migrants.

Finally, the state has committed itself to rooting out and deporting migrants without work permits or contracts. Farming areas have been targetted by the police. Farmworkers are an easy target for aliens control units wishing to boost the volume of arrests and magnify the country’s "crime-fighting" statistics.5

The public debate over farm labour has generated a good deal more heat than light. That a debate could take place at all is testimony to the greater openness of the new dispensation. The apartheid state kept the farm labour issue firmly behind closed doors. Only when independent investigators, such as the late Ruth First, published damning reports of conditions on the farms did the state reluctantly act.6 More often it simply acquiesced to the demands of white farmers and their political representatives. In the 1960s, for example, the state became concerned about the numbers of foreign migrants in South Africa and appointed a commission of enquiry (the Froneman Commission). Froneman recommended that foreign labour dependence be reduced. At the time, these recommendations were so politically explosive (and so roundly condemned by the mines and the farms) that the apartheid government simply shelved the report and ignored the recommendations.7

The development of a clear, consistent and implementable post-apartheid policy on the hiring of foreign workers by commercial agriculture has nevertheless been hampered by a number of factors:

The Southern African Migration Project (SAMP) is committed to the principle of delivering accurate and reliable information on migration to better enable policy-makers to devise workable solutions. The FFSA 1996 found that nationally, 16% of farms had some workers from Lesotho, 11% had workers from Mozambique, 8% had workers from Zimbabwe and 5% had workers from Malawi.13 SAMP has focused its attention on three different case study areas where foreign migrants from neighbouring countries are known to be present in even greater numbers: the Mpumalanga lowveld (mainly Mozambicans), the Northern Province (Zimbabweans) and the eastern Free State (Basotho). These studies aimed to shed light on patterns and processes of foreign migrant employment in three border areas. There were four main objectives:

These three studies were conducted by different teams of researchers and, for logistical reasons, at different points in 1998. In addition, research methods used in one area (such as interviews with farmers) were impossible in another. The net result is that the three studies had slightly different research designs. The results are therefore not strictly comparable between areas and are reported in different formats by the various authors. SAMP’s decision to publish them together was prompted by our conviction that, despite the variations in methods and emphasis, the studies complement one another.

These studies provide unprecedented insights into conditions on the farms in each of the three areas and they allow the authors to draw policy-related conclusions out of local experiences rather than from the top down. What is abundantly clear from all three studies is that policy reform is urgently needed in this areas and that a draconian approach to managing migration from these three countries is likely to be singularly ineffectual.

The similarities and differences between the three study areas need to be highlighted at the outset. This is because commercial agriculture is complex and variable in its ownership patterns, production regimes, deployment of technology and labour requirements. The mechanics of foreign labour usage and the mechanisms of official control also vary between each of the three regions. The reasons are part geopolitical and inter-governmental, part historical and part local. This poses real challenges for policy standardization.

In all three areas, non-South African citizens are employed in considerable numbers and in each case there are clear ties to only one source country: Mozambique and Mpumalanga, Zimbabwe and Northern Province, Lesotho and the Free State. However, it is a major fallacy to assume that all non-South Africans are recent migrants. The FFSA Survey concluded that over 50% of "immigrant farmworkers" had been on the farm for more than 5 years. As many as 16% had been there for 11-20 years and another 10% for over 20 years.14 The major difference with South African workers was that only 22% of the latter had been on the farms for less than 5 years (compared to 47.4% of non-South African workers). The findings suggest, first, a longstanding pattern of permanent farm work and residence in South Africa by non-South Africans; and second, increased employment of non-South Africans after 1990.

These general conclusions are confirmed by these studies. However, it also seems that border farmers are drawing, perhaps like never before, on cross-border migrants to meet their temporary and seasonal labour needs. Unlike the mining industry where, for example, Mozambican and Basotho migrants can be found on all mines, there is clearly a major distance and proximity effect operating in commercial agriculture. Most migrants travel on foot or by bus or taxi and have very limited means or incentive to travel the long distances that would be involved in getting to other farming districts. As farming in the Western Cape becomes more dependent on migrant labour, farmers have been trucking in labour from Lesotho.15 But this tends to be the exception.

In the case of Zimbabwe and Mozambique, the farms also straddle major migration and transportation routes to the south and west. As a result, many migrants simply use the farms as a "refuelling station" before moving on to their primary urban destination. The fact that farmers in these areas experience much greater rates of labour turnover is therefore not surprising.

Foreign farm labour in South Africa is thus primarily a "borderline" phenomenon. This gives it a very distinctive local character. Migrants move backwards and forwards across the border far more frequently. The borders are extremely porous. The migrants in turn are very well-versed in the legal and clandestine means of border-crossing. They know the district very well, which farms to avoid and where to seek work. If undocumented, they know where not to be found. On the other hand, officials and the security forces know how and where migrants travel, where to set up roadblocks, and which farms tend to employ foreign migrants. Most migrants cross the border for several months at a time. In the case of Lesotho and border farms, there is a daily movement back and forward across the Caledon.

Despite their common features as migrant border zones, the employment and labour market situation in each of the three areas varies considerably. The Mpumalanga lowveld is undoubtedly the most complex; the eastern Free State the simplest. Mpumalanga, unlike the other districts, has a large resident population of non-citizens; mainly Mozambicans who came to the area during the Mozambican War in the 1980s. These residents are a ready source of farm labour and are often confused with recent migrants in the media and by officials. On any large farm on the average harvesting day, one can find permanent farm workers (mainly South Africans), seasonal workers (mainly Mozambicans and South Africans living in the former homeland areas) and temporary workers (almost exclusively Mozambican migrants and many who are undocumented).

That workforce would also exhibit a complex combination of degrees of legality. There would be South African citizens, legal and illegal Mozambican ex-refugees (some with South African ID’s acquired in the recent amnesty and some without), Mozambican "legal illegals" (migrants with forged or counterfeit South African documentation) and Mozambican undocumented migrants (some with Mozambican ID’s, many without). In addition, the legal temporary workers from Mozambique would include some recruited legally by Algos in Mozambique (under a longstanding bilateral treaty between Mozambique and South Africa), some with legal contracts acquired for them by the farmer after they were hired illegally, and some with so-called "farm ID’s" (name tags issued by farmers to protect their workers from arrest, under an informal understanding with the local police).

In the eastern Free State and Northern Province, the situation is less complicated. Farmers in both areas have been allowed to legally hire foreign migrants although the policies and regulations that permit this to happen differ considerably. Lesotho and South Africa have a bilateral treaty which allows employers and recruiters to hire Basotho legally and on contract through Labour Offices in Lesotho. At present, upwards of 7,000 migrants are recruited this way for work on asparagus farms in the Free State. This is simply a modernized version of the old migrant labour system with some new wrinkles: many of the contract workers are women, there is no union like the NUM to fight for their rights, and there are few controls over working and living conditions.

Zimbabwe and South Africa have no bilateral labour treaty. In the Northern Province, officialdom has reacted pragmatically to a de facto situation of undocumented migration to farms by regularizing it. Informal border-crossing points and the issue of permits to border-crossers seeking farmwork (under an exemption clause in the Aliens Control Act) has cut the mortality rate from drowning and crocodiles in the Limpopo and allowed officials to get some handle on the magnitude of cross-border migration in the district. Chapter Two of this publication still shows, however, that despite the incentive to legalize, many migrants do not bother primarily because the disincentives are so slight. In general, as other SAMP research shows, migrants prefer to utilize legal means of border-crossing where these are available.16

Despite the differences of geography, production, workforce profile and regulatory mechanism, there is one major feature common to all three districts: the extremely poor working and living conditions endured by most migrants on many farms. Guaranteed anonymity and interviewed, in at least two areas, in their home countries we are confident that the majority of interviewees responded honestly and without malice. What is striking from the research is the consistency across regions:

In all three border areas, it is quite permissible for foreign migrants to work legally on South African farms. But it is still a curious anomaly that migrants working legally on the farms in three different parts of South Africa should be allowed entry under such variable legislative mechanisms and regulations. There is clearly a case for standardization and transparency, both for administrative ease and to address the perception that farmers still enjoy special privileges in the labour market which are denied to other employers.

Given the existence of these legal avenues, why do some migrants go to quite extraordinary lengths to avoid them? There are a number of reasons. First these avenues are neither clear, accessible to all nor efficient. As long as undocumented migrants can be legalized after coming to South Africa, there is little incentive to acquire or pay for permits up front. Second, at least in Mpumalanga, many temporary farmworkers are just that. They have come to South Africa with other destinations in mind. Third, migrants know that if they make it to the farms, they will be employed. Once there they are on "protected ground"; as long as they stay there, they know the police and other officials will leave them alone. Finally, at least on Free State border farms, undocumented migrants from Lesotho sacrifice marginally higher wages for the advantages of living at home, not being required to pay for a passport, not being tied to farms for months at a time, changing employees when they want and working throughout the year.

Farmers hire undocumented migrants, particularly during harvesting, for three reasons. First, they are available and accessible. As soon as the cane fires are spotted, the migrants arrive for work. It takes a brave or principled employer to turn the labour away when the crop is ready to go. Second, the legal channels for obtaining workers are slow and cumbersome by comparison. And third, some farmers undoubtedly like to hire workers without documents because they are, by definition, more vulnerable and exploitable. The threat of exposure (coupled with the knowledge that the farmer himself is immune from prosecution) can, and is, used to keep workers in line.


1. Alan Jeeves and Jonathan Crush (eds), White Farms, Black Labor: The State and Agrarian Change in Southern Africa, 1910-50 (Portsmouth NH, Pietermaritzburg and Oxford: Heinemann, University of Natal Press and James Currey, 1997).

2. Restructuring the Labour Market: Report of the Commission to Investigate the Development of a Comprehensive Labour Market Policy (Pretoria: Department of Labour, 1996); Guy Standing, John Sender and John Weeks, Restructuring the Labour Market: The South African Challenge (Geneva: ILO, 1996); South African Human Rights Commission, Investigation of Alleged Violations of Farmworkers’ Rights in the Messina/Tshipise District (Johannesburg, 1999).

3. Jeeves and Crush, White Farms, Black Labor, pp. 1-28.

4. R. Mattes, D.M. Taylor, D.A McDonald, A. Poore and W. Richmond, Still Waiting for the Barbarians: SA Attitudes to Immigrants and Immigration (SAMP Migration Policy Series No. 14, Cape Town, 1999).

5. Police reports consistently list arrests of "illegal immigrants" in the same breath as statistics for arrests for car-hijackings, drug busts, murders and armed robbery.

6. Ruth First, The Farm Labour Scandal (Johannesburg, 1949); Michael Scott, A Time to Speak (Garden City NY: Doubleday, 1958); Ben Turok, "The African on the Farm" Africa South 4(1), (1959): 28-33.

7. Ken Owen, Foreign Africans: Summary of the Report of the Froneman Committee (Johannesburg, 1965).

8. Stephen Greenberg, Meshack Hlongwane, David Shabangu and Elken Sigudla, State of South African Farmworkers 1996 (Farmworkers Research and Resource Project, 1997).

9. Ibid., p. 50-51.

10. Ibid., p. 50.

11. Jonathan Crush and Clarence Tshitereke, "Hidden Treaties: South Africa’s Bilateral Agreements with Its Closest Neighbours" (unpublished SAMP report, 1999).

12. The drafters of the recent White Paper on International Migration argue that the entry of contract workers from neighbouring countries (Mozambique and the BLS countries) is governed purely by labour agreements. Because there is no labour agreement with Zimbabwe, "no contract workers may thus be recruited from Zimbabwe and prospective workers from Zimbabwe are to follow the normal procedures of obtaining a work permit." This is clearly incorrect, as Chapter 2 shows. In Mpumalanga, it is also obvious that there is a yawning gap between the text of the bilaterals and actual practice on the ground.

13. State of South African Farmworkers, p. 50.

14. Ibid., p. 51.

15. "Boland Slave Trade Exposed" Cape Times 29 August 1996.

16. David McDonald, John Gay, Lovemore Zinyama, Robert Mattes and Fion de Vletter, Challenging Xenophobia: Myths and Realities about Cross-Border Migration in Southern Africa (SAMP Migration Policy Series No 7, 1998); and David McDonald, Lephophotho Mashike and Celia Golden, The Lives and Times of African Migrants and Immigrants in Post-Apartheid South Africa (SAMP Migration Policy Series No 14, 1999).

17. See State of South African Farmworkers.

18. White Paper on International Migration, Clause 8.5.

19. W.R. Böhning, Employing Foreign Workers: A Manual on Policies and Procedures of Special Interest to Middle- and Low-Income Countries (Geneva: ILO, 1996); International Labour Organization, Protecting the Most Vulnerable of Today’s Workers (Geneva: ILO, 1997).

20. Jonathan Crush, "Devising a Temporary Work Scheme for South Africa" (Report submitted to Department of Home Affairs, March 1999).