AFRICAN HUMAN MOBILITY REVIEW Volume 3 Number 2 May – August 2017 Guest Editor’s Note: Mythologies of Migration and the Informal Sector Benign Neglect or Active Destruction? A Critical Analysis of Refugee and Informal Sector Policy and Practice in South Africa Jonathan Crush, Caroline Skinner and Manal Stulgaitis Refugee Entrepreneurial Economies in Urban South Africa Jonathan Crush, Godfrey Tawodzera, Cameron McCordic, Sujata Ramachandran and Robertson Tengeh Comparing Refugee and South African Migrant Enterprise in the Urban Informal Sector Jonathan Crush and Cameron McCordic Security Risk and Xenophobia in the Urban Informal Sector Sujata Ramachandran, Jonathan Crush and Godfrey Tawodzera The Owners of Xenophobia: Zimbabwean Informal Enterprise and Xenophobic Violence in South Africa Jonathan Crush, Godfrey Tawodzera, Abel Chikanda and Daniel Tevera Linking Harare and Johannesburg through Informal Cross-Border Entrepreneurship Godfrey Tawodzera and Abel Chikanda Informal Entrepreneurship and Cross-Border Trade between Mozambique and South Africa Abel Chikanda and Ines Raimundo ii AHMR _________________________________________________________________________________ Chief Editor Dr Mulugeta F. Dinbabo University of the Western Cape, South Africa Board Members Dr Beneberu Assefa Wondimagegnhu Bahir Dar University, Ethiopia Dr Delali Margaret Badasu University of Ghana, Ghana Dr Edmond Agyeman University of Education, Winneba, Ghana Dr Ernest Angu Pineteh University of Pretoria, South Africa Dr Joseph Yaro University of Ghana, Ghana Prof Laurence Piper University of the Western Cape, South Africa Dr Linda Oucho African Migration and Development Policy Centre, Kenya Prof Loren Landau University of Witwatersrand, South Africa Dr Lothar Smith Radboud University, Netherlands Dr Meselu Alamnie Mulugeta Bahir Dar University, Ethiopia Prof Raul Delgado Wise Universidad Autónoma de Zacatecas, Mexico Dr Razack Karriem University of the Western Cape, South Africa Dr Sharon Penderis University of the Western Cape, South Africa Prof Simon Bekker University of Stellenbosch, South Africa Prof Shimelis Gulema Stony Brook University, New York Prof Thomas Faist Bielefeld University, Germany ____________________________________________________________________________________ AHMR is an interdisciplinary peer-reviewed on-line journal created to encourage and facilitate the study of all aspects (socio-economic, political, legislative and developmental) of Human Mobility in Africa. Through the publication of original research, policy discussions and evidence research papers, AHMR provides a comprehensive forum devoted exclusively to the analysis of contemporaneous trends, migration patterns and some of the most important migration-related issues. AHMR is published by the Scalabrini Institute for Human Mobility in Africa (SIHMA) a member of the Network of the Scalabrinian Centers for Migration Studies, with institutions in New York, Paris, Rome, Buenos Aires, Sao Paulo and Manila. Articles and reviews in AHMR reflect the opinions of the contributors. All rights reserved. No part of this publication may be reproduced without the written permission from the publisher. ISSN 2410-7972 (online) ISSN 2411-6955 (print). Webpage: www.sihma.org.za Copyright © 2017 by the SCALABRINI INSTITUTE FOR HUMAN MOBILITY IN AFRICA 47, Commercial St, 8001 Cape Town – South Africa Tel. 0027 021 461 4741, Email: editor@sihma.org.za http://sihma.org.za/mulugeta-dinbabo http://sihma.org.za/beneberu-wondimagegnhu/ http://sihma.org.za/delali-badasu/ http://sihma.org.za/edmond-agyeman/ http://sihma.org.za/laurence-piper/ http://sihma.org.za/linda-oucho/ http://sihma.org.za/loren-landau/ http://sihma.org.za/lothar-smith/ http://sihma.org.za/meselu-alamnie-mulugeta/ http://sihma.org.za/raul-delgado-wise/ http://sihma.org.za/sharon-penderis/ http://sihma.org.za/shimelis-gulema/ http://sihma.org.za/thomas-faist/ mailto:editor@sihma.org.za iii African Human Mobility Review ___________________________________________________________________________ Volume 3 Number 2 May – August 2017 – Special Issue IV-XIV Guest Editor’s Note: Mythologies of Migration and the Informal Sector Jonathan Crush 751 Benign Neglect or Active Destruction? A Critical Analysis of Refugee and Informal Sector Policy and Practice in South Africa Jonathan Crush, Caroline Skinner and Manal Stulgaitis 783 Refugee Entrepreneurial Economies in Urban South Africa Jonathan Crush, Godfrey Tawodzera, Cameron McCordic, Sujata Ramachandran and Robertson Tengeh 820 Comparing Refugee and South African Migrant Enterprise in the Urban Informal Sector Jonathan Crush and Cameron McCordic 854 Security Risk and Xenophobia in the Urban Informal Sector Sujata Ramachandran, Jonathan Crush and Godfrey Tawodzera 879 The Owners of Xenophobia: Zimbabwean Informal Enterprise and Xenophobic Violence in South Africa Jonathan Crush, Godfrey Tawodzera, Abel Chikanda and Daniel Tevera 910 Linking Harare and Johannesburg through Informal Cross-Border Entrepreneurship Godfrey Tawodzera and Abel Chikanda 943 Informal Entrepreneurship and Cross-Border Trade between Mozambique and South Africa Abel Chikanda and Ines Raimundo iv Guest Editor’s note: Mythologies of Migrants in the Informal Sector On a recent visit to Washington DC, I had an animated discussion with a taxi driver who was an avid supporter of Donald Trump. The driver was not the stereotypical white, middle-aged, small-town, working-class, non-college- educated, angry voter who put Trump in the White House. Rather, he was a former television producer and poet from India who had immigrated to the US in the 1990s, and was adamant that Trump would stop the “flood of illegal aliens” into the country. His vigorous defence of Trump was a surprise, as was his buy-in to Trump’s anti-immigrant discourse of threat. There are few issues in the contemporary world that generate so much uninformed debate and misinformation as immigration (Blinder, 2015; de Haas, 2008; Hellwig and Sinno, 2017; Valentino et al., 2013). In a recent three-volume study on the mythologies of migration, Arcarazo and Wiesbrock (2015) argue that there are three foundational and intertwined myths structuring citizen reactions and government responses to established and new migration flows. The first myth is that sovereign territories and local labour markets are being “swamped” by migrants to the detriment of citizens. The second is that all migration is driven by poverty in countries of origin. And the third is that migration is economically negative for receiving countries. These myths, and variants thereof, are present in public and policy responses to migrants running informal-sector businesses in South Africa (Crush and Ramachandran, 2015). The swamping myth is particularly evident in the numbers game popular with South African politicians, government officials and the media. In a briefing to an Ad Hoc Parliamentary Committee on the work of the Inter-Ministerial Committee on Migration (IMC) in 2015, for example, Minister Jeff Radebe observed that “a heavy influx of foreign nationals has led to migration laws not being adhered to due to border management laxity.” As a result, “it was very clear [to the Minister at least] that in townships and in villages foreign nationals were dominating” (PMG, 2015). Here, in just a few words, he performed the common linguistic trick of juxtaposing migration, unlawful entry, and the disadvantaging of South Africans. As if to emphasise that there could be no other outcome from migration, he also asserted that there were 5-6 million migrants in the country, representing 10% of the population (NA, 2015: 5.4). A local current affairs television programme, Carte Blanche, later claimed that there were as many as 6 million Malawians in South Africa, representing one-third of that country’s population (Africa Check, 2017). Others have implausibly claimed that there are 800,000 Nigerians in South v Africa. Statistics South Africa (SSA) is thus currently trying to work out why Census 2011 recorded a total of 2,188,872 foreign-born people in the country while the 2016 Community Survey recorded only 1,578,541 (SSA, 2016:25). The implication that the foreign-born population might conceivably be declining does not, as the SSA admits, “conform to expected outcomes” (SSA, 2016). Similar hyperbole has clouded the issue of the number of asylum-seekers in the country. The UNHCR (2016: 44) controversially claimed in 2015 that the number of asylum-seekers in South Africa at the end of 2014 was 1,057,600. South Africa, according to the UNHCR, therefore hosted more asylum-seekers than any other country in the world, and more than double the number in Germany in second place (UNHCR, 2016: 44). In its next annual report on forced displacement, however, the UNHCR (2017: 45) noted that there had been a “sharp reduction” from 1.1 million to 218,300 asylum-seekers at the end of 2016. It would have much more helpful to admit that the 2015 figure was a mistake and that South Africa was never, in fact, host to the largest number of asylum-seekers (Stupart, 2016). Regrettably, the claim and the numbers became part of government’s case for abandoning the rights-based post-apartheid model of refugee protection and its replacement with the draconian proposals of the recent Refugees Amendment Act (Cape Times, 2016). The White Paper on International Migration admits that the number is actually less than 100,000 but still goes on to endorse and justify the new exclusionary approach. The damage has been done, with fake numbers and the authority of the UNHCR used to support the myth of swamping with extremely detrimental implications for refugee protection. The second foundational myth is that poverty is the root cause of all migration. This is the myth that collapses the many complex and varied causes of migration into one and homogenises all migrants into a single impoverished category. This myth expresses itself in various ways in South African policy discourse. For example, politicians regularly claim that South Africa is a destination for migrants from poverty-stricken Africa, conveniently overlooking that many countries are growing much faster economically than South Africa and that the South African economy and job creation in the country are major beneficiaries of investment in Africa. The motives of migrants coming to South Africa from the rest of Africa are extremely diverse and, in general, it is not the poorest who migrate. Even if we grant that poverty is a root cause of some migration from neighbouring countries like Zimbabwe, Mozambique and Lesotho, it does not follow that migrants are desperate and vi faceless victims. They are, in the main, highly motivated individuals with extraordinary degrees of ingenuity, tenacity and resilience. The homogenising and (mis)naming of migrants is a regular South African pastime. There are the derogatory epithets that are used on the streets and in communities to insult non-South Africans to their faces. As one Zimbabwean cited in a paper in this special issue commented: “The children learn it from their parents. They call us makwerekwere. Do you know even small kids can call you makwerekwere? Is that not xenophobic?” However, there are other, only slightly more sanitised, homogenising labels that erase diversity and emphasise exclusion in policy discussion. In the 1990s, the apartheid-era term “aliens” was used by politicians and the media to describe all migrants. This term was replaced in the 2002 Immigration Act, but by an equally alienating term: the “foreigner.” The terms “foreigner” and “illegal foreigner” are now common currency, courtesy of the Act, and are used incessantly to draw a sharp distinction between those who belong in South Africa and all those (non-South African-born) who supposedly do not. Government also frequently divides migrants into just two categories: economic migrants (said to constitute 90% of the population) and refugees (the other 10%). An associated claim repeated in the White Paper, for which there is also no proof, is that 90% of refugees are really economic migrants. The third foundational myth relates to the supposed negative economic impacts of migrants on receiving countries. This myth has consistently been undermined by research evidence that points to the positive impacts of migration and migrant entrepreneurship on host economies (Kloosterman and Rath, 2003; Fairlie and Lofstrom, 2015). In the South African case, this mythology takes on a particular inflection: that is, that migrants are “stealing our jobs”; a claim with superficial plausibility given the high rates of unemployment in the country. However, past surveys have shown that the proportion of South Africans who have actually lost a job to a migrant is relatively low. And the calculus fails to take into account the economic advantages to South Africans from hiring migrants and, by extension, the benefits to the state in increased tax revenues. The negative economic impacts argument has crystallised in recent years around opposition to the activities of asylum-seekers and refugees in the informal sector, whose right to establish and operate businesses in the informal economy has nevertheless been upheld by the courts (Crush et al., 2015; Rogerson, 2016). The IMC, for example, maintains that migrants are “dominating trade in certain sectors such as consumable goods in informal settlements which has had a negative vii impact on unemployed and low skilled South Africans” (NA, 2015). The IMC ignores the positive economic impacts of refugee and migrant informal entrepreneurship including job creation for South Africans, profits for local retailers and wholesalers, licensing fees paid to municipalities, cheaper goods for poor consumers, and rent paid to South African property owners (Basardien et al., 2014; Jinnah, 2010; Peberdy, 2016, 2017; Radipere, 2012; Tawodzera et al., 2015). Indeed, South Africans have been publicly condemned by government ministers for renting their properties to refugee and migrant entrepreneurs (NA, 2015: 7.17). The papers in this special issue on migrant entrepreneurship (and the projects on which they are based) test all three foundational migration myths. First, they examine claims that South Africa and, by extension, the South African informal sector is being “swamped” by “foreigners.” Gauteng is generally reckoned to be the province with the largest number of migrants. SSA (2016) calculates that the province had 801,308 non-South African-born residents in 2016 or 6% of the total population. In most other provinces, the proportion is 1-3%. A survey of the Gauteng informal sector found that over 80% of small businesses in the informal sector were still owned by South Africans (Peberdy, 2015). Certainly, there are more migrants in the informal sector in 2017 than there were in 2000 or 2010 and there are more migrant than South African entrepreneurs in parts of some cities (and vice-versa in other areas). Because refugees, asylum-seekers and other migrants are shut out of sectors of the formal labour market (as they have been in the security industry, or because employers will not accept Section 22 and 24 permits as bona fide legal entitlements to work), the numbers working informally have inevitably swelled. Working in the tough and unrelenting environment of the informal sector is not necessarily the first choice of the many migrants and refugees who are more than qualified for formal employment. Second, these papers provide a more nuanced and humanised picture of so- called foreigners and their activities in the informal sector than is suggested by the notion that they are all desperate survivalists fleeing poverty and strife in other African countries. While a significant number of refugees say that they work in the informal sector because they cannot get formal-sector jobs, many also exhibit strong entrepreneurial characteristics, orientation and ambition. This emerges both in their attitudinal profile of suitability for running their own businesses and in the innovative business strategies that they have developed. Perhaps most surprisingly, given the blanketing clouds of myth, refugees are far more likely than their South African counterparts to want to viii contribute to the development of South Africa through running an informal business. As the two papers in this collection focused on informal cross- border traders clearly show, there is also significant entrepreneurial orientation amongst these primarily female informal sector participants. Like their less mobile counterparts, cross-border traders working the transportation corridors between Harare, Maputo and Johannesburg are dedicated and enterprising individuals with high levels of motivation. Third, the papers in this collection take issue with the predominantly negative stereotyping of migrant entrepreneurs to justify and promote harsher sanctions against them. National, provincial and local level campaigns by the police and army – with telling names such as Operation Fiela (‘sweep’ in English), Hardstick and Clean Sweep – aim to create city environments swept of informal entrepreneurship by migrants and refugees. However, in case such “operations” are viewed (as they were by the courts) as cynical and unconstitutional, it also seems necessary to target informal entrepreneurs with the most negative language possible. The IMC, for example, asserts that the impact of foreign national domination “is compounded by business models used by migrants to discourage competition such as forming monopolies, evading taxes, avoiding customs and selling illegal and expired goods” (NA, 2015). Or again, the previous Minister of Home Affairs pointed to the seriousness of “the dynamics of migration, crime, drugs, prostitution, fraud and unfair labour practices” (Gigaba, 2017). The most efficient, and misleading, way to highlight the supposedly negative impact of migrant and refugee entrepreneurship is to directly associate it with criminality. The papers in this collection instead show that crime is a very real business challenge for refugee and migrant entrepreneurs. Their many positive economic contributions are made despite, not because of, crime. In addition to replicating the foundational mythology of migration, South Africa has several of its own homegrown myths about migrant entrepreneurs in the informal sector. First, there is the myth, articulated by the Minister of Small Business Development, that refugees enjoy a competitive advantage over South Africans because trading and vending is ingrained and instinctual. She claimed in an interview that foreign spaza shop owners are “better at running shops than the local owners – they have a great network system. And also that’s how they live. From the moment they are born, they are introduced to trade. Their mothers, uncles – everyone trades. They start at an early age…How are they able to make it when our people can’t? It is because they know business. It is in their blood” (Zwane, 2014). Only a small proportion of ix the refugees and other migrant business owners interviewed for these studies had experience running an informal sector enterprise before they came to South Africa, however. The success of refugees and migrants in the informal economy is not because entrepreneurship is innate, as the minister claimed, but rather a function of hard work, innovation and competitive business strategies. The tools of the trade were acquired on site. The second homegrown myth is that South Africans are bad informal entrepreneurs unable to compete with the “foreigner.” Many South Africans have decided to opt out of the spaza business because it requires less work and is more remunerative to rent their property to refugees and other migrants. That said, there is still stiff competition in the informal economy between South Africans and non-South Africans, as there is between South Africans and other South Africans and between non-South Africans of one nationality and another (Piper and Charman, 2016). There is also evidence that antagonism towards refugee business owners is stoked by South African trader associations who would rather not compete openly and fairly. However, as the papers in this collection demonstrate, South Africans are certainly not incompetent or incapable of competing. They, too, have their own competitive strategies and carve out market niches, especially within the informal food economy, where they operate with success. Third, there is the myth that the post-2008 upsurge in violent attacks against migrant-owned businesses in the informal sector is purely the work of fringe criminal elements. According to the IMC, “the main causes of the violent attacks (are) criminal actions that started with stealing of goods from foreign owned spaza shops by South African criminals who are often drug addicts” (NA, 2015). This is certainly called into question by the large crowds, including many parents and children, that surround refugee and migrant- owned shops during episodes of collective looting. Whenever there is a particularly vicious flare-up, or after a spate of looting and murder, government ministers and Cabinet are quick to proclaim that crime, not xenophobia, is the driver (Misago, 2016). Indeed, there has been a persistent strain of xenophobia denialism ever since former president Thabo Mbeki proclaimed in 2008 that South Africans were not xenophobic. Most recently, the IMC has vigorously denied the existence of xenophobia in the country: “South Africans (are) not xenophobic; no evidence was found to indicate that South Africans were xenophobic” (NA, 2015). The IMC has clearly not acquainted itself with the large body of attitudinal research that proves precisely the opposite (Crush et al., 2013; Gordon, 2016, 2017). As the paper x in this collection on Zimbabweans in the informal sector suggests, migrants themselves have no hesitation in naming xenophobia for what it is. They say South Africans are the owners of xenophobia, refer to violence against their small businesses as “the xenophobia”, and recount numerous harrowing incidents that are either triggered by xenophobia or, as in the case of service delivery protests against government, quickly turn xenophobic. Finally, there is the argument that xenophobia is not a factor because violence and vulnerability to crime comes equally to all who work in the informal sector. The suggestion here is that South Africans and non-South Africans are equally vulnerable and to single out violence against the latter is to ignore the identical plight faced by South Africans. Some statistical evidence on relative vulnerability has been advanced in support of this argument (Piper and Charman, 2016). The paper in this collection on business risk presents alternative evidence. While it is true that South Africans in the informal sector are not immune from crimes such as robbery and looting, the prevalence is significantly higher amongst refugees and in different parts of the country. Many refugees in towns in Limpopo had started businesses in large cities but moved to what they assumed would be a safer operating environment after the levels of violence and xenophobia became overwhelming. It is safer, but it is certainly not safe. South Africa is the process of a major overhaul of its migration and refugee protection systems. In the case of migration for work, there are some grounds for optimism in the recognition that South Africa is integrated into, and benefits from, a regional SADC economy. The White Paper on International Migration contains proposals for a streamlined skills-based immigration policy and a system of work and trading permits for SADC citizens. In the case of cross-border traders, the proposed system seems unnecessarily bureaucratic unless the intention is to try to control numbers through quotas. This is unworkable and it would be far simpler to allow visitors to the country to both buy and sell goods. Whether proposals to issue quota-based permits to migrants from other SADC countries will ever see the light of day is questionable given the current upsurge in public complaints, now supported by the unions, that migrants are taking jobs from South Africans. However, it is the proposals for restructuring South Africa’s post-apartheid refugee protection regime that are likely to have the most far-reaching impact on livelihoods in the informal economy. As the first paper in this special issue argues, the intention is to make South Africa undesirable by moving from an urban integration towards a border encampment model, denying asylum- xi seekers their current right to pursue a livelihood while waiting for a hearing, and ensuring that no refugee ever qualifies for permanent residence, no matter how long they have been in the country. The proposed changes rest on the shaky foundation of myth. This issue is a contribution to the ongoing task of testing anti-migration mythologies with fact. It comprises a selection of papers from three projects implemented by the Southern African Migration Programme (SAMP) (see also www.samponline.org): (a) the Growing Informal Cities project, an IDRC- funded partnership between the African Centre for Cities (ACC), the Gauteng City Regional Observatory (GCRO), Eduardo Mondlane University and Queen’s University; (b) the Migrants in Countries in Crisis (MICIC) project between the University of the Western Cape, the University of Limpopo, the Balsillie School of International Affairs, the International Centre for Migration Policy Development and the International Migration Institute at Oxford University; and (c) the UNHCR-funded Refugee Economic Impacts project, a partnership between researchers at the Universities of Cape Town, Limpopo and Western Cape and the International Migration Research Centre. Prof Jonathan Crush* CIGI Chair in Global Migration and Development, International Migration Research Center Balsillie School of International Affairs Waterloo, Canada jcrush@balsillieschool.ca * A recipient of numerous academic awards and honours, Professor Crush founded the Southern African Migration Programme (a consortium of Canadian and African researchers) in 1997. In 2005, he co-founded the African Food Security Urban Network; a network of Canadian and African universities, NGOs and municipal governments. In 2014, he established the Hungry Cities Partnership; a research, policy and capacity-building network with partners in Canada, Mexico, Jamaica, South Africa, Mozambique, Kenya, India and China. He also works with the African Centre for Cities at the University of Cape Town, where he holds an Honorary Professorship. He has authored or edited 14 books, over 150 journal articles and book chapters and numerous published reports for governments, international agencies, and civil society organizations. mailto:jcrush@balsillieschool.ca xii Acknowledgements The contributors to this Special Issue wish to thank the following for their support: the International Migration Research Centre, the International Development Research Centre, the European Union and the UNHCR Geneva. References Africa Check (2017). “How Many Foreigners Really Live in South Africa?” Business Tech, 14 January 2017. Arcarazo, D. A., & Wiesbrock, A. (Eds.). (2015). Global Migration: Old Assumptions, New Dynamics [3 volumes]. ABC-CLIO. Basardien, F., Parker, H., Bayat, M., Friedrich, C. and Sulaiman, A. (2014). Entrepreneurial Orientation of Spaza Shop Entrepreneurs: Evidence from a Study of South African and Somali-Owned Spaza Shop Entrepreneurs in Khayelitsha. Singaporean Journal of Business Economics and Management Studies, 2: 45-61. Blinder, S. 2015. Imagined Immigration: The Impact of Different Meanings of ‘Immigrants’ in Public Opinion and Policy Debates in Britain. Political Studies, 63(1): 80–100. Cape Times (2016). Gigaba Looking to Lighten Asylum Burden Cape Times 22 June 2016. Crush, J. and Ramachandran, S. (2015). “Migration Myths and Extreme Xenophobia in South Africa” In D. Acosta and A. Wiesbrock (eds.), Global Migration: Myths and Realities Vol. 3 (Santa Barbara: Praeger, 2015), pp. 71- 96. Crush, J., Chikanda, A. and Skinner, C. (Eds.). 2015. Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa: IDRC. Crush, J., Ramachandran, S. and Pendleton, W. 2013. Soft Targets: Xenophobia, Public Violence and Changing Attitudes to Migrants in South Africa after May 2008. SAMP Migration Policy Series No. 64, Cape Town. xiii de Haas, H. 2008. The Myth of Invasion: The Inconvenient Realities of African Migration to Europe. Third World Quarterly, 29: 1305-1322. Fairlie, R. and Lofstrom, M. 2015. Immigration and Entrepreneurship. CESifo Working Paper Series No. 5298. Available at http://bit.ly/2wX3Y27. Gigaba, M. 2017. Media Statement by Minister Gigaba on Tensions Between Citizens and Foreign Nationals. From At: http://bit.ly/2v8rN4Y. Gordon, S. 2016. Welcoming Refugees in the Rainbow Nation: Contemporary Attitudes Towards Refugees in South Africa, African Geographical Review, 35: 1-17. Gordon, S. 2017. A Desire for Isolation? Mass Public Attitudes in South Africa Toward Immigration Levels. Journal of Immigrant & Refugee Studies, 17: 18- 35. Kloosterman, R. and Rath, J. 2003. Immigrant Entrepreneurs: Venturing Abroad in the Age of Globalization. New York: University of New York Press. Hellwig, T. and Sinno, A. 2017. Different Groups, Different Threats: Public Attitudes Towards Immigrants, Journal of Ethnic and Migration Studies, 43(3): 339-358. Jinnah, Z. 2010. Making Home in a Hostile Land: Understanding Somali Identity, Integration, Livelihood and Risks in Johannesburg. Journal of Sociology and Social Anthropology, 1(1-2): 91-99. Misago, J-P. 2016. Responding to Xenophobic Violence in Post-Apartheid South Africa: Barking Up the Wrong Tree? African Human Mobility Review, 2(2): 443-467. National Assembly (NA). 2015. Report of the Ad Hoc Joint Committee on Probing Violence Against Foreign Nationals. Dated 19 November 2015. Cape Town: National Assembly and National Council of Provinces. Peberdy, S. 2015. Informal Sector Enterprise and Employment in Gauteng. GCRO Data Brief No. 6, Johannesburg. Peberdy, S. 2016. International Migrants in Johannesburg’s Informal Economy, SAMP Migration Policy Series No. 71, Cape Town. xiv Peberdy, S. 2017. Competition or Cooperation? South African and Migrant Entrepreneurs in Johannesburg. SAMP Migration Policy Series No. 75, Cape Town. Piper, L. and Charman, A. 2016. Xenophobia, Price Competition and Violence in the Spaza Sector in South Africa. African Human Mobility Review, 2(1): 332- 361. Parliamentary Monitoring Group (PMG). 2015. Inter-Ministerial Committee Briefing. At http://bit.ly/2v7Mfmg. Rogerson, C. 2016. South Africa’s Informal Economy: Reframing Debates in National Policy. Local Economy, 31: 172-86. Radipere, N. 2012. An Analysis of Local and Immigrant Entrepreneurship in the South African Small Enterprise Sector (Gauteng Province). PhD Thesis, Pretoria: University of South Africa. Stupart, R. 2016. Is South Africa Home to More Than a Million Asylum Seekers? The Numbers Don’t Add Up. Africa Check 15 August 2016. Statistics South Africa (SSA). 2016. Community Survey 2016. Pretoria: Statistics South Africa. Tawodzera, G., Chikanda, A., Crush, J. and Tengeh, R. 2015. International Migrants and Refugees in Cape Town’s Informal Economy. SAMP Migration Policy Series No. 70, Cape Town. United Nations High Commissioner for Refugees (UNHCR). 2016. UNHCR Global Trends: Forced Displacement in 2015. Geneva: UNHCR. United Nations High Commissioner for Refugees (UNHCR). 2017. UNHCR Global Trends: Forced Displacement in 2016. Geneva: UNHCR. Valentino, N., Brader, T. and Jardina, A. 2013. “Immigration Opposition Among U.S. Whites: General Ethnocentrism or Media Priming of Attitudes About Latinos? Political Psychology, 34 (2): 149–166. Zwane, T. 2014. Spazas: Talking Shop is Good for Business. Mail & Guardian, 6 November 2014. AHMR, Vol.3 No2, May-August 2017 – Special Issue 751 Benign Neglect or Active Destruction? A Critical Analysis of Refugee and Informal Sector Policy and Practice in South Africa Jonathan Crush*, Caroline Skinner** and Manal Stulgaitis*** Abstract To fully comprehend the disabling policy environment in which refugees in South Africa attempt to carve out a livelihood, it is important to analyse two largely independent but overlapping streams of policy-making. This paper first examines the post-apartheid refugee protection regime and traces how and why a generous right-based approach has been progressively comprised by growing restrictionism, exclusion and bureaucratic ineptitude. The 2017 Refugees Amendment Act and White Paper on International Migration represent the culmination of this process. While both are probably unimplementable and will be the subject of numerous court challenges, they can be seen as a major retreat and an increasing failure to protect. The second part of the paper traces the history of national and municipal informal sector governance since the early 1990s. Since so many refugees are forced or choose to work informally, the uncertainty and confusion this history has produced is of particular relevance. Refugee entrepreneurs have regularly been the victims of general and targeted informal sector eradication campaigns. Therefore, there is a fundamental contradiction between a refugee protection policy that demands self-reliance from refugees and informal sector policies that undermine self-reliance at every turn. Keywords Refugee protection, Refugees Act, detention centres, informal sector, Home Affairs. * International Migration Research Centre, Balsillie School of International Affairs, Waterloo, Ontario, Canada N2L 6C2 Email: jcrush@balsillieschool.ca ** African Centre for Cities, University of Cape Town, Cape Town, South Africa *** Independent Consultant, Cape Town AHMR, Vol.3 No2, May-August 2017 – Special Issue 752 Introduction The past decade has seen South Africa’s apparently generous asylum and refugee system flounder, characterised by ever-growing wait times for status decisions, increased barriers for application and renewal of permits, and growing disregard for refugee law and court orders (Amit, 2011, 2012, 2015; Johnson, 2015; Polzer Ngwato, 2013). The South African Government has increasingly taken the position that the country’s post-apartheid refugee protection legislation is far too generous and needs to be revised in the direction of more restrictions and fewer rights. This has resulted in major changes to the seminal 1998 Refugees Act in the form of the 2017 Refugees Amendment Act, and the promise of a new approach to migration and refugee protection in the Green Paper and White Papers on International Migration in South Africa. These developments seek to align South Africa with the more exclusionary and restrictive treatment of asylum-seekers and refugees in many other countries (Fiddian-Qasmiyeh et al., 2014). In the absence of material support from the government or the United Nations High Commission for Refugees (UNHCR), one of the primary livelihood strategies of asylum-seekers and refugees has been to create work for themselves in the informal sector. The policy environment in which refugee entrepreneurs run informal businesses on the streets and in residential areas is framed by policy and legislation at national, provincial and local levels. Those working in the informal sector face an ambiguous policy environment that has occasionally supported but largely ignored – and at times actively destroyed – informal sector livelihoods and those of migrant and refugee businesses in particular. To fully understand the disabling policy environment within which migrants and refugees establish and operate their enterprises in the South African informal sector, we need to bring together these two streams of analysis. Therefore, this paper begins with a discussion of South Africa’s changing refugee policies and practices and traces the erosion of the protective and progressive refugee policy approach that characterised the immediate post- apartheid period (Handmaker et al., 2011). In the context of the refugee livelihoods which are central to the country’s local integration approach, it is AHMR, Vol.3 No2, May-August 2017 – Special Issue 753 vital to have effective and efficient recognition of refugee status and an accompanying basket of rights to support survival through employment or entrepreneurship (Jacobsen, 2005). The paper then reviews post-apartheid informal sector policy and practice. While the informal sector largely fell through policy gaps in the first ten years of democracy, the analysis suggests that increasing attention has been paid to this issue in recent years. While there have been longstanding tensions between foreign and South African informal sector operators, from 2012 an overtly anti-foreign migrant sentiment has been expressed in official policy and practice. The paper underlines the need for both a rights-based asylum system and more progressive policy towards the informal sector. Refugee entrepreneurs and service providers agree that obtaining refugee status is key to enabling refugee entrepreneurship and sustainable livelihoods. Despite the obstacles put in their way, refugee business owners appear to be succeeding, at least on par with their South African counterparts. This suggests that secure status and the associated basket of rights are serving refugee entrepreneurs in the context of sustainable livelihoods. While these arguments should mark a road map to successful local integration, the South African Government continues to pile on administrative and logistical barriers to the asylum process and prospective refugees. These measures add to the ambiguity around migration management in the country and complicate the prospects for refugees to provide for themselves in a safe and sustainable manner. The paper is based on a review of media and official government sources, published and grey literature, and extended interviews with key informants in Cape Town, Limpopo and Gauteng during 2015. A total of 30 in-depth interviews were conducted including with sector researchers, refugee and diaspora associations, refugee rights NGOs, law enforcement, the City of Cape Town and Western Cape Governments and international organisations (such as the IOM and UNHCR). Interviews were also conducted with national government departments including Home Affairs, Labour and Small Business Development. AHMR, Vol.3 No2, May-August 2017 – Special Issue 754 Refugee Policy and Practice South Africa is a signatory to the 1951 United Nations Convention on the Status of Refugees and the 1969 OAU Convention Governing the Specific Aspects of Refugee Problems in Africa. Since the end of apartheid, it has built a reputation as a protective and progressive refugee receiving country. South Africa’s 1998 Refugees Act integrated international refugee protections into domestic law and exceeded international standards in important respects. The Act made generous allowances for freedom of movement, access to health and education services, some social protection and the right to work. Most notably, the Act embraced local integration over encampment, which was a progressive choice in Africa at the time (Handmaker, 2001). Underpinning the Act was a political and ideological approach which posited that refugees were permitted and fully expected to temporarily integrate into the host country and benefit from all attendant protections and rights granted to citizens by the Constitution. Refugees who had been in the country for five years and were still unable to return were entitled to apply for permanent residence. In return for these progressive policies, both government and the UNHCR incurred minimal costs in providing material support for asylum-seekers and refugees in the country. In the years since these somewhat idealistic beginnings, South Africa has seen a distinct deterioration in the rights-based approach to refugee protection (Amit, 2011, 2015; Handmaker et al., 2011; Igglesden & Schreier, 2011; Landau, 2006). International praise for South Africa’s liberal approach has been eroded by chronic processing delays, poor and ill-informed adjudication and the corruption and mismanagement which has become endemic to the asylum process. South Africa has one of the longest asylum adjudication periods in the world, in some cases it lasts many years. After 2007, economic free fall in Zimbabwe imposed significant pressure on the asylum system as migrants from Zimbabwe moved in significant numbers to South Africa (Crush & Tevera, 2010; Crush et al., 2015). Arguably, South Africa’s failure to anticipate and account for the entry of Zimbabweans en masse created a situation whereby the asylum system became contorted into something of a ‘catch all’ for generalised migration into South Africa. It is worth noting, AHMR, Vol.3 No2, May-August 2017 – Special Issue 755 however, that part of the reason for the dramatic increase in asylum-seeking from Zimbabwe was the establishment of a Refugee Reception Office in Musina, close to Zimbabwe, specifically to speed up the issue of asylum-seeker permits. Those close to the process argue that the demand for a place in South Africa’s asylum queue, with its attendant right to work, rendered the system ineffectual at conducting legitimate asylum adjudication. In government discourse, this has translated into a strident denunciation of “bogus” claimants and “abuse of the system” by economic migrants (Mabuza, 2016). The official position is that 90% of asylum-seekers are economic migrants, a figure apparently based on the rate of acceptance of refugee claims (DHA, 2016: 29). However, this conclusion is a non-sequitur, given the well-documented delays in adjudication, the arbitrariness of many decisions, and the practice of adjudicating claims by country or origin and not the personal experiences of the individual claimant (Amit, 2012). The reasons for this shift from rights and protections towards exclusion and control are seen by some as the inevitable consequence of life in a country where the majority still struggle to meet basic needs and there is competition for scarce public resources such as education, health care and shelter, as well as employment and other livelihood opportunities (Hassim et al., 2008). In this zero sum game, every advantage that a refugee or asylum-seeker enjoys necessarily disadvantages a South African. However, Gordon (2016) shows that South Africans do not oppose refugee protection for reasons of economic self-interest. Rather, their opposition, and that of many policy-makers, is further evidence of the deeply xenophobic character of South African society, with its attendant failure to acknowledge the positive economic, social and cultural contribution that refugees and asylum-seekers make to the country (Crush et al., 2015). As Gordon (2016: 1) argues, “public animosity towards refugees in South Africa has motivated anti-immigrant riots, violence, and prejudice which has negatively impacted on refugee protection.” Landau and Duponchel (2011: 19) further suggest that “a protection strategy dedicated to maximizing refugees’ freedom and integration may prove politically untenable in an era of pronounced anti-immigrant hostilities.” AHMR, Vol.3 No2, May-August 2017 – Special Issue 756 Government argues that the breakdown of the asylum system in South Africa, and the need for a new approach, is because it has been overwhelmed by economic migrants. For a while, it argued that there were a million unprocessed asylum-seekers in the country, a figure that was uncritically reproduced by the UNHCR and the media. The fact that the system itself was badly under-resourced, staffed by small numbers of poorly-trained officers and riddled with corruption was less-often acknowledged. The Green Paper now admits that this figure was erroneous and that in mid-2015, South Africa had only 78,339 active asylum-seekers (Section 22 permit holders) and had issued 119,600 refugee (Section 24) permits since 2002 of which only 96,971 were still active. Having previously claimed that the country was one of the largest refugee destinations in the world, this admission represented a considerable climb-down by the Department of Home Affairs (DHA) (Stupart, 2016). Therefore, rather than being motivated by a need to address an overrun system, the recent developments are better interpreted as a response to the desire of South Africans to make the country an undesirable destination for asylum-seekers by narrowing refugee rights and imposing additional limitations on the ability of refugees to find safety and security in South Africa. Within government, four inter-connected strategies have been developed to achieve these ends. All are embodied in recent administrative decisions including the 2016 Refugees Amendment Act (which passed Parliament in March 2017 and currently awaits Cabinet approval before becoming law) and the Green Paper and White Papers on International Migration in South Africa (DHA, 2017). The first strategy has been to move away from an integration towards an encampment model of protection. The DHA has now publicly declared its intention of establishing what it calls Asylum-Seeker Processing centres away from the country’s urban areas (which is widely and correctly seen by critics as a euphemism for encampment) (DHA, 2016). Plans have apparently been drawn up for the location and physical infrastructure of detention centres close to the Zimbabwe and Mozambique borders (Mah & Rivers, 2016) and, according to one source, construction has begun on a detention centre at Lebombo. The White Paper notes that these detention centres will accommodate all asylum-seekers during their status determination process (DHA, 2016: 65). AHMR, Vol.3 No2, May-August 2017 – Special Issue 757 Following from the proposal that asylum-seekers should be kept in secure facilities while their claims are adjudicated, freedom of movement and integration into local communities (as at present) would be halted. So-called “low risk” asylum-seekers might be released into the care of national or international organisations and family or community members. However, asylum-seekers would not have the automatic right to work or study “since their basic needs will be catered for in the processing centres” (DHA, 2016: 68). To try and pre-empt the inevitable, and justifiable, criticism towards South Africa’s introduction of a policy of encampment, the Green Paper awkwardly asserts that “these centres should not be considered as contrary to the policy of non-encampment but as centres for mitigating security risks posed by irregular migration. Only refugees and not asylum seekers will be allowed to integrate into communities” (DHA, 2016: 66). The protestation that this somehow represents a continuation of the country’s non-encampment policy is disingenuous in the extreme. Under this policy, asylum-seekers will be sequestered in detention centres for the duration of the adjudication process, which is unrealistically envisioned to be a 60 to 90 day process. Whether that process takes place within the proposed period or not, during that time asylum-seekers will be fully dependent on government or the UNHCR for food, shelter, health care, education and other basic needs. Refugee service providers overwhelmingly agree that this is a no-win situation: if services to refugees are better than services available to South Africans, South Africans will cry foul and poor South Africans may even present as refugees in order to access direly needed services. On the other hand, if services in the detention centres are worse, South Africa will suffer the criticism of the international community for failing in its duty to respect the rights of refugees within its borders. The cost of constructing and maintaining camps for large numbers of asylum-seekers will be massive and the UNHCR has indicated that it will not underwrite detention costs despite appeals from the South African Government (which has previously constrained the UNHCR from offering material assistance to asylum-seekers.) If South Africa faced an influx of asylum-seekers in the future akin to the over 200,000 Zimbabweans between 2006 and 2009, it is hard to see how these centres would even begin to cope. AHMR, Vol.3 No2, May-August 2017 – Special Issue 758 The second major strategy is to steadily put in place procedural, administrative and logistical hurdles that complicate refugees’ already tenuous status and sustainability in the belief that this will act as a disincentive to asylum-seekers coming to the country and make life extremely difficult for them if they do come. The most obvious example is the DHA cutting the number of Refugee Reception Offices in the country in half, closing busy offices at Crown Mines (Johannesburg), Cape Town and Port Elizabeth Refugee Reception Offices (RRO). Only three RRO’s remained open: in Musina, Durban and Pretoria. The closure was “not merely a technical, operational decision, but one which impacts on the basic principles of the asylum system, namely access (for initial applications, renewals, status determination interviews and appeals) and administrative efficiency and fairness” (Polzer Ngwato, 2013). Legal challenges have produced contradictory outcomes. The Supreme Court of Appeal (SCA) ordered the re-opening of the Port Elizabeth RRO, a judgment which the Department has been very slow to implement. In contrast, the Cape Town High Court decided that the Cape Town RRO could remain closed. The 2016 Refugees Amendment Act gives the Director General of Home Affairs the power to “dis-establish” any RRO and to force a whole category of asylum- seeker (defined in terms of country of origin or “a particular gender, religion, nationality, political opinion or social group”) to report at a designated RRO. Second, and partly as a result, the administrative requirement that asylum- seekers should renew their permits every one to six months at one of the RROs (rather than an ordinary Home Affairs Office) imposes considerable financial and other hardship. The period granted on renewal appears to be entirely arbitrary and, according to some refugees, depends on how much they are willing to pay under the table. Individuals and families who have found safety, shelter, kin, communities, work or school in other parts of the country are forced to travel to one of the RROs to ensure that they remain in status. Furthermore, wait-times for receiving or renewing a permit can be considerable. The scene outside the Marabastad (Pretoria) RRO was described by one organisation as follows: They go to Home Affairs to Marabastad to get their asylum permit. So they come to Pretoria, there are queues and queues, never-ending queues. And AHMR, Vol.3 No2, May-August 2017 – Special Issue 759 then each country has a day. So now you have come to the Home Affairs office and it’s not your turn, your turn only comes after 4 days. And then you are told that if you have ZAR 2,000, these officials walk around and if you have money, you give them the money and go to the front of the queue. If you don’t have money, then you are right at the back. And then you have to come back the following week on the day of your country. In a situation where asylum-seekers are almost exclusively self-supporting, without the assistance of government or the international community, they sacrifice valuable time and money, risk jeopardising employment, and travel with or leave small children behind when they do so. Under the 2016 Act, failure to renew an asylum-seeker permit within one month of expiry now leads to automatic revocation of status, forfeiture of the right to renewal and treatment as an “illegal foreigner” under the Immigration Act (that is, arrest and deportation). Asylum-seekers whose claims are refused are also to be treated as “illegal foreigners.” An asylum-seeker with an expired permit is also guilty of an offence and liable to a fine and imprisonment of up to five years or both. Any individual or group of asylum-seekers or refugees can now be arrested and deported on the vaguely-worded grounds of “national interest, national interest or public order.” These provisions would seem draconian even if the system was efficient and transparent, a description that certainly does not apply here. The third strategy is to undercut court judgements that have affirmed the right of asylum-seekers and refugees to employment and self-employment, the essence of the post-apartheid model of refugee integration. The 2016 Act explicitly seeks to overturn a judgment that permitted asylum-seekers to work in South Africa while awaiting adjudication of their claim. Prior to the future holding of asylum-seekers in detention centres (as envisaged by the Green Paper), the onus will now be on “family and friends” to support the asylum- seeker for their first four months in the country. If such support is not available, the UNHCR and NGOs are permitted to provide “shelter and basic necessities.” In both situations, the asylum-seeker is prohibited from working, while government assumes no responsibility for their care and protection. The Standing Committee on Refugee Affairs is also now empowered to unilaterally AHMR, Vol.3 No2, May-August 2017 – Special Issue 760 decide under what conditions asylum-seekers may work or study. If they are permitted to do so, they are required to provide a letter from the employer or institution within fourteen days from the date of employment or enrolment. The employer or institution can be fined ZAR 20,000 if they fail to provide the documentation in the prescribed period. The right to work can also be revoked by the Director General. The Act says nothing about the right to access informal work or self-employment, a key component of earlier court judgments. The fourth strategy which will directly affect those with refugee status is to try and ensure that protection is ever only temporary by making it extremely difficult for refugees to progress to permanent residence and eventual citizenship. The 1998 Act stated that refugees were entitled to apply for permanent residence after five years of continuous residence in South Africa and refugees “of good and sound character” could be issued with permanent residence permits irrespective of the length of sojourn in the country. This is clearly one reason why the maximum length of a refugee permit was set at four years. It is unclear how many long-term refugees have tried to access or been granted permanent residence. DHA (2016: 29) notes that there were 1,175 applications between mid-2014 and early 2016, but does not say how many were successful. In 2013, the DHA stripped all refugees from Angola of their refugee status irrespective of length of residence and then issued them with two-year non- renewable temporary residence permits (called Angolan Cessation Permits or ACPs) (Carciotto, 2016). In 2016, after extensive negotiations with the DHA, it was agreed that former Angolan refugees could apply for permanent residence. The Western Cape High Court then issued an order by which all former Angolan refugees with expired ACPs could apply for permanent residence. In February 2017, the Scalabrini Centre of Cape Town submitted 1,757 applications on behalf of Angolan refugees to the DHA. The successful court action not only prevents the summary deportation of former Angolan refugees but potentially provides an important precedent for future cases of cessation. However, the 2016 Act gives the Minister of Home Affairs new powers to issue an order which ceases recognition of an individual refugee or AHMR, Vol.3 No2, May-August 2017 – Special Issue 761 group of refugees or to revoke refugee status, without the obligation to provide any justification for such an action. The right of a refugee to apply for permanent residence has also been extended from five to ten years. A final policy issue of relevance to this paper is asylum-seeker and refugee access to financial services. As other papers in this special issue show, refugee entrepreneurs have very limited access to start-up capital and other loans from formal banking institutions. Prior to 2010, the Financial Intelligence Centre Act (FICA) also prohibited refugees and asylum-seekers from opening bank accounts in South Africa. That policy was later eased but among FICA’s anti-money laundering provisions is a requirement for banks to verify the identity of persons wishing to open a bank account. Banks are given wide discretion as to how they implement the requirements, with the result that many refuse to open bank accounts for refugees and asylum-seekers based on the fear that they will not be able to correctly validate refugee identity documents. In response to legal action, the DHA and FICA reached an agreement for the former to provide banks with means to verify the authenticity of refugee and asylum permits issued by the DHA. In practice, opening a bank account remains a challenge for refugees and asylum-seekers, with banks remaining distrustful of the various types of documentation issued by Home Affairs, viewing it as less formal or secure than a South African national identity card. At those institutions where it is possible for a refugee to open a bank account, there have been instances of refugees and asylum-seekers having their assets frozen when identity documents have not been renewed on time, when identity documents change, or when the DHA has failed to respond to verification enquiries in a timely manner. Service providers report that ongoing challenges around the banking sector present a very real risk to families struggling to meet basic survival needs. A frozen bank account raises grave protection concerns, threatening the ability to pay rent, buy food, care for children and even cover costs for long distance travel for the purposes of renewing status documentation (Washinyira, 2012). The end result, noted by several interviewees, is that in the townships and informal settlements where many refugees own and operate their businesses, shops and homes offer a ready target for criminals who are well aware that cash is AHMR, Vol.3 No2, May-August 2017 – Special Issue 762 likely stored on the premises. This dynamic puts refugee families and livelihoods at risk not only of loss of profits and other assets, but also of violence and trauma when break-ins and robberies occur. Informal Sector Policy and Practice As noted in the introduction to the paper, due to the lack of formal job opportunities, international migrants have little choice but to work in the informal sector, suggesting that informal sector policy and practice shapes livelihood opportunities. This section critically analyses this policy arena in the post-apartheid period at national and provincial level and in the three largest metropoles. The apartheid state was particularly averse to informal sector activities. Lund (1998: 6) reminds us that apartheid policies controlled where black South Africans could live, what they could own and, through job reservation policies, what work they could do. Black South Africans were forbidden by law from engaging in manufacturing businesses and access to business premises was strictly regulated to prevent them from operating businesses in ‘white’ areas (Manning & Mashigo, 1994). In the mid-1980s, influx control laws became increasingly unenforceable and were abolished in 1986. In 1987, the National White Paper on Privatisation and Deregulation introduced a more tolerant approach to black small business as part of a broader new economic philosophy informed by the Reagan-Thatcher era of deregulation. The change of attitude culminated in the Businesses Act 71 of 1991 (which repealed numerous restrictive laws and secured a more liberal approach to business licensing, premises and hours for both formal and informal businesses). This legislation was a key measure for removing barriers to the operation of informal activities and was, in effect, a complete reversal of the apartheid approach. This piece of legislation is still in place today. Subsequent to the passing of the Act, informal-sector activities increased in all cities and towns. Local authorities, however, complained that they were unable to cope, particularly with trading in public spaces. This led in 1993 to the Businesses Amendment Act 186 of 1993 (RSA, 1993). The Act gave provinces the discretion to develop their own legislation and allowed local authorities to formulate street-trading bylaws, outline what was allowed and AHMR, Vol.3 No2, May-August 2017 – Special Issue 763 declare restricted and prohibited trade zones. Since then, local authorities across the country have promulgated such bylaws, evidently near mirror- images of each other. In all of the major metropoles, for example, the sanction in the case of violation was inappropriately criminalised – either a fine or imprisonment, suggesting a punitive approach to street-trader management. At a national level, the 1995 White Paper on the Development and Promotion of Small Businesses was one of the first economic policy initiatives of the post- apartheid government. The White Paper and the legislation that stemmed from it – the National Small Business Act of 1996 – both acknowledge survivalist and micro-enterprises as components of small business, thus making them, on paper, beneficiaries of government support. However, both documents are silent on the specific needs of these smaller players, suggesting the role played by this group in the small business/informal sector was not seen as a critical issue. Neither refer to foreign migrants or refugees. To implement this new approach the Department of Trade and Industry (DTI) set up the Ntsika Enterprise Promotion Agency as a facilitation and promotion body for small businesses and Khula Finance to secure small-business access to financial services. They supported the establishment of a country-wide network of Local Business Development Centres (LBDCs) to provide non- financial support to SMMEs. In 2004, a comprehensive review of the impact of the post-apartheid government’s small, medium and micro enterprise (SMME) programmes concluded that “existing government SMME programmes largely have been biased towards the groups of small and medium-sized enterprises and to a large extent have by-passed micro-enterprises and the informal economy” (Rogerson 2004: 765). Devey et al. (2003; 2008) evaluated the national-government skills-development system, concluding similarly that those working in the informal-sector had “fallen into the gap” between small businesses and the unemployed. These findings were echoed in Budlender et al.’s (2004) informal sector budget analysis and confirmed by survey data conducted with informal sector operators in Johannesburg and Durban (Chandra & Rajaratnam, 2001; Skinner, 2005). In 2003, President Mbeki publicly advocated for the idea of the ‘second economy’ in an address to the National Council of Provinces. Mbeki’s (2003) AHMR, Vol.3 No2, May-August 2017 – Special Issue 764 second economy was characterised by “underdevelopment, contributes little to GDP, contains a large percentage of our population, incorporates the poorest of our rural and urban poor, is structurally disconnected from both the first and the global economy, and is incapable of self-generated growth and development.” According to Mbeki (2003), Cabinet had resolved that the development of the second economy required “the infusion of capital and other resources by the democratic state to ensure the integration of this economy within the developed sector.” This would be achieved in a number of ways, including the development of SMMEs and cooperatives, the expansion of micro-credit and skills development. Although the concept was not new, its application to South Africa was a watershed moment for national informal sector policy. For the first time since the end of apartheid, the informal sector was given a policy profile. The whole idea of the second economy elicited a flurry of criticism (Devey et al., 2006; du Toit and Neves, 2007). According to Devey et al. (2006: 242), second economy arguments were based on the premise that “the mainstream of the economy is working rather well, and government action is needed to enhance the linkages between the first and second economy and where appropriate to provide relief, such as public works programmes, to those locked into the informal economy.” While the critics pointed to the conceptual flaw of seeing the formal and informal as structurally disconnected, subsequent policy pronouncements suggested that the informal sector should be eradicated altogether. For example, the next major statement on economic policy imperatives, the Accelerated Shared Growth Initiative of South Africa (or ASGISA), called for the “elimination” of the second economy (RSA, 2006: 11). In 2008, the Presidency initiated the Second Economy Strategy Project which proposed the progressive incorporation of the second into the first economy (Philip, 2009; Philip & Hassen, 2008). The final strategic framework and headline strategies were approved by Cabinet in January 2009 (after Mbeki’s recall as President by the ANC). While the Community Works Programme (CWP) was implemented, the rest of these headline strategies were not translated into the activities at national, provincial and local government levels. AHMR, Vol.3 No2, May-August 2017 – Special Issue 765 Since 2012, increasing attention has been paid to the informal sector at national level – although in somewhat haphazard and uncoordinated fashion. Different initiatives represent simultaneous neglect, support and suppression. The National Development Plan (NDP), for example, assigns a large role to small businesses in its employment scenarios and plans. The NDP’s ideal scenario projects that 11 million jobs will be created by 2030, suggesting that 90% of these new jobs will be created by small and growing enterprises. Depending on the scenario, the plan projects that the informal sector (and domestic work) will create 1.2 million to 2.1 million jobs (NPC, 2012: 121). However, the NDP chapter on the economy says nothing about strategies for the informal sector as such, or how existing operators in the informal sector will be supported, or how barriers to entry will be addressed to help generate new jobs. Meanwhile, also in 2012, the DTI established a new directorate for Informal Business and Chamber Support. Rogerson (2016a: 175) notes that this was an initial recognition by the department of the role of the informal sector in broadening economic participation. By November 2012, the directorate had established a reference group charged with developing a National Informal Business Development Strategic Framework. Under the guidance of the reference group, the DTI staff conducted consultations with stakeholders in the informal sector, formal business and local government officials over a few months, reporting back to the reference group in February 2013. This would lead to the launch of the National Informal Business Upliftment Strategy (NIBUS) in 2014. In the same period, unbeknown to the Reference Group, another section of the DTI was working on new legislation to replace the Businesses Amendment Act of 1993. In March 2013, the DTI released the Draft Business Licensing Bill (DTI, 2013). The draft Bill’s stated aim was “to provide for a simple and enabling framework for procedures for application of business licences by setting norms and standards [providing] a framework for co-operative governance” (DTI, 2013: 5). The DTI Minister, Rob Davies, publicly claimed that the Bill was put in place to deal with illegal traders and semi-illegal practices in South Africa, citing illegal imports, sub-standard goods, counterfeit goods and drug AHMR, Vol.3 No2, May-August 2017 – Special Issue 766 and illegal liquor trading. In fact, such issues are already adequately dealt with through other laws like the Customs and Excise Act of 1964, the Foodstuffs, Cosmetics, and Disinfectants Act of 1972, the Counterfeit Goods Act of 1997, the Drug and Drug Trafficking Act of 1992 and a draft of provincial level legislation aimed at regulating (through the issuing of licences) informal liquor outlets or shebeens. The draft Bill specified that anyone involved in business activities – no matter how small – would need a licence. Foreign migrants would only be licensed if they had business permits (which, according to the 2002 Immigration Act, have to be applied for in the country of origin and are only granted if the applicant can demonstrate having ZAR 2.5 million to invest in South Africa). The Bill suggested wide ranging discretionary powers be given to both the licensing authority and inspectors, far greater than those granted by the 1993 Business Amendment Act. Section 5.1a of that Act mandates an upper limit on fines of ZAR 1,000 or imprisonment not exceeding three months. The draft Bill proposed no upper limits on the fines charged; those found guilty of contravening the Act could be imprisoned for up to ten years. The Draft Bill elicited a flurry of criticism, with many organisations pointing out that the proposals were largely punitive and would result in large-scale criminalising of the informal sector. A submission from StreetNet International, on behalf of street trader organisations from all nine provinces, was particularly vociferous: What the [1991] Businesses Act added to the new South Africa was a developmental approach […] instead of the old abolitionist approach which characterised the Apartheid era. We believe that the repeal of the Businesses Act and replacement with this Bill […] would take us back to the era of forced removals (Horn, 2013: 2). Tensions between foreign and South African informal sector operators have characterised the informal sector since the mid-1990s (Peberdy & Rogerson, 2002). In trying to understand the motivation behind the draft bill, analysts pointed to an upsurge of anti-foreign sentiment not only in the informal sector itself but now from within government. Crush and Ramachandran (2015: 49) cite the following examples: AHMR, Vol.3 No2, May-August 2017 – Special Issue 767 A senior official in the Department of Home Affairs […] is reported to have informed South African MPs that “if you go to Alexandra, you go to Sunnyside, you go everywhere, spaza shops, hair salons, everything has been taken over by foreign nationals […] they displace South Africans by making them not competitive.” At an official meeting, then National Police Commissioner Bheki Cele characterized immigrants and refugees as “people who jump borders,” were flooding into the country and destroying the livelihoods of South African informal traders. He continued: “our people have been economically displaced. All these spaza shops [in the townships] are not run by locals […] One day our people will revolt, and we’ve appealed to the Department of Trade and Industry to do something about it. The onerous conditions imposed on foreign migrants for accessing a licence contained in the Bill would mean that few migrant informal operators would qualify and would therefore be criminalised. Indeed, it has been argued that the Bill was introduced to regulate foreign migrants in the interests of their South African counterparts (Crush et al., 2015: 15-17). According to Rogerson (personal communication), by May 2013, DTI officials conceded the Bill was ‘too blunt’ and in need of re-drafting. At the time of writing (April 2017), no revised draft has been gazetted. In part stemming from the work of the informal-business reference group, the DTI released the National Informal Business Upliftment Strategy (NIBUS) in early 2014 (Rogerson, 2016a). This is the first post-apartheid and nationally- coordinated policy approach to the informal sector. NIBUS has two key delivery arms – the Shared Economic Infrastructure Facility (SEIF) and the Informal Business Upliftment Facility (IBUF) – tackling infrastructure and skills deficits respectively. SEIF provides funding for new infrastructure and upgrading or maintaining existing infrastructure shared by informal businesses. Up to ZAR 2-million financing is available to municipalities on a 50:50 cost-sharing grant basis. IBUF focuses on skills development, promotional material, product improvement, technology support, equipment and help with registration. Part of the IBUF pilot involved training 1,000 informal traders in a partnership with the Wholesale and Retail Sector Education and Training Authority. AHMR, Vol.3 No2, May-August 2017 – Special Issue 768 This was the first time that the DTI had explicitly focused on the informal sector through a strategy to tackle two critical needs, infrastructure and skills development. The policy approach was also characterised by implicit and explicit anti-migrant sentiment. It identifies a supposed “foreign trader challenge” in the informal sector, noting that “there is evidence of violence and unhappiness of local communities with regard to the takeover of local business by foreign nationals” [our emphasis] (DTI, 2014: 10, 22). There are numerous suggestions to strengthen specifically South African informal sector businesses. The proposed solution to reduce the “xenophobia associated with foreign national traders” is to “influence the type of businesses that foreign nationals should run and the demarcated areas where these businesses should be active” (DTI, 2014: 57). NIBUS cites the precedent of the Ghana Investment Promotion Centre Act, which has reserved the sale of any goods in a market, petty trading and hawking, and the operation of metered taxis, car hire services, beauty salons and barber shops to nationals only, as well as India and Malaysia’s restrictions on foreign economic participation (DTI, 2014: 22-3). Much critical reference is made to the policies of the DHA with the DTI noting incorrectly there are “no regulatory restrictions in controlling the influx of foreigners” (DTI, 2014: 22). Rogerson (2016a: 184) concludes that “NIBUS is a pro-development approach for South African informal entrepreneurs which is allied to an anti-developmental agenda towards migrant entrepreneurs.” Anti-foreign sentiment reinforces a generally punitive approach to the informal sector that focuses on regulation and control. The November 2015 report of the Ad Hoc Parliamentary Committee investigating the 2015 xenophobic attacks on migrants and refugees working in the informal sector, recommended the regulation of their township businesses. The report states, for example, that municipal governments should improve systems for providing and monitoring business permits, noting a “tendency of issuing too many licenses” to businesses operating out of residential dwellings, many of which do not comply with municipal by-laws (Parliament of South Africa, 2015: 38-39). There are also reports that disarray around licensing creates a situation where refugee business owners are applying for licences they do not need and paying fines for violating licences they are not required to have. This punitive approach runs counter to good practice in management and support AHMR, Vol.3 No2, May-August 2017 – Special Issue 769 of the informal economy both in South Africa and globally (Chen, 2012). Ironically, increased regulation of township business would be destructive for South African and migrant informal operators alike. The focus on curtailing migrant entrepreneurship diverts attention from what Crush and Ramachandran (2015: 53) identify as “the real, urgent need to support and enhance opportunities for all small entrepreneurs.” Provinces are also mandated to play a role in regulating and supporting the informal sector, but have been slow in addressing the issue. While the 1993 amendment to the Businesses Act empowered provinces to develop dedicated provincial business acts, to date no provinces have done that. For example, in KwaZulu-Natal (KZN) an informal economy policy process was initiated in 2003. This resulted, after an eight-year process – in the KZN Informal Economy Policy of 2011 – but it still has not been developed into a White Paper, the precursor to new legislation. Such foot dragging suggests that it is not a priority. In other provinces, reference is made to the informal sector in local economic development strategies (as in Limpopo) as well as township development strategies. The Western Cape promulgated its first dedicated Informal Sector Framework (Western Cape Province, 2014) and Gauteng recently released the Gauteng Informal Business Development Strategy (Gauteng Province, 2015). Both focus on aligning the relevant departments’ work with NIBUS. All of the documents echo the need for financial and non- financial support to informal businesses (especially through small-business development centres), supporting informal trading in townships, improved access to business-related infrastructure facilities and reviewing regulations and bylaws to support the informal business sector (Gauteng Province, 2015: 47-59). On paper, all of these recommendations seem to be uncontroversial and important steps forward. However, a clear anti-foreign sentiment is also reflected at provincial level. The Gauteng strategy, for example, states: The existing competition for trading permits among local and foreign nationals is evident. Unfortunately, there are no regulatory restrictions in controlling the influx of foreign nationals. The Departments of Trade and Industry and Home Affairs should assist the province in devising strategies AHMR, Vol.3 No2, May-August 2017 – Special Issue 770 and policies to control foreign business activities (Gauteng Province, 2015: 45). Similar sentiments are expressed in the Western Cape document (Western Cape Province, 2014: 46). In KZN, the provincial government’s thrust has been to form and fund the KZN Provincial Association of Traders, and fund traders’ training academies in various districts. The purpose of this initiative, as outlined at its launch, is to “bring back our general dealer stores that used to be seen in our townships and villages” adding that these “stores […] have been sold to foreign nationals” (News 24, 2015). Statements like this suggest that the provincial government aims to ‘level the playing field,’ favouring South Africans over migrant and refugee operators. The provincial government in Limpopo Province has perhaps the most overtly hostile approach to migrants and refugees working in the informal sector. In 2012, it launched Operation Hardstick, an aggressive military-style campaign that targeted small informal businesses run by migrants and refugees. The Somali Association of South Africa supported by Lawyers for Human Rights contested the action in the Courts. Court documents show that despite being labelled a crime-fighting initiative, Operation Hardstick was selectively enforced, affecting only migrant entrepreneurs and not South African businesses in the same locations. Police shuttered over 600 businesses, detained owners, confiscated stock, imposed fines for trading without permits and verbally abused the owners. Affected business owners were informed that “foreigners” were not allowed to operate in South Africa, that their asylum- seeker and refugee permits did not entitle them to run a business and that they should leave the area. Thirty displaced migrants from Ethiopia were forced to flee when the house they had taken refuge in was fire-bombed (Supreme Court, 2014). The Court noted that police actions “tell a story of the most naked form of xenophobic discrimination and of the utter desperation experienced by the victims of that discrimination” (Supreme Court, 2014:6-7). The judgment observed that “one is left with the uneasy feeling that the stance adopted by the authorities in relation to the licensing of spaza shops and tuck-shops was in order to induce foreign nationals who were destitute to leave our shores” AHMR, Vol.3 No2, May-August 2017 – Special Issue 771 (Supreme Court, 2014:25). The Court ruled in favour the Somali Association and effectively established the right to self-employment for all asylum-seekers and refugees. Opposing the appeal were all three tiers of government – national, provincial and municipal – including the Limpopo Member of the Executive Council (MEC) for Safety, Security and Liaison; the Provincial Commissioner of Police; the National Police Commissioner; the Standing Committee on Refugee Affairs; the Ministers of Police, Labour and Home Affairs; and two municipalities. At local government level, there is a preoccupation with the most visible element of the informal sector – street vendors – who operate in public spaces over which there are often competing interests. However, a scan of policy statements on street trading or the informal sector shows that, on paper, the positive contribution of the informal sector is often recognised. For example, in its street trading policy, the City of Johannesburg (2009: 3) states that “informal trading is a positive development in the micro business sector as it contributes to the creation of jobs and alleviation of poverty and has the potential to expand further the City’s economic base.” The City of Cape Town’s (2013: 8) policy advocates a “thriving informal trading sector that is valued and integrated into the economic life, urban landscape and social activities within the City of Cape Town.” The eThekwini Informal Economy Policy (2001:1) states that the “informal economy makes an important contribution to the economic and social life of Durban.” Despite the positive rhetoric, city-level actions reveal an ambivalent, if not actively hostile, approach to street traders. Wafer (2011) provides details of the aggressive approach to street trading in Johannesburg over the post- apartheid period. In late 2013, this culminated in a draconian action when the Johannesburg City Council violently removed and confiscated the inventory of about 6,000 inner-city street traders, many of them migrants (Zack, 2015; Rogerson, 2016a; Rogerson 2016b). A group of traders took the City to court and the Constitutional Court ruled in their favour. Acting Chief Justice Moseneke stated that Operation Clean Sweep was an act of “humiliation and degradation” and that the City’s attitude “may well border on the cynical” (Constitutional Court, 2014). Street traders have returned to the streets but in AHMR, Vol.3 No2, May-August 2017 – Special Issue 772 more limited numbers with the city declaring large inner- city areas to be restricted and prohibited trade zones. An analysis of Cape Town’s approach to street traders indicates systematic exclusion in the inner-city, exemplified by the allocation of only 410 street- trading bays in the whole inner city (Bukasa, 2014). Township trading is characterised by long-term neglect. Zulu (2016) shows that in Khayelitsha, the city council had invested very little in infrastructure for street traders and had devolved the management of street trading to a small group of traders, with negative consequences for many. Crush et al. (2015: 15) argue that, although the policy environment in Cape Town varies across parts of the city and between segments of the informal economy, “the modernist vision of a ‘world- class city’ with its associated antipathy to informality dominates, and informal space and activity is pathologized.” Foreign migrants face an additional set of challenges. In an interview for this study, the Cape Town Department of Economic Development claimed that “the City, in terms of its policy around trading, doesn’t differentiate and we don’t discriminate. There’s set criteria in terms of who qualifies (for a trading bay and permit) and how that person qualifies. We don’t look at what nationality the person is.” But, as the interviewed official admitted, the City is forced to discriminate in practice because refugees have to produce documentation that South Africans do not. In particular, the renewal of asylum and refugee permits is extremely unpredictable (in terms of wait times and length granted) and became much more difficult once the DHA declared that renewal had to be effected at the office of application, which could be thousands of kilometres away in Johannesburg or even Musina. One common way around this challenge is for South Africans to obtain the permit and then rent the space to refugees at a profit. Gastrow and Amit (2015) detail the regulation of Somali-owned spaza shops in Cape Town townships and show how mediation efforts led by the police and non-governmental organisations have culminated in agreements prohibiting the opening of new Somali shops in certain areas. They outline various formal regulatory attempts to control and curtail the operations of Somali businesses including fines, drafting new by-laws, issuing policy statements about foreign AHMR, Vol.3 No2, May-August 2017 – Special Issue 773 shops and proposing laws tightening the regulation of the spaza market (Gastrow & Amit 2015). They also note that both formal and informal measures skirt the law, are applied in a discriminatory manner and stifle free competition. The City thus appears to be protecting the interests of South African spaza shop owners over their foreign counterparts. With respect to the informal sector more generally, the modernist vision of the ‘world-class city’ with its associated antipathy towards informality and the pathologising of informal space and activity, predominates. Durban was once hailed for its relatively liberal stance on the informal sector (Lund & Skinner, 2004; Dobson & Skinner, 2009). A progressive informal sector policy was unanimously accepted by the City Council in 2001 and remains official policy today. The Council’s actions reflect a more ambivalent approach, however. For example, a Council-approved shopping mall development at the inner-city Warwick Junction transport node threatened 6,000 traders operating there, and was only halted by a legal challenge (Skinner, 2010). In 2013, traders in both the inner city and outlying areas identified harassment by the police as their key business challenge (Dube et al., 2013). In 2015, traders won a legal case challenging the constitutionality of confiscating their goods, forcing the city to redraft the street trader by-laws. Again, court action proved to be the only way to secure relief. Conclusion Distinct from many other refugee receiving countries, South Africa’s rights- based refugee legislation has historically allowed for refugees and asylum- seekers to access a broad array of rights from health services to education and employment. South Africa has never hosted a dedicated refugee camp or detention centre. In this environment, refugees and asylum-seekers have independently found their way into South Africa’s social and economic fabric, sending their children to South African schools, finding employment in South African businesses and households and establishing their own formal and informal businesses. The 2017 Amendment Act and White Paper are clear evidence of a new and less generous policy direction which is intended to shrink asylum space and further constrain the rights and protections afforded to refugees and asylum-seekers. AHMR, Vol.3 No2, May-August 2017 – Special Issue 774 Cumulatively, the changes documented in this paper illustrate a significant shift in South Africa’s long held policy for the local integration of refugee populations. By removing the right to work and confining asylum-seekers to detention centres, it is assumed that the practice of asylum-seeking by economic migrants will cease. This, of course, completely ignores evidence of the positive economic contribution of refugees and asylum-seekers who, under existing law, are permitted to pursue economic livelihoods. The ongoing move away from a rights-based approach to refugee protection is seen most clearly in the 2017 Refugees Amendment Act and the proposals in the 2017 White Paper on International Migration. International trends which increasingly stress the positive development impacts of refugee populations are being completed ignored (UNHCR, 2014; World Bank, 2016). Rather, the emphasis is on the ‘exceptionalism’ of forced migrants and the need to craft a coercive, non-developmental approach to dealing with refugees. This represents a profound shift in the country’s approach to refugee rights, protections and associated international obligations, moving away from an integration approach towards a containment approach. While the new approach may appear to be a local response to intemperate local demands, it is part of a more global trend which seeks to inhibit access to the physical territory and refugee protection systems of those countries by erecting physical, economic and social barriers (Mountz, 2013; Mountz et al., 2013). If the proposed South African refugee policy is ever operationalised in the face of the inevitable court and constitutional challenges, it may deter non-refugee migrants from further afield, but it is unlikely to act as a deterrent to economic migrants from neighbouring countries. They will simply find other ways to come, live and work in South Africa, a fact which seems to be acknowledged in the White Paper’s proposals to make SADC-specific work, traders’ and micro-enterprise permits available to migrants. While there may be a belief that detention centres will reduce the flow of genuine asylum-seekers to South Africa, there remains a whole set of unanswered questions about whether there will be new policies that directly affect those who have refugee status. Here the White Paper is almost silent, AHMR, Vol.3 No2, May-August 2017 – Special Issue 775 although it does assert that under the existing system refugees are allowed to apply for permanent residence “even though their status is inherently temporary.” The Amendment Act extends the period of qualification to 10 years while the White Paper recommends abolishing this right altogether. There is no indication of whether those with refugee status will be denied the right to work, to self-employment, to freedom of movement and to access health and educational services. On the other hand, they will not be given any additional resources and will be expected to pursue their own livelihoods, as at present. To understand the challenges and obstacles that refugees face in securing these livelihoods, it is important to examine the policy and regulatory environment within which those in the informal sector try to survive. Refugees and asylum-seekers confront a formidable set of challenges in operating their informal enterprises in South African cities (Crush et al., 2015). At best tolerated, and at worst hounded out of communities by xenophobic mobs and violent entrepreneurship, South Africa is certainly not a safe haven for those fleeing violence and persecution at home. However, as this paper argues, xenophobic violence and police inaction are certainly not the only difficulties they face. South African city managers oscillate between benign neglect and active destruction of the vibrant and economically-productive informal sector. Migrants and refugees who operate informal enterprises have been the main targets in a series of national, provincial and local-level “operations” designed to inhibit or eradicate their businesses from urban space. Thus, there is a fundamental contradiction between a refugee protection policy that demands self-reliance from refugees and informal sector policies that undermine self- reliance at every turn. A comparison of the 2015 Draft Refugees Amendment Bill and the final 2016 Act suggests that the petitions and representations of human rights groups and refugee and migrant associations had little or no impact in softening the legislation, so litigation in the courts is the most likely way to roll back these draconian provisions. The courts have clearly played an increasingly important role in securing the livelihoods of informal sector operators in general, and migrant entrepreneurs in particular, in post-apartheid South AHMR, Vol.3 No2, May-August 2017 – Special Issue 776 Africa. Litigation has been one of the key sources of support to migrant entrepreneurs, highlighting the core contradiction between the rights enshrined in the South African Constitution and South Africa’s relatively progressive refugee protection regime, on the one hand, and the policies and actions of key government departments and officials, on the other. Protecting unalienable rights relies on a cohort of legal human rights non-governmental organisations prioritising migrant livelihood rights and being willing and able to pursue time-consuming and costly litigation on their behalf. Acknowledgements We wish to thank the UNHCR Geneva for funding the Refugee Economic Impacts SAMP research project. Our thanks also go to Vanya Gastrow and Sujata Ramachandran for their assistance. References Amit, R. 2011. No refuge: Flawed status determination and the failures of South Africa’s refugee system to provide protection. International Journal of Refugee Law, 23: 458-488. Amit. R. 2012. No Way In: Barriers to Access, Service and Administrative Justice at South Africa’s Refugee Reception Offices. ACMS Report, Johannesburg: ACMS. Amit, R. 2015. Queue Here for Corruption: Measuring Irregularities in South Africa’s Asylum System. Report by Lawyers for Human Rights and African Centre for Migration and Society. Johannesburg: LHR and ACMS. Budlender, D., Skinner C. and Valodia, I. 2004. Budgets and the Informal Economy: An Analysis of the Impact of the Budget on Informal Workers in South Africa. School of Development Studies Research Report, Durban: University of KwaZulu-Natal. Bukasa, A. 2014. Securing Sustainable Livelihoods: A Critical Assessment of the City of Cape Town’s Approach to Inner City Street Trading. MA Thesis, University of Cape Town, Cape Town. Carciotto, S. 2016. Angolan refugees in South Africa: Alternatives to permanent repatriation? African Human Mobility Review, 2: 362-382. AHMR, Vol.3 No2, May-August 2017 – Special Issue 777 Chandra V & Rajaratnam B. 2001. Constraints to Growth and Employment in the Informal Sector: Evidence from the 1999 Informal Survey Firm. Washington: World Bank. City of Cape Town. 2013. Informal Trading Policy 2013 (Policy Number 12664). Cape Town. City of Johannesburg. 2009. Informal Trading Policy 2009. Johannesburg. Chen, M. 2012. The Informal Economy: Definitions, Theories and Policies. WIEGO Working Paper No. 1, Cambridge, MA: WIEGO. Constitutional Court. 2014. South African Informal Traders Forum and Others v City of Johannesburg and Others; South African National Traders Retail Association v City of Johannesburg and Others (CCT 173/13; CCT 174/14) [2014] ZACC 8; 2014 (6) BCLR 726 (CC); 2014 (4) SA 371 (CC) (4 April). Crush, J. and Ramachandran S. 2015. Doing business with xenophobia. In Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa: IDRC and Cape Town: SAMP, pp. 25-59. Crush, J. and Tevera, D. (Eds.). 2010. Zimbabwe’s Exodus: Crisis, Migration, Survival. Ottawa: IDRC and Cape Town: SAMP. Crush, J., Chikanda, A. and Skinner, C. (Eds.). 2015. Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa: IDRC and Cape Town: SAMP. Devey, R., Skinner, C. and Valodia I. 2003. Human resource development in the informal economy. In Kraak, A. and Press, K. (Eds.). Human Resource Development Biennial Directory. Cape Town: Human Sciences Research Council. Devey, R., Skinner, C. and Valodia, I. 2006. The state of the informal economy. In Buhlungu, S., Daniel, J., Southall, R. and Lutchman, J. (Eds.). The State of the Nation, 2005-2006. Cape Town: HSRC Press. AHMR, Vol.3 No2, May-August 2017 – Special Issue 778 Devey, R. Skinner, C. and Valodia I. 2008. The informal economy. In Kraak, A. and Press, K. (Eds.). Human Resource Development Biennial Directory. Cape Town: HSRC. DHA. 2016. Green Paper on International Migration in South Africa. Pretoria: Department of Home Affairs. Dobson, R. and Skinner, C. 2009. Working in Warwick: Integrating Street Traders into Urban Plans. Durban: University of KwaZulu-Natal. DTI (Department of Trade and Industry). 2013. The Draft Business Licensing Bill. Pretoria: Department of Trade and Industry. DTI. 2014. The National Informal Business Upliftment Strategy (NIBUS). Pretoria: Department of Trade and Industry. Du Toit, A. and Neves, D. 2007. In search of South Africa’s ‘second economy’. Africanus, 37: 145-174. Dube, G., Mkhize, S. and Skinner, C. 2013. Informal Economy Monitoring Study: Street Vendors in Durban, South Africa. Inclusive Cities Research Report, Manchester: WIEGO. Fiddian-Qasmiyeh, E., Loescher, G., Long, K. and Sigona, N. (Eds.). 2014. The Oxford Handbook of Refugee and Forced Migration Studies. Oxford: OUP. Gastrow, V. and Amit, R., 2015. The role of migrant traders in local economies: A case study of Somali spaza shops in Cape Town. Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa: IDRC, pp.162-177. Gauteng Province (Economic Development Department). 2015. Gauteng Informal Business Upliftment Strategy. Johannesburg: Policy Memo. Gordon, S. 2016. Welcoming refugees in the Rainbow Nation: Contemporary attitudes towards refugees in South Africa. African Geographical Review, 35: 1- 17. Handmaker, J. 2001. No easy walk: Advancing refugee protection in South Africa. Africa Today, 48: 91-113. AHMR, Vol.3 No2, May-August 2017 – Special Issue 779 Handmaker, J., de la Hunt, L. and Klaaren, J. (Eds.). 2011. Advancing Refugee Protection in South Africa. 2nd Edition. Oxford: Berghahn Books. Hassim, S., Kupe, T. and Worby, E. (Eds.). 2008. Go Home or Die Here: Violence, Xenophobia and the Reinvention of Difference in South Africa. Johannesburg: Wits University Press. Horn, P. 2013. Submission on behalf of StreetNet International on the Licensing of Businesses Bill 2013 – Notice 231 of 2013. From Igglesden, V. and Schreier, T. 2011. Expanding the Protection Space for Refugees in South Africa. Refugee Rights Unit Working Paper 2, University of Cape Town. Jacobsen, K. 2005. The Economic Lives of Refugees. Bloomfield, CT: Kumarian Press. Johnson, C. 2015. Failed asylum seekers in South Africa: Policy and Practice. African Human Mobility Review, 1: 201-228. Landau, L. 2006. Protection and dignity in Johannesburg: Shortcomings of South Africa’s urban refugee policy. Journal of Refugee Studies, 19: 308-327. Landau, L. and Duponchel, M. 2011. Laws, policies, or social position? Capabilities and the determinants of effective protection in four African cities. Journal of Refugee Studies, 24: 1-22. Lund, F. 1998. Women Street Traders in Urban South Africa: A Synthesis of Selected Research Findings. CSDS Research Report No. 15, University of Natal: Durban. Lund, F. and Skinner, C. 2004. Integrating the informal economy in urban planning and governance: A case study of the process of policy development in Durban, South Africa. International Development Planning Review, 26: 431- 456. Mabuza, E. 2016. Economic refugees abuse South Africa's liberal asylum regime: Gigaba. Times Live, 20 June 2016. AHMR, Vol.3 No2, May-August 2017 – Special Issue 780 Mah, K. and Rivers, P. 2016. Refugee housing without exception. Space and Culture, 19: 390-405. Manning, C. and Mashingo, A. 1994. Manufacturing and South African microenterprises. IDS Bulletin, 25(1): 30-36. Mbeki, T. 2003. Address to the National Council of Provinces, 11 November. From Mountz, A. 2013. Shrinking spaces of asylum: Vanishing points where geography is used to inhibit access to asylum. Journal of Human Rights, 19: 29- 50 Mountz, A., Coddington, K., Catania, R. and Loyd, J. 2013. Conceptualizing detention: Mobility containment bordering and exclusion. Progress in Human Geography, 37: 522-541 NPC (National Planning Commission). 2012. National Development Plan: Vision for 2030, Our Future – Make it Work. Pretoria: NPC. News24. 2015. MEC to uplift smaller businesses. News24, 14 May 2015. Parliament of South Africa. 2015. Report of The Ad Hoc Joint Committee On Probing Violence Against Foreign Nationals, Dated 19 November 2015. Peberdy, S. and Rogerson, C. 2002. Transnationalism and non-South African entrepreneurs in South Africa’s small, medium and micro-enterprise (SMME) economy. In Crush, J. and McDonald, D. (Eds.). Transnationalism and New African Immigration to South Africa. Toronto: CAAS, pp. 20-40. Philip, K. 2009. Second Economy Strategy: Addressing Inequality and Economic Marginalisation. A Strategic Framework. Second Economy Strategy Project Report. Johannesburg: TIPS. Philip, K. and Hassen, E. 2008. The Review of Second Economy Programmes, An Overview for the Presidency’s Fifteen-Year Review. From Polzer Ngwato, T. 2013. Policy Shifts in the South African Asylum System. Pretoria and Johannesburg: LHR and ACMS. AHMR, Vol.3 No2, May-August 2017 – Special Issue 781 RSA (Republic of South Africa). 1991. Businesses Act 71 of 1991. Government Gazette, 311(13266). RSA. 1993. Businesses Amendment Act No. 186 of 1993. Government Gazette, 342(15374). RSA. 1995. White Paper on National Strategy for the Development and Promotion of Small Business in South Africa. Cape Town: South African Parliament. RSA. 2006. Accelerated Shared Growth Initiative of South Africa. From Rogerson, C. 2004. The impact of the South African Government’s SMME programmes: A ten year review. Development Southern Africa, 21: 765-784. Rogerson, C. 2016a. South Africa’s informal economy: Reframing debates in national policy. Local Economy, 31(1-2): 172-186. Rogerson, C. 2016b. Progressive rhetoric, ambiguous policy pathways: Street trading in inner-city Johannesburg, South Africa. Local Economy, 31: 204-218. Skinner, C. 2005. Constraints to Growth and Employment: Evidence from the Informal Economy. School of Development Studies Research Report No. 65, Durban: University of KwaZulu-Natal. Skinner, C. 2010. Challenging city imaginaries: Street traders’ struggles in Warwick Junction. Agenda, 23: 101-109. Stupart, R. 2016. Is South Africa home to more than a million asylum seekers? The numbers don’t add up. Africa Check, 15 August 2016. Supreme Court of Appeal. 2014. Somali Association of South Africa and Others v Limpopo Department of Economic Development Environment and Tourism and Others (48/2014) [2014] ZASCA 143; 2015 (1) SA 151 (SCA); [2014] 4 All SA 600 (SCA) (26 September). UNHCR. 2014. Global Strategy for Livelihoods. Geneva: UNHCR. Unicity, E., 2001. Durban's Informal Economy Policy. AHMR, Vol.3 No2, May-August 2017 – Special Issue 782 Wafer, A. 2011. Informality, infrastructure and the state in post-apartheid Johannesburg. PhD Thesis, Open University. Washinyira, T. 2012. Asylum seekers struggle as FNB freezes their accounts. GroundUp, 5 December 2012. World Bank. 2016. Forcibly displaced: Toward a development approach supporting refugees, the internally displaced, and their hosts. Washington DC: World Bank. Western Cape Province (Department of Economic Development and Tourism). 2014. Draft Informal Sector Framework, Cape Town, Policy Memo. Zack, T. 2015. Making an area hot: Interrupting trade in an ethnic enclave in Johannesburg’s inner city. In Crush, J., Chikanda, A., Skinner, C. (Eds.). Mean streets: Migration, xenophobia and informality in South Africa. Ottawa: IDRC. Zulu, T. 2015. Interchange for the Informal Market? An Investigation of the Nature of the Nolungile Station Upgrades in Khayelitsha, Cape Town and their Implications for Informal Traders. MA Thesis, University of Cape Town. AHMR, Vol.3 No2, May-August 2017 – Special Issue 783 Refugee Entrepreneurial Economies in Urban South Africa Jonathan Crush*, Godfrey Tawodzera**, Cameron McCordic***, Sujata Ramachandran**** and Robertson Tengeh***** Abstract The case-study literature on refugees and asylum-seekers in South Africa is dominated by an overwhelming focus on the problems they face and their marginalised existence, reinforcing an image of victimhood, exploitation and vulnerability. In this paper, we seek to broaden the image of refugees and their economic impacts beyond a narrow focus on their marginal status and vulnerable position. They are viewed here as dynamic agents with skills and capabilities who can play an integral role in transforming local settings and contributing to economic development. This paper presents and discusses the results of a survey of over 1,000 refugee informal business-owners in Cape Town and small-town Limpopo. As well as providing new insights into the nature, achievements and challenges of refugee entrepreneurship, the paper addresses the question of whether geographical location makes a difference to the nature of refugee entrepreneurial economies by contrasting the two groups of entrepreneurs. Refugee entrepreneurs are more vulnerable to xenophobic violence in Cape Town and to police misconduct in Limpopo. Otherwise, there are remarkable similarities in the manner in which refugee entrepreneurs establish and grow their businesses in large cities and small provincial towns. Keywords Refugees, informal sector, entrepreneurial economies, xenophobia, Cape Town, Limpopo. * International Migration Research Centre, Balsillie School of International Affairs, 67 Erb Street W, Waterloo, ON, N2L 6C2. Email: jcrush@balsillieschool.ca ** Department of Geography and Environmental Studies, University of Limpopo, P. Bag X1106, Sovenga. *** Balsillie School of International Affairs, 67 Erb Street W, Waterloo, ON, N2L 6C2. **** Southern African Migration Programme, International Migration Research Centre, Balsillie School of International Affairs, 67 Erb St West, Waterloo, Ontario, Canada N2L6C2. ***** Faculty of Informatics and Design, Cape Peninsula University of Technology, Cape Town, South Africa. AHMR, Vol.3 No2, May-August 2017 – Special Issue 784 Introduction More than 60% of the world’s refugees now live in urban areas, according to the UNHCR. Therefore, the social and economic impacts of refugee movements are increasingly being felt in the towns and cities of host nations. In general, the presence of refugees among urban populations is more likely to be assessed in terms of perceived burdens rather than benefits (Omata & Weaver, 2015). Only recently has interest emerged in what has been characterised by Jacobsen (2005) as the positive “economic lives of refugees.” Some researchers have adopted the notion of “refugee economies” to highlight the strong involvement of refugees in the many overlapping processes of production, consumption, exchange, entrepreneurship and development of financial and capital markets in host countries and beyond (Betts et al., 2014). However, it would be a mistake to imagine that there are discrete spatially and economically isolated “refugee economies” in the urban environment. Refugees may dominate particular economic and geographical niches in the urban economy but they continuously interact with and contribute to that economy in ways that are poorly understood. As refugees increasingly become the norm in the urbanising Global South, more research on the specifically urban economic impacts of protracted refugee situations is therefore urgently needed (Koizumi & Hoffstaedter, 2015). This paper aims to contribute to recent literature that takes “refugee livelihoods” as its point of departure (Buscher, 2011; Jacobsen, 2006; Omata & Kaplan, 2013; de Vriese, 2006). A livelihoods perspective underscores the need to adopt a rights-based approach to the economic activities of forced migrants; to identify their resources, assets and skills; and to understand how they respond to their situations and secure their livelihood needs. This perspective represents a move away from welfare-centred models of engagement focused on meeting basic needs, towards a capability approach which emphasises refugees’ agency, right to choose and ability to control their own environments (Landau, 2008). Identifying the economic advantages and benefits associated with the presence of refugees is a key precondition for making governments and local elites more responsive to refugee needs and removing the barriers they face (Lyytinen & Kullenberg, 2013). This objective AHMR, Vol.3 No2, May-August 2017 – Special Issue 785 coheres with the UNHCR’s (2014) Global Strategy for Livelihoods, which underscores the need to enhance the economic capabilities of refugee populations and build on their own resilience. The sizable case-study literature on refugees and asylum-seekers in South Africa is dominated by an overwhelming focus on the problems they face and their marginalised existence, reinforcing an image of victimhood, exploitation and vulnerability (Crush & Chikanda, 2014). In recent years, a more troubling image has emerged of refugees as unfair and underhand competitors who disadvantage poor South Africans (Desai, 2008; Gordon, 2016). Both characterisations suffer from obvious problems. The former treats them in a restricted manner as undifferentiated, homogeneous populations defined by their marginalised status as recipients of protection in the host country. The latter views them in a prejudicial manner through a warped understanding of the nature of economic competition and refugee rights. Undue emphasis on their legal standing as ‘refugees’ tends to minimise their education, credentials and work experience in various fields. Furthermore, it underestimates their creative energies, determination and ability to overcome some of these challenges and re-build their lives in South Africa. Post-apartheid South Africa has relied on a “self-settlement” approach to refugees (Handmaker et al., 2011). The country imposes no restrictions on freedom of movement and the geographical locations where refugees can reside. They are not confined to refugee camps or physically separated from citizens, which means they have direct interaction with South Africans with the potential for both conflict and integration into local communities. However, the South African self-settlement model does emphasise self- sufficiency on the part of refugees in the process of resettlement, with little assistance from state authorities (Polzer Ngwato, 2013). Existing studies suggest that many urban refugees are unable, for various reasons, to access formal employment in the cities and turn to the informal economy for their livelihoods (Achiume, 2014; Jacobsen, 2006; Kavuro, 2015). Employment or self-employment in the informal economy becomes a desirable option, if not a necessity. Because urban informality is generally invisible to policy-makers, scorned by politicians and seen as a site of desperation rather than AHMR, Vol.3 No2, May-August 2017 – Special Issue 786 opportunity, its economic significance is often overlooked (Pavanello et al., 2015; Williams, 2013). The primary aim of this paper is to examine what we call “refugee entrepreneurial economies” in urban South Africa. The paper builds on case- study evidence on the activities of refugees in the South African informal economy (Greyling, 2015; Jinnah, 2010; Owen, 2015; Smit & Rugunanan, 2014; Thompson, 2016; Zack, 2015). Within the informal sector, refugee economies have often been viewed as stagnant pools of desperation, providing narrow opportunities and limited scope for advancement. But this is a misleading characterisation. It is important to recognise the dynamism and growing complexity of South African refugee economies and to re-shape our ideas about the economic impacts of the informal sector (Roever & Skinner, 2016). The important question is whether all forms of refugee activity in the informal sector are associated with economic precariousness and social marginality, or whether possibilities for real economic advancement exist (Crush et al., 2015a). In this paper, we seek to broaden the image of refugees and their economic impacts beyond a narrow focus on their marginal status and vulnerability. We view them as dynamic agents with skills and capabilities who can play an integral role in transforming local settings and contributing to economic development. This is not to undervalue the divergent ways through which refugees earn a living in South Africa, nor to gloss over their struggles and the pervasive discrimination, hostility and xenophobic violence they regularly encounter, nor to exaggerate the significance of refugee entrepreneurship. However, there is a need to re-calibrate the narrow, partial and negative lens through which their economic potential is evaluated by politicians and policy- makers. Here, we also address the question of whether geographical location makes a difference to the nature of refugee entrepreneurial economies by contrasting refugee enterprise in a major South African city (Cape Town) with that in several smaller towns in a different part of the country (Limpopo Province). Our aim is to ascertain whether geography matters in framing the activities, challenges and impacts of refugee entrepreneurial economies. AHMR, Vol.3 No2, May-August 2017 – Special Issue 787 Methodology Most refugees are initially drawn to the major South African cities of Johannesburg, Durban, Port Elizabeth and Cape Town. In this study of refugee entrepreneurship, the city of Cape Town was chosen as one of two study sites because there is a significant concentration of refugees and a contextual literature on refugee entrepreneurship in the city (Basardien et al., 2014; Gastrow & Amit, 2013, 2015; Owen, 2015; Tawodzera et al., 2015). In recent years, refugee entrepreneurs have also been establishing businesses in smaller urban areas around the country. In 2012, a police campaign called Operation Hardstick in Limpopo Province forcibly closed as many as 600 small businesses run by migrants and refugees, suggesting that this province had become a significant operating location for small-town entrepreneurs. In contrast to Cape Town, however, very little research exists on refugee livelihoods and entrepreneurship in Limpopo, which was chosen as the second site for this study (Idemudia et al., 2013; Mothibi et al., 2015; Ramathetje & Mtapuri, 2014). The study focused only on informal sector business owners who hold refugee (Section 24) permits under the Refugees Act. Holders of asylum-seeker (Section 22) permits were not included as many of these migrants are unlikely to be refugees, as conventionally defined. The South African Government claims that 90% of asylum-seeker permit holders are “economic migrants” and not genuine refugees as defined by the Refugee Conventions. While this figure has been disputed by researchers, it is not possible to predict whether any particular asylum-seeker permit holder will make the transition to full refugee status. Indeed, there is every likelihood that the majority will not (Amit, 2011; Amit & Kriger, 2014). The number and geographical location of refugee enterprises in South Africa is unknown, which means that there is no comprehensive database from which to draw a sample. Therefore, this paper relied on “maximum variation sampling” (MVS) using the methodology suggested by Williams and Gurtoo (2010) in their study of the informal economy in Bangalore, India. In Cape Town, the areas of the city in which the informal economy is known to operate were first identified from the existing literature, personal knowledge and a AHMR, Vol.3 No2, May-August 2017 – Special Issue 788 scoping of the city. This allowed for the exclusion of significant areas of the city from the study, particularly high-income residential areas. Second, five different types of area were identified – commercial, formal residential, informal residential, mixed formal and informal residential and industrial – and within each of these types contrasting and geographically separated areas were chosen. There were two commercial areas (the CBD and Bellville), two industrial areas (Maitland and Parow), three formal residential areas (Observatory, Delft and Nyanga), three informal settlements (Khayelitsha, Imizamo Yethu and Masiphumelele) and two mixed formal and informal residential areas (Philippi and Dunoon). In Limpopo, the majority of migrant entrepreneurs live in urban centres. Since many of these centres are relatively small, the primary criterion for the application of MVS in Limpopo was variable urban size. The selected towns are scattered around the province and include Polokwane (population of 130,000), Thohoyandou (70,000), Musina (43,000), Louis Trichardt (25,000), Tzaneen (14,500) and Burgersfort (6,000). Sampling was conducted in each site using the mapped grid-pattern exhibited by streets. Streets were sampled one after another in successive fashion moving from west to east. After identifying the first five enterprises on a street, and randomly selecting the first of the five for the sample, every third refugee- owned enterprise was selected for interview. Where business owners were not available, field workers made three call backs to the enterprise, after which a substitution was made. In each of the two study areas, a total of 504 refugee entrepreneurs were interviewed. The survey instrument was administered using tablet technology and collected information on a wide variety of issues pertaining to business set-up, activities, operation, profitability, challenges and economic impacts. The enterprise survey was complemented with fifty in- depth interviews with business-owners in each research location and three focus groups with refugees from the same country in each location. This paper presents some of the results of the survey supplemented with a selection of observations from the entrepreneurs about their experiences running businesses in South Africa. AHMR, Vol.3 No2, May-August 2017 – Special Issue 789 Refugee Profile This section of the paper compares refugee business owners in Cape Town with those in small town Limpopo to ascertain if there are any significant differences in their respective socio-demographic and migration profiles. First, male refugee entrepreneurs clearly dominate in both areas, with only 20-25% of business owners being women (Table 1). This is generally consistent with the predominantly male character of refugee migration to South Africa. The entrepreneurs in both locations tend to be relatively young, with 80% in Cape Town and 77% in Limpopo being under the age of 40. Very few entrepreneurs in either location were over the age of 50. Accepting the International Labour Organisation definition of “young people” in Africa as anyone under the age of 35, this would mean that 54% of those in Cape Town and 57% of those in Limpopo are young entrepreneurs (Kew, 2015). There are no significant differences between the two groups of refugee entrepreneurs in terms of age profile. However, the difficulties and challenges facing youth entrepreneurs in South Africa, and how these are addressed by refugee youth, warrant further investigation (Fatoki & Chindoga, 2011; Gwija et al., 2014). Despite the similarities in sex and age breakdown between the two samples, there are marked differences in the countries of origin of refugee entrepreneurs. In the Cape Town sample, the most numerous group was from Somalia, which is consistent with other studies showing a significant Somali presence in the informal economy of the city (Basardien et al., 2014; Gastrow & Amit, 2013, 2015; Tawodzera et al., 2015). While there are some Somali- owned businesses in Limpopo and Ethiopian-owned businesses in Cape Town, the largest group in this sample of refugee entrepreneurs in Limpopo towns was from Ethiopia (28%). Some national groups are well-represented in both places, including refugees from the Democratic Republic of the Congo (DRC) (17% in Cape Town and 11% in Limpopo) and Zimbabwe (7% in Cape Town and 11% in Limpopo). AHMR, Vol.3 No2, May-August 2017 – Special Issue 790 Table 1: Demographic Profile of Refugee Entrepreneurs Cape Town % Limpopo % Sex Male 75.2 80.0 Female 24.8 20.0 Age <21 0.0 0.2 21-25 8.5 10.8 26-30 20.1 23.4 31-35 25.2 22.2 36-40 26.6 21.0 41-45 12.2 11.4 46-50 4.3 6.8 >50 3.0 4.4 Country of origin Somalia 26.9 7.4 DRC 16.9 11.0 Cameroon 11.6 0.8 Ethiopia 9.2 28.1 Zimbabwe 7.2 11.1 Congo Brazzaville 3.0 0.6 Burundi 1.4 0.6 Angola 1.0 0.6 Rwanda 0.8 0.2 AHMR, Vol.3 No2, May-August 2017 – Special Issue 791 In both the Cape Town and Limpopo samples, there are small numbers of refugees from the same countries, notably Burundi, Cameroon, Congo Brazzaville, Eritrea, Rwanda and Sudan. In sum, both areas are dominated by refugees from one or two (albeit different) countries, but beyond that there is considerable diversity in the makeup of the refugee entrepreneurial population. The vast majority of sampled refugee business owners (over 90% in both research locations) arrived in South Africa after 2000 (Figure 1). Limpopo has a slightly greater proportion of recent arrivals than Cape Town. For example, 71% of the refugees in Limpopo migrated to South Africa after 2005, compared to 61% of those in Cape Town. Or again, 26% of those in Limpopo arrived in South Africa after 2010, compared to only 16% in Cape Town. This raises the possibility that more recent forced migrants may be avoiding the large cities and going straight to smaller urban centres. However, in Limpopo there is a clear lag between year of migration to South Africa and year of arrival in the province (Figure 2). As many as 64% of the sampled refugee entrepreneurs in Limpopo had first lived in another South African town or city (compared to only 27% in Cape Town) (Table 2). In other words, Cape Town was the first destination for nearly three-quarters of the refugee entrepreneurs in that sample, while Limpopo was the first destination for only 13% of the entrepreneurs sampled there. Sudan 0.2 1.0 Eritrea 0.2 4.6 Other 21.6 34.4 AHMR, Vol.3 No2, May-August 2017 – Special Issue 792 Figure 1: Year of Arrival in South Africa Figure 2: Migration of Limpopo Entrepreneurs 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 1991 - 1995 1996 - 2000 2001 - 2005 2006 - 2010 2011 - 2016 % Year Cape Town Refugees (n=501) Limpopo Refugees (n=502) 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 1991 - 1995 1996 - 2000 2001 - 2005 2006 - 2010 2011 - 2016 % Year Arrival in South Africa (n=502) Arrival in this City (n=502) AHMR, Vol.3 No2, May-August 2017 – Special Issue 793 Table 2: Previous Places of Residence in South Africa Cape Town Refugees % Limpopo Refugees % Major cities Johannesburg 15.7 29.0 Pretoria 4.2 12.9 Durban 3.2 5.0 Cape Town n/a 4.6 Port Elizabeth 1.6 2.4 Other urban centres Other Western Cape 0.2 1.2 Other Limpopo 1.2 16.9 Other Gauteng 0.0 1.6 Other KwaZulu-Natal 0.2 1.2 Other Eastern Cape 0.2 0.2 Free State 0.4 1.6 North West 0.2 1.6 Northern Cape 0.6 1.4 Mpumalanga 0.4 6.7 Note: Multiple response question In the Limpopo sample, previous places of residence were dominated by the large cities, particularly Johannesburg (29%) and Pretoria (13%), but also Durban (5%) and even Cape Town, which is nearly 2,000km away (5%). The movement of refugees from Limpopo to Cape Town was much more limited. As many as 17% of the Limpopo refugees had lived in another town in the province, suggesting the existence of some intra-provincial mobility. This kind of movement was virtually absent in the case of Cape Town and the Western Cape Province. In the Cape Town sample, the majority of the 28% who had lived in another urban area came from Johannesburg, with much smaller numbers from cities such as Pretoria and Durban. The major reason for relocation from large cities to small-town Limpopo concerns the pattern of violence against non-South African informal AHMR, Vol.3 No2, May-August 2017 – Special Issue 794 businesses (Crush et al., 2015b). Nationwide xenophobic attacks on migrants and refugees in 2008 led to considerable loss of life, damage to property and internal displacement (Hassim et al., 2008; Landau, 2012). In 2015, widespread xenophobic violence again wracked the country. Unlike the indiscriminate violence of 2008, these attacks mainly targeted small informal businesses run by migrants, asylum-seekers and refugees. Drawing on an extensive database of media coverage of xenophobic incidents, Crush and Ramachandran (2015a) chronicle an escalating pattern of xenophobic violence directed at migrants and refugees between 2008 and 2015. Refugee- owned businesses are regular targets of attacks and certain locations have witnessed collective violence in repeated cycles, including loss of store contents through looting and arson, damage to shop structure, forcible store closures, temporary or permanent displacement and loss of life (Crush & Ramachandran, 2015b). The impact of xenophobic violence is exacerbated by poor police response and follow-up. Refugees have little faith in police protection or judicial reparation, and even see the police as perpetrators or instigators of violence (Haile, 2012; Okpechi, 2011). Steinberg (2012) argues that it is the perceived failure of the authorities to control migration that has exacerbated retaliatory xenophobic mob violence. Some studies have suggested that attacks on refugee-owned businesses are instigated or orchestrated by South African competitors, including various shadowy informal business associations (Crush & Ramachandran, 2015a). This phenomenon – dubbed “violent entrepreneurship” by Charman and Piper (2012) – involves the use of intimidatory violence as a business strategy to drive non-South African competitors out of an area. The most frequent and intense xenophobic attacks and instances of violent entrepreneurship have occurred in low-income neighbourhoods in large cities. The in-depth interviews with Limpopo refugee entrepreneurs for this study confirmed that xenophobic violence was a major factor in relocation from cities of first refuge. One Somali refugee recounted how he had arrived in Johannesburg after a long and difficult journey through several countries; experiences consistent with those of other Somali refugees (Jinnah, 2010; Sadouni, 2009). In Johannesburg, he lived with his brother and neighbours from the same village in Somalia. His brother operated a business buying goods in the CBD and selling them at his AHMR, Vol.3 No2, May-August 2017 – Special Issue 795 shop in Soweto. He worked for his brother for a while but found the experience unnerving: “I did not want to operate in Soweto. The people there were nice but there were others who were just rough. I had seen two people being killed in broad daylight and they were all foreigners and their shops were robbed. So I wanted to go somewhere else.” With his brother’s assistance he bought a small spaza (informal shop) in Orange Farm south of Johannesburg and ran it for a year: Orange Farm was a good area for business but it was not safe. As a foreigner you are always conscious of your security and you can feel that this place is not good. It is far from the Johannesburg CBD and there are few police there. So even when there are robberies, the police come very late and sometimes they do not come at all. I was robbed seven times in the period that I stayed in Orange Farm. Most of the time the robbers would come when it is at night and you are still operating. They pounce on you with sticks, spanners or iron bars and they hit you hard. So I was almost killed twice and I thought this is enough. Let me leave this place. Then I left and came here (Interview with Somali Refugee, Polokwane, 12 March 2016). Experience of violent crime and fears over personal safety were recurrent motives for moving to Limpopo. Interviewees moved to escape being injured or murdered rather than specifically to set up a new business. One Eritrean refugee, for example, had his spaza looted and burnt to the ground: “I came here because I was running for my life. I was not thinking of doing business, but of surviving. I was almost killed that night” (Interview with Eritrean Refugee, 21 March 2016). Piper and Charman (2016) have argued that South African business owners are just as vulnerable to violent attacks and crime as their refugee counterparts and suggest that xenophobia is, therefore, not a factor. However, the interviewees in Limpopo clearly saw the violence in Johannesburg and other centres as motivated by xenophobia. As the former Orange Farm business owner noted: “There were many spaza shops around me, but they kept stealing from me. Is that not xenophobia? Why not steal from the locals? The customers sometimes harass you because you are a foreigner. They take AHMR, Vol.3 No2, May-August 2017 – Special Issue 796 your goods and are a problem in paying. So, yes, I have been affected a lot by xenophobia.” The province of Limpopo appears to be a safer haven and less inhospitable business environment than South Africa’s large cities. As many as 45% of sampled Cape Town refugees said that xenophobia had affected their business operations to some extent (20%) or a great deal (25%). The equivalent figures in Limpopo were 31% in total (19% to some extent and 12% not at all). Refugees in Cape Town, for example, appear to be more vulnerable to theft and physical attack from South African citizens. For example, in the sample, 57% of refugees in Cape Town reported often or sometimes having their goods and stock stolen, compared to 39% of those in Limpopo. Similarly, 54% of refugees in Cape Town often or sometimes had money stolen, compared to 32% in Limpopo. The incidents of prejudice and verbal and physical assault by citizens were similar in both places. But while relocation to Limpopo may lessen the chance of victimisation, it certainly does not eliminate it. Many of the refugees interviewed in Limpopo told stories of being robbed and having their business premises destroyed. However, police misconduct also emerged as a greater problem for refugees in Limpopo. Alfaro-Velcamp and Shaw (2016) show that police disregard for the rights of refugees and migrants is a significant problem in Cape Town and other large cities. This survey found that police harassment, demands for bribes, confiscation of goods and physical violence were all more common in Limpopo than in Cape Town (Table 3). AHMR, Vol.3 No2, May-August 2017 – Special Issue 797 Table 3: Challenges Faced by Refugee Entrepreneurs in South Africa Cape Town Refugees (% often/ sometimes) Limpopo Refugees (% often/ sometimes) Theft of goods/stock 56.7 38.7 Theft of money/income 44.2 31.8 Physical violence by South Africans 23.5 19.2 Confiscation of goods by police 10.2 19.1 Harassment/demands for bribes by police 10.7 26.5 Physical violence by police 6.8 8.8 Verbal insults 32.6 35.7 Prejudice against my nationality 48.0 47.9 Business Characteristics and Strategies Despite being located in very different parts of South Africa, and nearly 2,000 km apart, the sampled Cape Town and Limpopo refugee entrepreneurs engaged in a similarly wide range of economic activities. The vast majority in both locations is in the retail sector (75-79%), followed by services (25-28%) and manufacturing (4.4%-7.5%) (Table 4). The kinds of goods being sold and services offered are very similar as well. Among the sampled Limpopo refugee entrepreneurs, the most common items sold are clothing/footwear, confectionary, soft drinks and toiletries/cosmetics. In Cape Town, the most common items sold are cigarettes, clothing/footwear, personal accessories and confectionary and beverages (Table 5). Comparing the two, the main difference lies in the greater numbers of refugees selling confectionary, household products, and toiletries and cosmetics in Limpopo. AHMR, Vol.3 No2, May-August 2017 – Special Issue 798 Table 4: Economic Sector of Refugee Enterprises Cape Town Refugees Limpopo Refugees No. % No. % Retail, trade and wholesale 400 79.4 378 75.0 Manufacturing 38 7.5 22 4.4 Services 128 25.4 143 28.4 Other 1 0.2 1 0.2 Table 5: Goods and Services Offered by Refugee Entrepreneurs Cape Town Refugees % Limpopo Refugees % Retail Soft drinks (e.g. Coke, Fanta) 23.0 21.6 Cigarettes 21.4 18.5 Clothing and footwear 18.8 20.2 Personal accessories (e.g. bags, sunglasses) 14.5 18.5 Confectionary (sweets/candies and cakes) 13.7 20.2 Fresh produce (fruits and vegetables) 11.3 15.5 Toiletries and cosmetics 9.5 20.0 Cooked food ready to eat 7.5 6.0 Electronics 6.3 8.7 Arts and crafts 5.0 0.2 Books/newspapers 3.8 2.0 Household products 3.6 15.1 Hardware/tools 1.6 2.8 Music/film CDs/DVDs 1.0 4.2 Car parts 0.8 0.2 Livestock (e.g. chickens) 0.2 0.2 Alcohol 0.4 0.4 AHMR, Vol.3 No2, May-August 2017 – Special Issue 799 Other 9.9 18.5 Services Haircutting and braiding 10.9 19.2 IT/internet 4.4 1.0 Telephone 1.6 1.6 Shoe repairs 1.2 0.2 Traditional doctor 0.4 0.0 Accommodation 0.2 0.2 Car repairs 0.2 0.8 Medicine (pharmacy) 0.2 0.0 Transportation (taxi/passengers) 0.2 0.2 Car wash 0.0 0.8 Financial 0.0 0.2 Rentals 0.0 0.2 Transportation (goods) 0.0 0.2 Construction 0.0 0.0 Car parking/guarding 0.0 0.0 Other 5.4 5.8 Note: multiple response question The primary sources of business start-up capital in both areas were very similar, suggesting that being in a large city does not provide additional financing opportunities (Table 6). Approximately 85% of the sampled refugees in both Cape Town and Limpopo used their personal savings to start their businesses. Around 20% of both groups used loans from relatives, 12- 14% used loans from non-relatives and 12% used gifts from relatives. A small number in both areas used remittances from outside the country to establish their businesses. Only a handful of refugees in both areas were able to access funding from banks, NGOs or the UNHCR. The only difference worth noting was that more entrepreneurs in Limpopo were able to access goods on credit with which to start their businesses. Despite the similarities in sources of start-up capital, the amounts needed to establish a business did differ. In Limpopo, the sampled refugee entrepreneurs AHMR, Vol.3 No2, May-August 2017 – Special Issue 800 seemed to be able to start businesses with less capital (Figure 3). 34% percent had less than ZAR 5,000 compared to 21% of the Cape Town refugee entrepreneurs, and 63% had less than ZAR 10,000 compared to 52% of the Cape Town entrepreneurs. At the other end of the scale, 27% of the sampled Cape Town refugees had more than ZAR 50,000 in start-up capital compared to only 15% of those in Limpopo. Table 6: Primary Sources of Start-Up Capital Cape Town Refugees % Limpopo Refugees % Personal savings 85.9 84.3 Loan from relatives 19.8 20.2 Gift from relatives 13.7 11.7 Loan from non-relatives 12.1 14.3 Loan from church 2.4 0.2 Money from relatives in another country 2.4 2.4 Loan from mosque 1.4 0.2 Loan from informal financial institutions 1.2 1.4 Bank loan 0.8 1.2 Loan from government agency 0.2 0.0 Goods on credit 0.0 9.1 Loan or grant from NGO 0.0 1.4 Loan or grant from UNHCR 0.0 0.4 Usurers/mashonisa (money lenders) 0.0 0.2 Other source 2.2 6.9 Note: Multiple response question AHMR, Vol.3 No2, May-August 2017 – Special Issue 801 Figure 3: Amount of Start-Up Capital Used by Refugee Entrepreneurs Despite their lack of prior business experience, refugees in South Africa have developed a range of strategies to maximise their returns. Table 7 lists a variety of business strategies and shows how many sampled refugees employ them in the conduct of their business. The most common strategies include bulk purchasing, extended hours of operation, keeping business records, negotiating prices with suppliers, allowing customers to buy goods on credit and competitive prices. Refugees consult other entrepreneurs, suppliers and the media for information about the price of goods. These are all commonsense business strategies and certainly do not fall into the category of 0 5 10 15 20 25 30 35 40 Less than R5,000 R5,000 – R9,999 R10,000 – R19,999 R20,000 – R29,999 R30,000 – R49,999 R50,000 – R99,999 R100,000 – R199,999 R200,000 – R499,999 R500,000 – R999,999 R1,000,000 and above % S ta r t- U p A m o u n t (i n R a n d s ) Limpopo Refugees (n=468) Cape Town Refugees (n=462) AHMR, Vol.3 No2, May-August 2017 – Special Issue 802 underhand or secret tactics that the South African Minister of Small Business Development accused them of employing. There was very little difference in the frequency of use of business strategies by the two groups of refugees. However, living in a large city seems to provide more opportunity for buying in bulk, negotiating with suppliers and getting information on prices from other entrepreneurs. Almost one-third of the Limpopo refugees travel to Johannesburg to purchase supplies, while most Cape Town refugees obtain their goods in the city. This probably affords the latter greater opportunity for bulk buying, negotiation with suppliers and information-sharing. Table 7: Business Strategies of Refugee Entrepreneurs Cape Town Refugees % Limpopo Refugees % Obtain price information from other entrepreneurs 75.6 57.9 Purchase stock in bulk 69.7 56.3 Extended hours of operation 69.5 65.5 Negotiate prices with suppliers 67.7 59.8 Keep business records 59.4 55.7 Obtain price information by calling suppliers 57.2 61.2 Negotiate prices with customers 47.8 60.5 Offer credit to customers 42.0 44.8 Sell goods more cheaply than competitors 42.0 38.6 Charge different prices for different customers 37.1 30.3 Give discounts to regular customers 34.9 39.1 Obtain price information from media 31.9 28.1 Partner with other businesses to distribute risk 28.1 34.2 Engage in shareholding 10.3 25.9 Purchase insurance 3.2 7.0 AHMR, Vol.3 No2, May-August 2017 – Special Issue 803 Entrepreneurial Economies In this study, we used two measures of business success: (a) net monthly profit and (b) the difference between the current value of the enterprise and the initial capital outlay. With regards to the first indicator, net profit, it is clear that although refugees in Limpopo are able to start a business with a smaller capital outlay than those in Cape Town, their enterprises are not as profitable. The sampled Limpopo refugees earned less on average than their Cape Town counterparts: ZAR 7,246 per month compared to ZAR 11,315 in Cape Town (Table 8). Only one-third of the sampled Cape Town entrepreneurs made less than ZAR 5,000 per month compared to 58% of those in Limpopo. Or again, 70% of the Limpopo entrepreneurs make less than ZAR 7,500 per month compared to only 49% of those in Cape Town. With regards to the second indicator of business success the difference between the amount of start-up capital and the current value of the enterprise in both Cape Town and Limpopo there are clear indications of increased value. For example, 21% of the sampled entrepreneurs in Cape Town started with less than ZAR 5,000 but only 3% of the businesses were currently valued at less than ZAR 5,000 (Table 9). In Limpopo, the equivalent figures were 34% (start-up) and 14% (current value). In both locations, therefore, there was significant upward movement out of the lowest value category. A similar pattern can be observed with businesses that started with less than ZAR 20,000. In Cape Town, 51% fell into this start-up category but only 26% of businesses fell into this value category. The proportion of sampled businesses in the ZAR 50,000-plus category increased from 27% to 47% and in Limpopo from 15% to 42%. This suggests that higher value businesses may actually be performing better in Limpopo than they are in Cape Town. AHMR, Vol.3 No2, May-August 2017 – Special Issue 804 Table 8: Net Monthly Profit of Refugee Entrepreneurs Cape Town Refugees Limpopo Refugees % cumulative % % cumulative % <= ZAR2500 14.7 14.7 18.1 18.1 ZAR2,501-ZAR5,000 18.6 33.3 40.0 58.1 ZAR5,001-ZAR7,500 15.5 48.8 10.9 70.0 ZAR7,501-ZAR10,000 15.3 73.1 12.7 82.7 ZAR10,001-ZAR12,500 6.2 79.3 3.0 85.7 ZAR12,501-ZAR15,000 8.8 88.1 5.1 90.8 ZAR15,001-ZAR17,500 1.4 89.5 2.3 93.1 ZAR17,501-ZAR20,000 6.5 96.0 3.9 95.0 >ZAR20,000 4.0 100.0 5.0 100.0 AHMR, Vol.3 No2, May-August 2017 – Special Issue 805 The establishment and growth of sampled refugee businesses has economic spin-offs for a variety of South African stakeholders. The first beneficiaries are formal sector suppliers including wholesalers, supermarkets, fresh produce markets, retailers and manufacturers. Around two-thirds of the sampled refugees in both Cape Town and Limpopo purchase supplies from wholesalers, who are easily the largest beneficiaries of their patronage (Table 10). Other beneficiaries vary in relative importance, reflecting the difference between operating in a large city and a small town. For example, refugees in Cape Town are more likely to patronise supermarkets and factories while those in Limpopo are more likely to patronise small shops. The fact that more Cape Town refugees purchase directly from farms is primarily a reflection of the existence of a market gardening area (the Philippi Horticultural Area) within the city limits (Battersby-Lennard & Haysom, 2012). The average monthly spend at wholesalers was very similar in both Cape Town and Limpopo. Otherwise, the amounts spent varied considerably but, in general, the more important the source, the greater the amount is spent. Table 9: Increase in Value of Refugee Enterprises Cape Town Refugees Limpopo Refugees Start up % Current value % Start up % Current value % ZAR1,000,000 0.0 0.4 0.2 0.7 AHMR, Vol.3 No2, May-August 2017 – Special Issue 806 Table 10: Main Suppliers Used by Refugee Entrepreneurs Cape Town Refugees % Limpopo Refugees %. Cape Town Ave spend (ZAR/ month) Limpopo Ave spend (ZAR/ month) Wholesaler 64.1 57.9 34,565 34,908 Direct from factory 12.7 3.6 9,575 20,278 Supermarkets 9.7 4.2 12,253 3,979 Fresh produce markets 5.6 4.2 28,021 2,000 Non-South African informal producers/retailers 4.8 8.5 9,121 15,549 Small shops/retailers 4.2 17.9 3,738 6,876 South African informal sector producers/retailers 2.6 3.8 7,631 3,859 Direct from farms 1.6 0.8 19,625 4,875 Other sources 5.6 3.6 14,855 4,586 Note: Multiple response question The second major beneficiary of the activities of refugee entrepreneurs is the South African treasury. While most businesses operate in the informal sector and are too small to pay income tax, they do pay value-added tax (VAT) on most goods purchased from formal sector suppliers. One Ethiopian refugee in a Limpopo focus group observed that they not only paid VAT on goods but also were unable to claim rebates: There are many things we do to help the economy. We operate our businesses and we buy our stock from the wholesalers. We pay VAT there so our money is going into the economy. We pay a lot of money that way. Some of our people go to buy goods that are worth ZAR 200,000 or even more. Some buy every week and some buy every day. So imagine if you are paying ZAR 200,000 for goods and VAT is 14% so that is more than AHMR, Vol.3 No2, May-August 2017 – Special Issue 807 ZAR 25,000 I am paying to the government. Even after paying this large amount of VAT we suffer because we cannot claim some of it because we do not have the documents. We are not registered. If we were registered at the end of the year we would claim some of the VAT like South Africans do. We are actually paying more and contributing more through VAT than most South African small businesses (Focus Group Participant, Burgersfort, 1 April 2016). Another participant claimed that business registration was impossible because “government does not want to register us.”1 Third, as these refugees noted, they pay heavy rents to South African property owners: South Africans are surviving on us because we pay them money to rent their shops. I pay ZAR 4,000 per month to my landlord to use the shop. That is a lot of money. How many local people can afford to pay that money? Locals do not pay that kind of money when they rent. They will pay ZAR 1,000 or ZAR 1,500 only, not ZAR 4,000. We pay because we have no choice, we want to do business and survive.2 While over half of both groups of refugees paid rent to a private South African owner, this was more common in Limpopo (almost 66%) (Table 11). However, the average monthly rentals paid to landlords were very similar: ZAR 4,838 in Cape Town and ZAR 4,555 in Limpopo. Renting also provides refugees with a modicum of protection. As one noted, “If your landlord is respected or feared, then you are okay because the thugs will not come and break in.”3 The Cape Town refugee entrepreneurs were more likely to own their business premises although the numbers were small (7% compared to 2% in Limpopo). Fourth, the municipal government has a direct financial interest in its dealings with refugees. Particularly in Cape Town, refugees pay into municipal coffers through rent for business sites. This amounted to 22% of the refugee entrepreneurs in Cape Town compared to only 4% in Limpopo. These rents 1 Focus Group Participant, Burgersfort, 1 April 2016. 2 Ibid. 3 Ibid. AHMR, Vol.3 No2, May-August 2017 – Special Issue 808 were higher in Cape Town, an average of ZAR 879 per month compared to only ZAR 311 per month in Limpopo. Additionally, as many as 28% of Limpopo- based refugees (and 21% of those in Cape Town) pay an annual license fee to the municipality. The cost of a business license is much higher in Cape Town at ZAR 1,959 per year compared to only ZAR 752 per year in Limpopo. Table 11: Occupancy Status of Business Premises Cape Town Refugees % Limpopo Refugees % Cape Town rent (R/month) Limpopo rent (R/month) Pay rent to private owner who is a South African (company or individual) 53.6 65.9 7,541 4,838 Pay rent to council/municipality 21.5 3.8 860 879 Pay rent to private owner who is not a South African (company or individual) 9.4 8.8 2,800 3,180 Owner or part-owner 7.4 2.2 Rent-free, with permission 5.2 6.4 Rent-free, without permission 1.2 10.6 Share space/premises with others 0.0 0.4 Other 1.6 1.8 Fifth, there is a growing consensus in the research literature that migrant businesses in the informal economy create jobs for South Africans, and that this needs to be acknowledged by government. As one refugee observed: “The government does not want to accept that we are contributing. When they talk AHMR, Vol.3 No2, May-August 2017 – Special Issue 809 about us they make it look like we are only taking, taking and taking from the South Africans.”4 Another pointed out the mutual benefits of employing South Africans: We employ a lot of South Africans. If your business starts to grow, you employ, not only our people from Ethiopia, but also locals. They help with customers and other things. They speak the language and understand quickly what the customer wants. Even though I understand the language, there are words that I do not know and so it is better when you employ a South African, they will talk and agree and I will communicate with my employee. The locals need work and we also need workers, so it is a mutual benefit.5 Some refugees also find that employing South Africans is a form of security against theft, although it can be a double-edged sword: It is good security because they will know the local thugs and thieves and so they may tell them to leave you alone. If you have a bad employee, they may connive with the thieves and steal from you, so it is both ways. Sometimes we employ locals that we know, when we know their parents and we talk to them so that they do not steal and run away. It is better to employ someone from around the area, someone in the community.6 Around half (52%) of the Cape Town refugee entrepreneurs and just under half (45%) of the Limpopo entrepreneurs employ people in their businesses (Table 12). In terms of the pro file of employees, the Cape Town and Limpopo entrepreneurs were equally likely to employ South Africans (around 50% of the total number of jobs created). While the Limpopo entrepreneurs favoured female employees (51% versus 45% of total employees), both groups preferred to hire South African women over men (with 65-70% of South African employees in both sites being female). What this suggests is that, in both large cities and small urban areas, refugee entrepreneurs may be 4 Ibid. 5 Ibid. 6 Ibid. AHMR, Vol.3 No2, May-August 2017 – Special Issue 810 providing jobs for more South African women than men and thereby contributing to lowering the female unemployment rate. Table 12: Employment Generation by Refugee Enterprises Cape Town Refugees Limpopo Refugees No. of employees % No. of employees % South African Males 66 17.7 37 11.8 South African Females 120 32.2 82 26.1 Home Country Males 101 27.2 81 25.8 Home Country Females 28 7.5 26 8.3 Other Country Males 35 9.4 37 11.8 Other Country Females 22 5.6 51 16.2 Finally, it is clear from the accounts of refugees in both Cape Town and Limpopo that one of the primary beneficiaries of their activities in the informal economy is the South African consumer who can access goods and services much more cheaply than from formal sector suppliers. These include necessities such as cheaper food for the food insecure, luxuries such as household and personal products and services such as panel-beating. For example, one group of five Zimbabwean refugees in Cape Town (including a former teacher), operate a panel-beating and spray-painting business in an industrial area of the city. In a focus group, they discussed at length how and why they established the business, their business challenges and the nature of the service they offer to South Africans: We are very good at this business. There are many people who come here because they cannot afford to repair their cars in these expensive garages, so we are offering them services otherwise they would not be driving their cars. That is a good service we are offering. Some of the customers actually go to the garages first and the charges there will make them come to us (Focus Group Discussion, Cape Town, 22 February 2016). AHMR, Vol.3 No2, May-August 2017 – Special Issue 811 These refugees also argued, in a prescient manner, that their activities saved the South African Government and the UNHCR from having to support them. As one commented: “In Europe the governments look after the refugees and every month they get paid like they are working. But here we are working for ourselves and are saving the government a lot of money.”7 Conclusion This paper aims to contribute to the literature on urban refugee livelihoods in the Global South and, in particular, to a small but growing body of work on urban refugee entrepreneurship in cities of refuge (Campbell, 2005, 2006; Gastrow & Amit, 2013; Omata, 2012; Pavanello et al., 2015; Thompson, 2016). The concept of ‘refugee economies’ is a valuable starting point for restoring agency, self-reliance and innovation to populations all too often represented as passive victims. One of the defining characteristics of many large cities in the rapidly urbanising South is the high degree of informality of shelter, services and economic livelihoods. As Simone (2004) argues, this involves “highly mobile and provisional possibilities for how people live and make things, how they use the urban environment and collaborate with one another.” Furthermore, these urban spaces provide for “upscaling a variety of entrepreneurial activities through the dense intersections of actors from different countries and situations.” It is these dynamic, shifting and dangerous informal urban spaces in which refugees often arrive, with few resources other than a will to survive, a few social contacts and a drive to support themselves in the absence of financial support from the host government and international agencies. South Africa is sometimes hailed as having an extremely progressive refugee protection regime by the standards of the rest of Africa. Refugees enjoy freedoms denied in other countries, including the right to move and live anywhere and the right to work and be self-employed. Many opt for employment and self-employment in the informal economy because they find it extremely difficult to access the formal labour market or to establish and operate a business in the formal economy. Newly-arrived refugees, including 7 Focus Group Discussion, Cape Town, 22 February 2016. AHMR, Vol.3 No2, May-August 2017 – Special Issue 812 those with advanced skills and professional qualifications, initially find themselves accepting menial work on construction sites or as dish-washers in restaurants. Those with relatives or friends already in the country often work in their informal businesses until they have saved enough money to launch their own. They start small and, with extremely hard work, self-sacrifice and hardship, their businesses take off. Thus, South Africa provides an important example of refugee self-reliance, motivation and agency through informal entrepreneurship. At the same time, South Africa is one of the most xenophobic countries in the world and migrants of any kind are extremely unwelcome. Large-scale violence directed at migrants and refugees wracked the country in 2008 and again in 2015. In this span of time, extreme xenophobia, in the form of violent attacks on migrant and refugee-owned shops and small businesses, has escalated. The state refuses to acknowledge that xenophobia exists, much less that something should be done about it. Attacks on foreign businesses are dismissed as general criminality, which might be an acceptable rhetorical position if there was evidence that the police provide adequate protection and justice for the victims. The evidence of human rights observers and previous studies, and confirmed here, is that policing is lackadaisical at best and certainly does very little to protect refugees from re-victimisation in their chosen country of refuge. The fact that informal businesses started and run by refugees have become a particular target means that the South African city has become a hazardous place for refugee entrepreneurs. Refugee economic opportunity and entrepreneurial activity is likely to vary significantly between camp and urban environments. This paper has addressed the question of variability between urban environments within the same destination country by comparing refugee entrepreneurship in the large city of Cape Town with a population of over 3 million and several small towns in the predominantly rural province of Limpopo, none with a population larger than 150,000. The research shows that refugee entrepreneurial activity in Limpopo is a more recent phenomenon and is largely a function of refugees moving away from large cities such as Johannesburg where their businesses and lives are in greater danger. While Somali refugees predominate in Cape AHMR, Vol.3 No2, May-August 2017 – Special Issue 813 Town and Ethiopian refugees in Limpopo, the refugee populations in both areas are equally diverse and tend to be engaged in the same wide range of activities. Less start-up capital is needed in Limpopo but the refugees in both areas pursue similar business strategies and make similar contributions to the local economies. They face many of the same business challenges, including problems with documentation and the refusal of banks to offer credit, although small town policing appears to be harsher and more corrupt. While South African consumers clearly benefit from and appreciate their presence, migrant entrepreneurs are more vulnerable to xenophobic violence in Cape Town than Limpopo. In short, different urban geographies do shape the local nature of refugee entrepreneurial economies, but there are also remarkable similarities in the manner in which unconnected refugee entrepreneurs establish and grow their businesses in large cities and small provincial towns. Acknowledgements We wish to thank the UNHCR Geneva for funding the Refugee Economic Impacts SAMP research project on which this paper is based. References Achiume, T. 2014. Beyond prejudice: Structural xenophobic discrimination against refugees. Georgetown Journal of International Law, 45: 323-381. Alfaro-Velcamp, T. and Shaw, M. 2016. ‘Please GO HOME and BUILD Africa’: Criminalising migrants in South Africa. Journal of Southern African Studies, 42: 993-998. Amit, R. 2011. No refuge: Flawed status determination and the failures of South Africa's refugee system to provide protection. International Journal of Refugee Law, 23: 458-488. Amit, R. and Kriger, N. 2014. Making migrants 'il-legible': The policies and practices of documentation in post-apartheid South Africa. Kronos, 40: 269- 290. http://ijrl.oxfordjournals.org/content/23/3/458.short http://ijrl.oxfordjournals.org/content/23/3/458.short AHMR, Vol.3 No2, May-August 2017 – Special Issue 814 Basardien, F., Parker, H., Bayat, M., Friedrich, C. and Appoles, S. 2014. Entrepreneurial orientation of spaza shop entrepreneurs: Evidence from a study of South African and Somali owned spaza shop entrepreneurs in Khayelitsha. Singaporean Journal of Business Economics and Management Studies, 2: 45-61. Battersby-Lennard, J. and Haysom, G. 2012. ‘Philippi Horticultural Area’ AFSUN and Rooftops Canada Abri International, Cape Town and Toronto. Betts, A., Bloom, L., Kaplan, J. and Omata, N. 2014. Refugee economies: Rethinking popular assumptions. Humanitarian Innovation Project Report, Refugee Studies Centre, Oxford: University of Oxford. Buscher, D. 2011. New approaches to urban refugee livelihoods. Refuge, 28: 17-29. Campbell, E. 2005. Formalising the Informal Economy: Somali Refugee and Migrant Trade Networks in Nairobi. Global Migration Perspectives No. 47, Geneva: Global Commission on International Migration. Campbell, E. 2006. Urban refugees in Nairobi: Problems of protection, mechanisms of survival, and possibilities for integration. Journal of Refugee Studies, 19: 396-413. Charman, A. and Piper, L. 2012. Xenophobia, criminality and violent entrepreneurship: Violence against Somali shopkeepers in Delft South, Cape Town, South Africa. South African Review of Sociology, 43: 81-105. Crush, J. and Chikanda, A. 2014. Forced migration in Southern Africa. In: Fiddian-Qasmiyeh, E., Loescher, G., Long, K. and Sigona, N. (Eds.). The Oxford Handbook of Refugee and Forced Migration. Oxford: Oxford University Press. Crush, J. and Ramachandran, S. 2015a. Doing business with xenophobia. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. AHMR, Vol.3 No2, May-August 2017 – Special Issue 815 Crush, J. and Ramachandran, S. 2015b. Migration myths and extreme xenophobia in South Africa. In: Arcarazo, D. and Wiesbrock, A. (Eds.). Global Migration: Old Assumptions, New Dynamics, Volume 3. Santa Barbara: Praeger. Crush, J., Chikanda, A. and Skinner, C. (Eds.). 2015a. Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. Crush, J., Skinner, C. and Chikanda, A. 2015b. Informal Migrant Entrepreneurship and Inclusive Growth in South Africa, Zimbabwe and Mozambique. SAMP Migration Policy Series No. 68, Cape Town: SAMP. Desai, A. 2008. Xenophobia and the place of the refugee in the rainbow nation of human rights. African Sociological Review, 12: 49-68. De Vriese, M. 2006. Refugee Livelihoods: A Review of the Evidence. Evaluation and Policy Analysis Unit, EPAU/2006/04. Geneva: UNHCR. Fatoki, O. and Chindoga, L. 2011. An investigation into the obstacles to youth entrepreneurship in South Africa. International Business Research, 4: 161-169. Gastrow, V. and Amit, R. 2013. Somalinomics: A Case Study on the Economics of Somali Informal Trade in the Western Cape. ACMS Research Report, Johannesburg: Wits University. Gastrow, V. and Amit, R. 2015. The role of migrant traders in local economies: A case study of Somali spaza shops in Cape Town. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. Gordon, S. 2016. Welcoming refugees in the rainbow nation: Contemporary attitudes towards refugees in South Africa. African Geographical Review, 35: 1- 17. Greyling, T. 2015. The Expected Well-Being of Urban Refugees and Asylum- Seekers in Johannesburg. ERSA Working Paper, 507. Cape Town: ERSA. Gwija, S., Eresia-Eke, C. and Iwu, C. 2014. Challenges and prospects of youth entrepreneurship development in a designated community in the Western Cape, South Africa. Journal of Economics and Behavioral Studies, 6: 10-20. AHMR, Vol.3 No2, May-August 2017 – Special Issue 816 Haile, T. 2012. Immigrants and Xenophobia: Perception of Judicial System Personnel and Experiences of Ethiopian Immigrants in Accessing the Justice System in Newcastle, South Africa. MA Thesis (Unpublished), KwaZulu-Natal: University of KwaZulu-Natal. Handmaker, J., de la Hunt, L. and Klaaren, J. (Eds.). 2011. Advancing Refugee Protection in South Africa 2nd Edition. New York and Oxford: Berghahn Books. Hassim, S., Kupe, T. and Worby, E. (Eds.). 2008. Go Home or Die Here: Violence, Xenophobia and the Reinvention of Difference in South Africa. Johannesburg: Wits University Press. Idemudia, E., Williams, J. and Wyatt, G. 2013. Migration challenges among Zimbabwean refugees before, during and post arrival in South Africa. Journal of Injury and Violence Research, 5: 17-27. Jacobsen, K. 2005. The Economic Lives of Refugees. Bloomfield, CT: Kumarian Press. Jacobsen, K. 2006. Refugees and asylum seekers in urban areas: A livelihoods perspective. Journal of Refugee Studies, 19: 273-286. Jinnah, Z. 2010. Making home in a hostile land: Understanding Somali identity, integration, livelihood and risks in Johannesburg. Journal of Sociology and Social Anthropology, 1: 91-99. Kavuro, C. 2015. Refugees and asylum seekers: Barriers to accessing South Africa's labour market. Law, Democracy and Development, 19: 232-260. Kew, J. 2015. Africa’s Young Entrepreneurs: Unlocking the Potential for a Brighter Future. Ottawa: IDRC. Koizumi, K. and Hoffsteadter, G. (eds.) 2015). Urban Refugees: Challenges in Protection, Services and Policy. London: Routledge. Landau, L. 2008. Protection as capability expansion: Practical ethics for assisting urban refugees. In: Hollenbach, D. (Ed.). Refugee Rights: Ethics, Advocacy, and Africa. Washington DC: Georgetown University Press. Landau, L. (Ed.). 2012. Exorcising the Demons Within: Xenophobia, Violence and Statecraft in Contemporary South Africa. Johannesburg: Wits University Press. https://www.ncbi.nlm.nih.gov/pubmed/?term=Idemudia%20ES%5BAuthor%5D&cauthor=true&cauthor_uid=22095004 https://www.ncbi.nlm.nih.gov/pubmed/?term=Williams%20JK%5BAuthor%5D&cauthor=true&cauthor_uid=22095004 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3591728/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3591728/ http://www.scielo.org.za/scielo.php?script=sci_serial&pid=2077-4907&lng=en&nrm=iso AHMR, Vol.3 No2, May-August 2017 – Special Issue 817 Lyytinen, E. and Kullenberg, J. 2013. Urban Refugee Research: An Analytical Report. New York: International Rescue Committee and Women’s Refugee Commission. Mothibi, K., Roelofse, C. and Tshivhase, T. 2015. Xenophobic attacks on foreign shop owners and street vendors in Louis Trichardt Central Business District, Limpopo Province. Journal for Transdisciplinary Research in Southern Africa, 11: 151-162. Okpechi, A. 2011. Access to Justice by Refugees and Asylum Seekers in South Africa. Ph.D. Thesis, Cape Town: University of Cape Town. Omata, N. 2012. Refugee Livelihoods and the Private Sector: Ugandan Case Study. Refugee Studies Centre Working Paper Series No. 86, Oxford: University of Oxford. Omata, N. and Kaplan, J. 2013. Refugee Livelihoods in Kampala, Nakivale and Kyangwali Refugee Settlements: Patterns of Engagement with the Private Sector. Refugee Studies Centre Working Paper Series No. 95, Oxford: University of Oxford. Omata, N. and Weaver, N. 2015. Assessing Economic Impacts of Hosting Refugees: Conceptual, Methodological and Ethical Gaps. Refugee Studies Centre Working Paper Series No. 11, Oxford: University of Oxford. Owen, J. 2015. Congolese Social Networks: Living on the Margins in Muizenberg, Cape Town. Lanham, MD: Rowman and Littlefield. Pavanello, S. Elhawary, S. and Pantuliano, S. 2015. Hidden and exposed: Urban refugees. In: Koizumi, K. and Hoffstaedter, G. (Eds.). Urban Refugees: Challenges in Protection, Services and Policy. London: Routledge. Piper, L. and Charman, A. 2016. Xenophobia, price competition and violence in the spaza sector in South Africa. African Human Mobility Review, 2: 332-361. Polzer Ngwato, T. 2013. Policy Shifts in the South African Asylum System: Evidence and Implications. Pretoria: LHR and ACMS. AHMR, Vol.3 No2, May-August 2017 – Special Issue 818 Ramatheje, G. and Mtapuri, O. 2014. A dialogue on migration: Residents and foreign nationals in Limpopo, South Africa. Mediterranean Journal of Social Sciences, 5: 573-585. Roever, S. and Skinner, C. 2016. Street vendors and cities. Environment and Urbanization, 28: 359-374. Sadouni, S. (2009). God is Not Unemployed: Journeys of Somali Refugees in Johannesburg. African Studies, 68: 235-249. Simone, A. 2004. People as infrastructure: Intersecting fragments in Johannesburg. Public Culture, 16: 407-442. Smit, R. and Rugunanan, P. 2014. From precarious lives to precarious work: The dilemma facing refugees in Gauteng, South Africa. South African Journal of Sociology, 45: 4-26. Steinberg, J. 2012. Security and disappointment: Policing, freedom and xenophobia in South Africa. British Journal of Criminology, 52: 345-360. Tawodzera, G., Chikanda, A., Crush, J. and Tengeh, R. 2015. International Migrants and Refugees in Cape Town’s Informal Economy. SAMP Migration Policy Series No. 70, Cape Town: SAMP. Thompson, D. 2016. Risky business and geographies of refugee capitalism in the Somali migrant economy of Gauteng, South Africa. Journal of Ethnic and Migration Studies, 42: 120-135. UNHCR (United Nations High Commissioner for Refugees). 2014. Global Strategy for Livelihoods, A UNHCR Strategy, 2014-2018. Geneva: United Nations. Williams, C. 2013. Out of the shadows: Classifying economies by the extent and nature of employment in the informal economy. International Labour Review, 154: 331-351. Williams, C. and Gurtoo, A. 2010. Evaluating competing theories of street entrepreneurship: Some lessons form a study of street vendors in Bangalore, India. International Entrepreneurship and Management Journal, 8: 391-409. https://www.researchgate.net/journal/1469-9451_Journal_of_Ethnic_and_Migration_Studies https://www.researchgate.net/journal/1469-9451_Journal_of_Ethnic_and_Migration_Studies AHMR, Vol.3 No2, May-August 2017 – Special Issue 819 Zack, T. 2015. ‘Jeppe’: Where low-end globalisation, ethnic entrepreneurialism and the arrival city meet. Urban Forum, 26: 131-150. AHMR, Vol.3 No2, May-August 2017 – Special Issue 820 Comparing Refugee and South African Migrant Enterprise in the Urban Informal Sector Jonathan Crush* and Cameron McCordic** Abstract Comparisons between the informal business operations of South Africans and international migrants are increasingly common. The conventional wisdom is that survivalist South Africans are being displaced by entrepreneurial migrants with a long tradition of informal enterprise. This paper is the first attempt to explicitly compare the informal enterprises established by refugees and South African migrants in urban areas. The paper is based on a comparative analysis of over 2,000 refugee and South African informal enterprises. The stereotyping of refugees in public discourse as undermining and destroying South African competitors is far-removed from the reality. The idea that refugees have a competitive advantage as experienced entrepreneurs is also clearly fallacious. Refugees are extremely motivated, hard-working and dedicated, and employ a number of legitimate business strategies to achieve success. To suggest, on the other hand, that South African migrants are poor business people is equally fallacious. While refugees seem able to access greater amounts of start-up capital (although neither they nor South Africans can access formal bank loans), both groups are seemingly able to grow their businesses. Thus, there is a need for much greater nuance in policy and academic discussions about the impact of refugee migration on the South African informal economy. Keywords Informal sector, business strategies, xenophobia, refugee entrepreneurs, South African migrant entrepreneurs. * International Migration Research Centre, Balsillie School of International Affairs, 67 Erb St West, Waterloo, Ontario, Canada N2L6C2. Email address: jcrush@balsillieschool.ca ** Balsillie School of International Affairs, 67 Erb St West, Waterloo, Ontario, Canada N2L6C2. Email address: cmccordic@balsillieschool.ca mailto:cmccordic@balsillieschool.ca AHMR, Vol.3 No2, May-August 2017 – Special Issue 821 Introduction In 2014, South Africa’s Minister of Small Business Development, Lindiwe Zulu, publicly compared South African and migrant informal entrepreneurs. She suggested that South Africans were largely inept business owners who should learn from the business practices of their foreign counterparts who were “better at running shops than the local owners” (Zwane, 2014). At the same time, South Africans were said to be at a natural disadvantage because they had no history of entrepreneurship. The reason for refugee success, she continued, is that business “is in their blood […] from the moment they are born, they are introduced to trade. Their mothers, uncles, everyone trades.” As a result, they “operate in the same communities in which we fail” (Zwane, 2014). Such stereotypical comparisons of refugee and South African informal business are echoed in the research literature. There is a common idea in the business literature, for example, that poor black South Africans lack entrepreneurial ambition and this, in turn, helps explain the relatively small size of the South African informal economy and the high rate of local informal business failure (Hutchinson & de Beer, 2013; Iwu et al., 2016; Ligthelm, 2011; Preisendörfer et al., 2012a, 2012b, 2014a, 2014b). This paper sets out to examine and contest the contrasting stereotypes that surround both South African and refugee entrepreneurs. Comparisons between the informal business operations of South Africans and international migrants are increasingly common. The conventional wisdom is that “survivalist” South Africans in the informal economy are being displaced by “entrepreneurial” migrants (Charman et al., 2012). South Africans supposedly display a “survivalist mentality and one dimensional [business] strategy,” leading to poorer performance than migrants (Basardien et al., 2014: 57). Comparing South African and Somali spaza shop owners in Cape Town, Basardien et al. (2014) found that the latter scored better on various indicators of entrepreneurial orientation, including achievement, innovation, personal initiative and autonomy. In addition, migrant businesses grew faster and created more jobs than South African businesses. By contrast, some have suggested that business failure is not inevitable and that South African survivalists can grow their enterprises and create jobs (Choto et al., 2014; Iwu AHMR, Vol.3 No2, May-August 2017 – Special Issue 822 et al., 2016). Other studies have suggested that the gap between South African and migrant entrepreneurs is not as great as is commonly supposed. One study of 500 retail enterprises in Gauteng, for example, found that motivations to start a business did not differ significantly between South Africans and immigrants (Radipere, 2012; Radipere & Dhliwayo, 2014). Callaghan and Venter’s (2011) study of street traders in inner-city Johannesburg concluded that South Africans were actually more innovative than migrants, although they did not display the same levels of proactiveness and competitive aggression. While migrant traders had earned more than their South African counterparts in 2008 and 2010, in 2009 the South Africans were the higher earners (Callaghan, 2013). What is often overlooked in the public discourse and research literature about informal business competition between South Africans and non-South Africans is the fact that many South Africans in the informal economy are themselves migrants. By consistently representing business and other competition as a conflict between ‘South Africans’ and ‘foreigners,’ the fact that much of the supposed competition is between two groups of migrants is lost. Therefore, this paper aims to systematically compare a group of South African and non-South African migrant entrepreneurs and to assess the similarities and differences between them. The two groups are (i) refugees (holders of Section 24 permits) in Cape Town and Limpopo and (ii) South African migrants operating businesses in the same localities. The survey drew a sample of 1,068 South African migrant entrepreneurs and 1,008 refugee entrepreneurs (split approximately equally across the two locations of Cape Town and urban Limpopo). The maximum variation sampling methodology used to select the refugees for interview is explained in Crush et al. (2015). Exactly the same procedure was used to select the comparator group of South African informal business; that is, random selection of respondents within each area identified. For the purposes of this comparative analysis, we have combined the two sub-groups of refugees (in Cape Town and Limpopo) into one group and have done the same with the South Africans. AHMR, Vol.3 No2, May-August 2017 – Special Issue 823 Motivating Entrepreneurship South Africans and refugees appear to face very different livelihood prospects in the country’s urban areas. Although South Africa does not have refugee encampment policy and refugees are permitted by law to pursue employment, there is much evidence to suggest that they face considerable barriers in accessing the formal labour market (Crea et al., 2016; Jinnah, 2010; Kavuro, 2015; Rugunanan & Smit, 2011). They have been shut out of the security industry (where many were initially employed) and they face considerable hurdles in getting employers to accept their documentation. South Africans, on the other hand, should theoretically have none of these problems but they face other hurdles including limited skills and training, job competition, and high rates of unemployment (currently around 30% nationally and as high as 45% amongst urban youth) (Graham & De Lannoy, 2016; Klasen & Woolard, 2009). South African migrants to the cities often end up living in informal settlements far from formal job opportunities, and also have to compete in the job market with long-time residents of the city who have a significant geographical and networking advantage. For both sets of migrants, then, the informal economy can often be the only livelihood niche they can find. The general literature on informal entrepreneurship conventionally divides participants into survival (or necessity) entrepreneurs and opportunity entrepreneurs (Williams, 2007, 2015; Williams & Gurtoo, 2012; Williams & Youseff, 2014). The former are driven to participate purely by the need to survive and because they have no other choice. The latter choose to work in the informal sector because they see greater opportunities for economic advancement, they prefer to work for themselves rather than for others or they feel that they have the right aptitude. Distinguishing between these two types of entrepreneur and their likely differences in entrepreneurial motivation and orientation has generated a large body of empirical and methodological literature. In the South African context, studies of entrepreneurial motivation have sought to go beyond the idea of survivalism and demonstrate that many participants in the informal economy are not driven there out of desperation but are highly motivated entrepreneurs AHMR, Vol.3 No2, May-August 2017 – Special Issue 824 (Callaghan & Venter, 2011; Fatoki & Patswawairi, 2012; Khosa & Kalitanyi, 2015; Venter, 2012). One of the most common ways of deciding what lies behind personal decisions to establish an informal enterprise is to measure what is known as entrepreneurial motivation. This involves the development of possible reasons why the informal enterprise was started and then asking respondents to rank them on a Likert scale from 1 (no importance) to 5 (extremely important). In this study, both refugees and South African migrants were presented with 24 different possibilities to rate. A mean score was calculated for each group on each statement (Table 1). For ease of interpretation, we have grouped the 24 statements under 4 main themes (a) economic survival; (b) provision of employment or a service to others; (c) business experience and appeal and (d) entrepreneurial orientation. Two things immediately stand out from a descriptive comparison of means. First, both refugees and migrants tend to assign the same relative importance to each of the 24 factors, which might suggest that they have a similar motivational profile, rating the same factors as relatively important and unimportant. The second notable finding is that almost across the board, even on statements that had a low mean score, refugees scored more highly than South African migrants. This could indicate a greater general degree of commitment to participation in the informal economy amongst refugees. AHMR, Vol.3 No2, May-August 2017 – Special Issue 825 Table 1: Entrepreneurial Motivation of Refugees and South Africans South Africans (Mean Score) Refugees (Mean Score) Economic Survival/Financial Support of Dependants I needed more money just to survive 4.10 4.31 I wanted to give my family greater financial security 3.69 3.97 I was unemployed and unable to find a job 3.43 2.89 I wanted to make more money to send to my family in my home area/country 3.00 3.57 I had a job but it did not pay enough 2.29 2.24 I had a job but it did not suit my qualifications and experience 1.44 1.62 Providing Employment/Product/Service I wanted to provide a product/service to South Africans 2.74 3.41 I wanted to contribute to the development of South Africa 2.69 3.35 I wanted to provide a service/product to non-South Africans/migrants and refugees 2.46 3.00 I wanted to provide employment for people from my home area/country 1.93 2.25 I wanted to provide employment for members of my family 2.19 2.27 AHMR, Vol.3 No2, May-August 2017 – Special Issue 826 I wanted to provide employment for other South Africans 2.10 2.29 Business Experience/Appeal I wanted more control over my own time/to be my own boss 3.08 3.72 I have always wanted to run my own business 3.06 3.75 Support and help in starting my business was available from other South Africans/refugees 2.05 3.03 I decided to go into business in partnership with others 1.62 2.37 My family has always been involved in business 1.81 2.34 Entrepreneurial Orientation I have the right personality to run my own business 3.01 3.45 I wanted to do something new and challenging 2.83 3.34 I like to learn new skills 2.83 3.41 I enjoy taking risks 2.73 3.24 I like to challenge myself 2.84 3.37 I wanted to increase my status in the community 2.48 2.99 I wanted to compete with others and be the best 2.48 3.07 Significantly, the only two reasons for entrepreneurship on which South Africans scored higher than refugees were “I was unemployed and unable to find a job” and “I had a job but it did not pay enough.” This suggests that for South African migrants, informal sector participation is more closely tied to the absence of formal employment than it is for refugees. Of the four groups of factors, economic survival motivations scored most highly for both groups, and providing an employment or service was the least important. The highest AHMR, Vol.3 No2, May-August 2017 – Special Issue 827 single factor for both groups was the need for more money to survive (both with means over 4.0). Also very important for both was the desire to provide family with greater financial security and the desire to make more money to remit to family at home. In other words, financial support of dependants is a strong motivating factor for informal sector entrepreneurship. Neither group was highly motivated by a desire to provide employment for others, but refugees were ironically much more likely to be motivated by a desire to provide a service or product to South Africans (3.41 versus 2.74) and to contribute to the development of South Africa (3.35 versus 2.69). Although both groups said that wanting to run their own business and be their own boss was important to them, the refugees scored significantly higher on both factors. One of the major differences between the two was the amount of help and support they could count on from others, with refugees scoring much higher than South Africans (3.03 versus 2.05). Refugees were also consistently more positive about their personal aptitude for running a business. This is clear in the grouping of entrepreneurial orientation factors where refugees scored above 3.0 on 6 of the 7 factors, compared to South Africans who scored above 3.0 on only 1 of the 7 factors. While these frequency distributions tell an interesting story about the differences and similarities between South African migrant and refugee entrepreneurs, it is difficult to gauge their statistical significance. The main challenge is that the dependent variable for the comparison (the importance ranking for each variable) is at an ordinal level of measurement with varying distributions across each sampled group. This means that we need to use non- parametric tests of difference and bin the motivation factors into binary level indicators. Each indicator was therefore assigned two values: not important (1 in the original scale) and important (2-5 in the original scale). A combination of odds ratio calculations and Pearson’s Chi-Square Test of independence were used to test for significance. The odds ratio calculations show how migrant status is associated with a change in the odds of ranking each motivation factor (where a value greater than 1 indicates increased odds and less than 1 indicates decreased odds). These calculations are supported by 95% confidence intervals and the p-values taken from a Chi-Square analysis (where AHMR, Vol.3 No2, May-August 2017 – Special Issue 828 an alpha of 0.05 is used as a threshold for a statistically significant difference in the distribution of scores across the two groups) (Table 2). Table 2: Odds Ratio Calculations of Motivational Factors Entrepreneurial Motivation Factor Odds Ratio 95% Confidence Interval Pearson Chi- Square Df P- Value (2- sided) Lower Upper I was unemployed and unable to find a job** 0.496 0.411 0.599 54.036 1 <.001 I had a job but it did not pay enough 1.012 0.851 1.204 .019 1 0.895 I had a job but it did not suit my qualifications and experience** 1.631 1.322 2.012 21.088 1 <.001 I wanted to provide employment for members of my family** 1.344 1.128 1.600 10.970 1 0.001 I wanted to provide employment for people from my home area/country** 1.840 1.540 2.198 45.499 1 <.001 I wanted to provide employment for 1.597 1.341 1.902 27.682 1 <.001 AHMR, Vol.3 No2, May-August 2017 – Special Issue 829 other South Africans** I needed more money just to survive** 1.770 1.275 2.457 11.880 1 0.001 I wanted to give my family greater financial security** 1.651 1.294 2.105 16.483 1 <.001 I wanted to make more money to send to my family in my home area/country** 2.942 2.393 3.618 109.114 1 <.001 I decided to go into business in partnership with others** 2.931 2.423 3.545 126.855 1 <.001 Support and help in starting my business was available from other South Africans/refugees ** 3.155 2.635 3.778 160.774 1 <.001 My family has always been involved in business** 2.149 1.793 2.575 69.422 1 <.001 I wanted to provide a service/product 2.085 1.741 2.496 64.788 1 <.001 AHMR, Vol.3 No2, May-August 2017 – Special Issue 830 to non-South Africans/migrant s and refugees** I wanted to provide a product/service to South Africans** 2.550 2.103 3.093 92.934 1 <.001 I have always wanted to run my own business** 2.806 2.268 3.471 94.245 1 <.001 I have the right personality to run my own business** 2.176 1.787 2.650 61.018 1 <.001 I wanted to do something new and challenging** 2.289 1.890 2.772 73.209 1 <.001 I like to learn new skills** 2.590 2.128 3.153 92.620 1 <.001 I enjoy taking risks** 2.299 1.901 2.781 75.044 1 <.001 I like to challenge myself** 2.466 2.028 2.998 83.975 1 <.001 I wanted more control over my own time/to be my own boss** 2.887 2.331 3.574 98.785 1 <.001 I wanted to increase my 2.193 1.832 2.626 74.095 1 <.001 AHMR, Vol.3 No2, May-August 2017 – Special Issue 831 status in the community** I wanted to compete with others and be the best** 2.581 2.151 3.096 106.434 1 <.001 I wanted to contribute to the development of South Africa** 3.677 3.001 4.505 166.788 1 <.001 * p<.05 on both Pearson’s Chi-Square and Fisher’s Exact Test ** p<.01 on both Pearson’s Chi-Square and Fisher’s Exact Test Note: Odds Ratios measure change in odds from refugee to South African entrepreneurs The major conclusions from the analysis are as follows: first, refugee entrepreneurs have about 50% lower odds of starting their business because of being unable to find a job. Second, refugees had four times the odds of desiring to contribute to the development of South Africa and three times the odds of stressing the importance of obtaining help from others in starting their business and going into partnership with others. Third, refugees had nearly three times the odds of starting a business with the intention of remitting money to family at home. Finally, refugees had two to three times the odds of assigning importance to the range of personal entrepreneurial orientation factors. Contrasting Business Profiles The survey highlighted a number of similarities and differences in the informal business activities of refugee and South African migrant enterprises. Firstly, more South Africans had been in business for a longer period of time (Table 3 and Figure 1). For example, 19% of the South African businesses were established before 2000, compared to only 2% of the refugee businesses. However, the majority of all businesses were started in the last decade, with 61% of refugee businesses and 44% of South African businesses established AHMR, Vol.3 No2, May-August 2017 – Special Issue 832 after 2010. This finding is certainly consistent with the general perception that refugees have been entering the informal economy in growing numbers. Table 3: Year of Business Establishment Year South Africans Refugees No. % No. % <= 1990 46 4.4 1 0.1 1991-1995 49 4.7 1 0.1 1996-2000 115 11.0 18 1.8 2001-2005 124 11.9 70 7.1 2006-2010 246 23.6 293 29.6 2011-2016 462 44.3 608 61.4 Total 1,042 100.0 991 100.0 Figure 1: Year of business establishment 0 10 20 30 40 50 60 70 <= 1990 1991 – 1995 1996 – 2000 2001 – 2005 2006 – 2010 2011 – 2016 % Year of Business Establishment South Africans Refugees AHMR, Vol.3 No2, May-August 2017 – Special Issue 833 Second, since both groups are migrants to the city, it is important to see if they go into business as soon as they arrive or if business start-up comes at a later time. Only 32% of refugees and 21% of South Africans started a business within the first year of arrival (Table 4). This general pattern of a greater time lapse on the part of the South Africans is further demonstrated by the fact that 41% of them started their business within two years of arrival, compared to 61% of the refugees. Both groups have similar numbers who waited 3 to 10 years, but starting a business after 10 years or more was definitely a South African preserve (at 24% compared to 3% of refugees). The general time lapse in both groups indicates that immediate start-up is not an option for most. Rather, both tend to work first in the formal or informal economy, often to raise the start-up capital to branch out on their own. Table 4: Time Lapse Between Year of Migration and Business Start-Up South Africans Refugees Years No. % No. % 0 213 21.4 304 32.0 1-2 196 19.7 277 29.2 3-5 173 17.4 201 21.2 6-10 172 17.3 140 14.7 >10 242 24.3 28 2.9 Third, it is theoretically possible that the shorter time-lag between migration and start-up amongst refugees is also because they have prior business experience. This would certainly be consistent with the views of Minister Zulu summarised at the beginning of this paper (Zwane, 2014). The respondents were all asked what their main occupation was prior to leaving their home country or area. Only 9% of the South Africans said they were operating their own informal sector business. The figure for refugees was higher, at 18%, but AHMR, Vol.3 No2, May-August 2017 – Special Issue 834 this does not suggest a massive competitive advantage conferred by prior experience. In other words, over 80% of the refugee entrepreneurs were not operating an informal sector business prior to migrating to South Africa. The stereotypical idea that refugees somehow have business “in their blood” is, therefore, not supported by the evidence of this survey. Fourth, the survey found that the majority of enterprises of both refugees and South African migrants are located in the retail sector (Table 5). A small number of businesses (9% of refugees and 6% of South Africans) are involved in more than one sector; for example, a business that manufactured and sold arts and crafts would count as both a retail and manufacturing enterprise. Or a business offering a service, such as a hair salon, may also be involved in retailing products. At this sectoral level of analysis, it appears that there is potential for significant intra-sectoral competition between the two groups. However, if the activity profile is disaggregated, the picture is more nuanced (Table 6). Table 5: Sectors of Informal Business Operation South Africans Refugees Sector No. % No. % Retail, Trade and Wholesale 828 77.5 778 77.2 Services 262 24.5 271 26.9 Manufacturing 41 3.8 60 6.0 Other 4 0.4 2 0.2 Note: Multiple response question AHMR, Vol.3 No2, May-August 2017 – Special Issue 835 Table 6: Main Goods and Services Provided % of Total Enterprises Owned by South Africans % of Total Enterprises Owned by Refugees % of South Africans Selling Product or Service % of Refugee s Selling Product or Service Retail Food and Beverages Fruit and vegetables 27.1 13.4 68.2 31.8 Cooked food (ready to eat) 18.4 6.7 74.3 25.7 Confectionary 17.5 17.0 52.2 47.8 Canned Drinks 13.1 22.3 38.4 61.6 Livestock (e.g. chickens) 1.7 0.2 90.0 10.0 Alcohol 0.9 0.4 71.4 28.6 Personal and Household Goods Cigarettes 13.3 19.9 40.8 59.2 Clothing and footwear 7.7 19.5 29.4 70.6 Accessories (bags, sunglasses) 6.1 16.5 28.1 71.9 AHMR, Vol.3 No2, May-August 2017 – Special Issue 836 Toiletries and Cosmetics 3.9 14.8 22.0 78.0 Household Products 3.4 9.3 27.7 72.3 Electronics 1.8 7.5 20.0 80.0 CDs/DVDs 1.1 2.6 31.6 68.4 Books/newspapers 0.7 2.9 19.4 80.6 Other Goods Arts and Crafts 1.1 2.6 31.6 68.4 Hardware/tools 1.1 2.2 35.3 64.7 Car parts 0.5 0.5 50.0 50.0 Other 15.3 14.2 53.2 46.8 Services Haircutting and braiding 6.4 15.1 30.9 69.1 Car washing 3.4 0.4 90.0 10.0 Car parking/guarding 2.3 0.0 100.0 0.0 Shoe repairs 2.0 0.7 75.0 25.0 Car repairs 1.2 0.5 27.8 72.2 Telephone 1.2 1.6 44.8 55.2 IT/Internet 0.9 2.7 27.0 73.0 Transportation (taxi/passengers) 0.5 0.2 71.4 0.2 Rentals 0.4 0.1 80.0 20.0 AHMR, Vol.3 No2, May-August 2017 – Special Issue 837 Financial (loans) 0.3 0.1 75.0 25.0 Accommodation 0.2 0.2 50.0 50.0 Construction (building) 0.2 0.0 100.0 0.0 Traditional doctor 0.2 0.2 50.0 50.0 Transportation (goods) 0.2 0.1 66.6 33.4 Medicine (pharmacy) 0.1 0.1 50.0 50.0 Other 5.3 5.6 50.4 49.6 Manufacturing Sewing/tailoring 1.7 2.0 47.4 52.6 Arts and crafts 0.7 0.4 66.6 33.4 Shoe repair 0.5 0.5 50.0 50.0 Furniture making 0.4 0.6 40.0 60.0 Security (gates and burglar bars) 0.1 0.4 20.0 80.0 Waste recycling 0.0 0.1 0.0 100.0 Other 0.6 1.9 24.0 76.0 Table 6 shows that at least some South Africans and refugees are involved in every activity. However, they also tend to occupy and dominate different niches. South Africans are more strongly represented in food retail (the main exception being confectionary and the sale of canned drinks with roughly equal participation). Around 70% of the entrepreneurs who were selling fresh produce and cooked food were South Africans. On the other hand, over 70% of those selling most types of personal and household products were refugees. In AHMR, Vol.3 No2, May-August 2017 – Special Issue 838 the service sector, refugees dominate hair cutting and braiding, as well as car repairs and IT. South Africans tend to dominate shoe repairs, transportation and car washing and guarding. In the manufacturing sector, there is less differentiation, although the overall numbers of participants are small compared with retail and services. Fourth, there was a significant difference in the amount of start-capital used by the two groups (Figure 2). Almost 80% of the South Africans started their businesses with less than ZAR 5,000, while the equivalent figure was only 27% for refugees. At the other end of the spectrum, only 6% of the South Africans had start-up capital of more than ZAR 20,000, compared to 43% of the refugees. This certainly suggests that refugees have access to greater amounts of start-up capital but it may also be that the barriers to entry are much lower in the food sector (which is dominated by South Africans) as the initial spend on stock is likely much lower than for businesses selling personal and household goods. It is significant that of the 28% of refugees who started with less than ZAR 5,000, most were food retailers. Figure 2: Amount of start-up capital 0 10 20 30 40 50 60 70 80 90 % Startup Capital Amount South Africans Refugees AHMR, Vol.3 No2, May-August 2017 – Special Issue 839 Fifth, both groups had added value to their businesses since start-up (Table 7). For example, while 78% of South Africans started with less than ZAR 5,000, only 40% valued their enterprise as still less than ZAR 5,000 (a fall of 28%). Similarly, with refugees the equivalent figures were 28% and 9% (a fall of 19%). The proportion of South African businesses with a current value of over ZAR 20,000 was 25% (compared to only 6% at start-up). In the case of refugees, the equivalent figures were 70% and 43%). In other words, 19% of South Africans and 27% of refugees had moved up into the highest value bracket. Table 7: Current Business Value South Africans Refugees No. % No. % Less than ZAR5,000 349 40.1 75 8.5 ZAR5,000-9,999 177 20.3 83 9.4 ZAR10,000-19,999 128 14.7 110 12.4 ZAR20,000-29,999 57 6.6 110 12.4 ZAR30,000-49,999 59 6.8 115 13.0 ZAR50,000-99,999 46 5.3 148 16.7 ZAR100,000-199,999 28 3.2 135 15.2 ZAR200,000-499,999 18 2.1 93 10.5 ZAR500,000-999,999 3 0.3 12 1.4 <=ZAR1,000,000 5 0.6 5 0.6 Total 870 100.0 886 100.0 AHMR, Vol.3 No2, May-August 2017 – Special Issue 840 To assess the statistical significance of these differences, key variable comparisons were drawn out from the frequency distributions and binned into binary-level variables. These included (a) year of establishment (<=2010 and > 2010); (b) start-up capital (ZAR 5,000); and (c) current business value (ZAR 5,000). The odds ratio calculations performed in Table 8 provide convergent validity for the observed frequency distributions. Independent of the influence of any other variables, the South African entrepreneurs had almost twice the odds of running a business established before 2011, almost ten times the odds of starting a business with less than ZAR 5,000 and almost seven times the odds of currently running a business valued at less than ZAR 5,000. All of these comparisons yielded p- values less than the alpha of 0.01 on both the Pearson’s Chi-Square Test and the Fisher’s Exact Test. Table 8: Odds Ratio Calculations of Business Characteristics 95% C.I. P-Values Variables Odds Ratio Lower Upper Chi- Square Fisher's Exact Test Established business before 2011** 1.993 1.67 2.379 <.001 <.001 Less than R5,000 in start- up amount** 9.579 7.772 11.807 <.001 <.001 Less than R5,000 in current value ** 7.243 5.515 9.514 <.001 <.001 * p<.05 on both Pearson’s Chi-Square and Fisher’s Exact Test ** p<.01 on both Pearson’s Chi-Square and Fisher’s Exact Test Note: Odds Ratios measure change in odds from refugee to South African entrepreneurs AHMR, Vol.3 No2, May-August 2017 – Special Issue 841 Business Strategies Given the official and business literature perception that non-South Africans are much better at running businesses than their South African counterparts, it is important to see if the two groups pursue different business strategies and activities. The first point of comparison concerns where the two groups choose to locate their business operations. In the case of Cape Town, there are areas of the city where each group tends to dominate: refugee businesses are more common in the CBD and Bellville, for instance, while South Africans are more commonly located along transport routes in and out of the city (such as on streets and at taxi ranks and bus terminals). This difference is clear from Table 9. Half of the South Africans operate stalls on roadsides and 21% operate at taxi ranks. This compares with only 31% and 2% of refugees, respectively. The other major difference is that half of the refugees operate from a fixed shop or workshop, compared to only 8% of the South Africans. Table 9: Usual Location of Business Activities South Africans Refugees Business Location No. % No. % Temporary stall on the street/roadside 290 27.2 166 16.5 Permanent stall on the street/roadside 246 23.0 147 14.6 Taxi rank 221 20.7 22 2.2 In my home 102 9.6 34 3.4 No fixed location, mobile 87 8.1 26 2.6 Workshop or shop 86 8.1 525 52.1 Permanent stall in a market 87 8.1 106 10.5 Bus terminal 52 4.9 9 0.9 Railway station 21 2.0 1 0.1 AHMR, Vol.3 No2, May-August 2017 – Special Issue 842 Vehicle (car, truck, motor bike, bike) 19 1.8 3 0.3 In customer’s home 16 1.5 16 1.6 Craft market 6 0.6 7 0.7 Restaurant or hotel 2 0.2 8 0.8 Other 37 3.5 38 3.8 Note: Multiple choice question In addition to the observed variations in business location, the reasons for locational decisions also varied between the two groups (Table 10). When compared with the refugee entrepreneurs, the South African migrant entrepreneurs had greater odds of choosing a business location based on it having the greatest number of customers, the tradition of doing business in a location, the cheapness of land and a limited number of police in the area. The refugees had higher odds of choosing their business location based on the other locational factors, especially access to services, property rentals, safety concerns and distance from other competitors. Table 10: Odds Ratio Calculations of Reasons for Business Location 95% C.I. for O.R. P-Values Reasons Odds Ratio Lower Upper Chi- Square Fisher's Exact Test Place with greatest number of customers** 2.231 1.707 2.915 <.001 <.001 Access to services such as water/electricity** 0.341 0.284 0.409 <.001 <.001 Have a permit to operate there** 0.746 0.624 0.893 .001 .001 Rents are cheaper 0.969 0.807 1.163 0.732 0.744 AHMR, Vol.3 No2, May-August 2017 – Special Issue 843 Safer than other locations** 0.673 0.564 0.803 <.001 <.001 Due to passing traffic 0.949 0.775 1.163 0.615 0.641 Close to home 0.869 0.73 1.036 0.117 0.118 Own/rent the land** 0.458 0.381 0.55 <.001 <.001 Close to other enterprises* 0.825 0.692 0.984 0.032 0.035 Distant from other competitors** 0.647 0.534 0.784 <.001 <.001 Always done business there** 1.262 1.061 1.501 0.009 0.009 Close to public transport 0.987 0.811 1.2 0.893 0.92 Cheap land** 1.739 1.424 2.123 <.001 <.001 Few or no police* 1.251 1.017 1.54 0.034 0.035 * p<.05 on both Pearson’s Chi-Square and Fisher’s Exact Test ** p<.01 on both Pearson’s Chi-Square and Fisher’s Exact Test Note: Odds Ratios measure change in odds from refugee to South African entrepreneurs Refugee entrepreneurs were much more likely than South Africans to rent their business premises (Table 11). Almost 60% of them paid rent to a South African landlord. Another 13% paid rent to the municipality (as did 10% of the South Africans). Nearly 50% of South Africans operated their businesses rent- free (compared to only 5% of refugees). What this means, in effect, is that around three-quarters of South Africans do not pay any rent for their premises, while over 80% of refugees do. The refugee entrepreneurs also pay a higher monthly rent, on average, than those South Africans who do pay rent (ZAR 4,000 per month versus ZAR 2,820 per month). In effect, many South Africans are able to augment their household income through renting business premises to refugees and therefore benefit from their presence. AHMR, Vol.3 No2, May-August 2017 – Special Issue 844 Table 11: Occupancy/Tenure Status of Business Premises South Africans Refugees Tenure Status No. % No. % Rent-free, with permission 276 26.1 58 5.8 I own it/am part owner 256 24.2 48 4.8 Rent-free, without permission (squatting) 214 20.3 59 5.9 Pay rent to private owner who is a South African (company or individual) 145 13.7 595 59.7 Pay rent to council/municipality 104 9.8 126 12.7 Share space/premises with others 28 2.7 2 0.2 Pay rent to private owner who is not a South African (company or individual) 17 1.6 91 9.1 Other 16 1.5 17 1.7 Another area of business strategy comparison concerns where the two groups source their goods and whether they tend to patronise the same outlets (Table 12). Most of the respondent refugees buy their supplies at wholesalers while South Africans patronise wholesalers and supermarkets in almost equal numbers. South African respondents also obtain goods from fresh produce markets and direct from farms in greater numbers. With the exception of factory purchase, refugees tend to spend more on average at all outlets. For example, while fewer refugees patronise supermarkets, their average monthly spend is ZAR 8,693 compared with only ZAR 3,219 by the South Africans. In total, the South African respondents spend more than the refugees at supermarkets, fresh produce markets and buying direct from farms. Refugees spend five times as much on average at wholesalers and a great deal more in total (ZAR 21 million compared to less than ZAR 2 million). AHMR, Vol.3 No2, May-August 2017 – Special Issue 845 Table 12: Patronage of Different Suppliers South Africans Refugees Source % of Total Using Source Mean Monthly Spend (ZAR) Total Annual Spend (ZAR) % Using Source Mean Monthly Spend (ZAR) Total Annual Spend (ZAR) Wholesaler 27.8 6,248 1,855,724 61.0 34,728 21,357,827 Supermarkets 27.3 3,219 936,642 8.5 8,693 747,640 Small shops/retailers 11.4 1,798 219,407 11.0 6,282 697,350 Fresh produce markets 9.6 4,751 489,364 4.9 16,869 826,600 Direct from farms 9.6 8,875 905,270 1.2 14,708 176,500 Direct from factory 7.4 32,216 2,545,050 8.2 11,924 977,800 SA informal sector enterprises 5.5 1,956 115,391 3.2 5,391 172,520 Non-SA informal sector enterprises 3.2 1,607 54,650 6.7 13,246 887,500 Other sources 5.1 3,486 188,250 4.6 10,838 498,525 There is a common assumption that other strategies adopted by refugees give them a strong competitive advantage over South Africans. In addition to greater business acumen and skills, they have been viewed, inter alia, as securing discounts through group purchasing, offering credit to consumers, operating for longer hours and selling goods more cheaply. Statistical AHMR, Vol.3 No2, May-August 2017 – Special Issue 846 comparison of these, and other, business strategies indicates their relative importance to each group (Table 13). The refugees had lower odds of adjusting their operating hours to times of the day when there were most customers and purchasing insurance. South African migrant entrepreneurs had lower odds of operating for extended hours (0.743) and individual bulk purchasing (0.67). However, they were two to five times as likely to keep business records (0.475), sell goods more cheaply than competitors (0.395), purchase in bulk with others (0.244) and negotiate with suppliers (0.340). Table 13: Odds Ratio Calculations of Business Strategies 95% C.I. for O.R. P-Values Odds Ratio Lower Upper Chi- Square Fisher's Exact Test I open my business only during the periods of the day when I have the most customers* 1.213 1.020 1.443 0.029 0.030 I purchase insurance 1.078 0.732 1.587 0.703 0.768 I offer credit for customers 0.918 0.770 1.093 0.336 0.348 I extend my hours of operation** 0.743 0.620 0.890 .001 0.001 I purchase stock in bulk myself** 0.670 0.562 0.800 <.001 <.001 I change different prices for different customers** 0.660 0.545 0.799 <.001 <.001 AHMR, Vol.3 No2, May-August 2017 – Special Issue 847 I look for the cheapest prices for goods by consulting the media** 0.656 0.538 0.800 <.001 <.001 I engage in shareholding** 0.562 0.437 0.722 <.001 <.001 I keep records of my business accounts** 0.475 0.398 0.566 <.001 <.001 I look for the cheapest prices for goods by asking other entrepreneurs** 0.439 0.367 0.525 <.001 <.001 I sell goods more cheaply than my competitors** 0.395 0.325 0.481 <.001 <.001 I purchase stock in bulk together with others** 0.344 0.278 0.424 <.001 <.001 I negotiate prices with my suppliers** 0.340 0.284 0.407 <.001 <.001 I look for cheapest prices for goods by calling suppliers** 0.230 0.191 0.278 <.001 <.001 * p<.05 on both Pearson’s Chi-Square and Fisher’s Exact Test ** p<.01 on both Pearson’s Chi-Square and Fisher’s Exact Test Note: Odds Ratios measure change in odds from refugee to South African entrepreneurs The final point of business strategy comparison concerns the hiring practices of the two groups of entrepreneurs. Almost half of the refugee entrepreneurs have paid employees compared to only 21% of the South Africans. The refugees in this sample provided three times as many jobs as the South AHMR, Vol.3 No2, May-August 2017 – Special Issue 848 Africans. Table 14 provides statistical confirmation of the greater employment-generating potential of refugees using odds ratio, Chi-Square and Fisher’s Exact Test calculations. A breakdown of employees by sex and national origin shows some differences in the hiring patterns of the two groups. In total, 9% of refugees hire South African men full-time and 1% part-time. The equivalent figures for South African enterprises are 8% and 4%. Refugees show a preference for hiring South African women over men, with 16% employing women full-time and 4% part-time (compared to 9% and 1% for men). In the sample as a whole, 30% are South Africans employed by refugees and 28% are South Africans employed by other South Africans. Table 14: Odds Ratio Calculations for Employment-Generation 95% C.I. P-Values Variables Odds Ratio Lower Upper Chi- Square Fisher's Exact Test Currently have employees** 0.273 0.225 0.332 <.001 <.001 * p<.05 on both Pearson’s Chi-Square and Fisher’s Exact Test ** p<.01 on both Pearson’s Chi-Square and Fisher’s Exact Test This suggests that both refugee and South African enterprises create jobs for South Africans in roughly equal numbers. The major difference is in the employment of non-South Africans. Less than 5% of the total number of employees are non-South Africans employed by South Africans, whereas 39% are non-South Africans employed by refugees. Conclusion This paper is the first attempt to explicitly compare the informal enterprises established by different categories of migrant in South African urban areas. This comparative analysis of refugees and internal migrants suggests that AHMR, Vol.3 No2, May-August 2017 – Special Issue 849 there is a need for much greater nuance in policy and academic discussions about the impact of refugee migration on the South African informal economy. The stereotyping of refugees in public discourse as undermining and destroying South African competitors is clearly far-removed from the reality. While refugees seem able to access greater amounts of start-up capital (although neither they nor South Africans can access formal bank loans), both groups are seemingly able to grow their businesses. This is partly because they tend to occupy different niches in the informal economy, with South Africans focused more on the food sector and refugees focused more on services and retailing household goods. This may help explain another difference between the two, with refugees tending to patronise wholesalers for their supplies and South Africans purchasing from supermarkets and fresh produce markets. The idea promulgated by the South African Minister of Small Business Development, that refugees have a competitive advantage as experienced or “in their blood” entrepreneurs, is clearly fallacious. South Africa’s refugee legislation and restrictive employment policies mean that working for, and then establishing, an informal enterprise is virtually the only available livelihood option. But to argue that refugees come to South Africa with a pre- existing skill and business experience is misplaced. Instead, refugees (like small business owners everywhere) are extremely motivated, hard-working and dedicated. They employ a number of business strategies to achieve monetary success, although business expansion is hampered by the fact that only a portion of business profits can be reinvested in the business as the rest go to support dependants in South Africa and the home country. These strategies are not illegal or even underhand but are quite transparent and could be emulated. To suggest, on the other hand, that South African migrants are poor business people, as the minister also suggested, is just as fallacious. It is true that the odds of refugees pursuing a particular strategy (such as giving goods on credit) are generally higher than a South African doing so, but this does not mean that no South Africans pursue the strategy, as many clearly do. Instead of constantly pitting refugees against South Africans as the official mind likes to do, it would be more productive to treat them in policy terms as a single group attempting, often against considerable odds, to establish and AHMR, Vol.3 No2, May-August 2017 – Special Issue 850 grow a small business in a hostile or indifferent economic and political environment. Acknowledgements We wish to thank the UNHCR in Geneva for funding the Refugee Economic Impacts SAMP research project on which this paper is based. References Basardien, F., Parker, H., Bayat, M., Friedrich, C. and Sulaiman, A. 2014. Entrepreneurial orientation of spaza shop entrepreneurs: Evidence from a study of South African and Somali-owned spaza shop entrepreneurs in Khayelitsha. Singaporean Journal of Business Economics and Management Studies, 2: 45-61. Callaghan, C. 2013. Individual values and economic performance of inner-city street traders. Journal of Economics, 4: 145-156. Callaghan, C. and Venter, R. 2011. An investigation of the entrepreneurial orientation, context and entrepreneurial performance of inner-city Johannesburg street traders. Southern African Business Review, 15: 28-48. Charman, A., Petersen, L. and Piper, L. 2012. From local survivalism to foreign entrepreneurship: The transformation of the spaza sector in Delft, Cape Town. Transformation, 78: 47-73. Choto, P., Tengeh, R. and Iwu, C. 2014. Daring to survive or to grow? The growth aspirations and challenges of survivalist entrepreneurs in South Africa. Environmental Economics, 5: 93-101. Crea, T., Loughry, M., O’Halloran, C. and Flannery, G. 2016. Environmental risk: Urban refugees’ struggles to build livelihoods in South Africa. International Social Work, 60: 667-682. Crush, J., Ramachandran, S. and Pendleton, W. 2013. Soft targets: Xenophobia, public violence and changing attitudes to migrants in South Africa after May 2008. SAMP Migration Policy Series No. 64, Cape Town: SAMP. Crush, J. and Chikanda, A. eds., 2015. Mean streets: Migration, xenophobia and informality in South Africa. Southern African Migration Programme. AHMR, Vol.3 No2, May-August 2017 – Special Issue 851 Fatoki, O. and Patswawairi, T. 2012. The motivations and obstacles to immigrant entrepreneurship in South Africa. Journal of Social Science, 32: 133- 142. Graham, L. and De Lannoy, A. 2016. Youth unemployment: What can we do in the short run? From . Hutchinson, M. and de Beer, M. 2013. An exploration of factors that limit the long-term survival and development of micro and survivalist enterprises: Empirical evidence from the informal economy in South Africa. Mediterranean Journal of Social Sciences, 4: 237-245. Iwu, C., Gwija, S., Tengeh, R., Cupido, C. and Mason, R. 2016. The unique role of the survivalist retail entrepreneur in job creation and poverty reduction: Implications for active stakeholder participation. Acta Universitatis Danubius (Economica), 12: 16-37. Jinnah, Z. 2010. Making home in a hostile land: Understanding Somali identity, integration, livelihood and risks in Johannesburg. Journal of Sociology and Social Anthropology, 1: 91-99. Kavuro, C. 2015. Refugees and asylum seekers: Barriers to accessing South Africa's labour market. Law, Democracy and Development, 19: 232-260. Khosa, R. and Kalitanyi, V. 2015. Migration reasons, traits and entrepreneurial motivation of African immigrant entrepreneurs: Towards an entrepreneurial migration progression. Journal of Enterprising Communities: People and Places in the Global Economy, 9: 132-155; Klasen, S. and Woolard, I. 2009. Surviving unemployment without state support: Unemployment and household formation in South Africa. Journal of African Economies, 18: 1-51. Ligthelm, A. 2011. Survival analysis of small informal businesses in South Africa, 2007–2010. Eurasian Business Review, 1: 160-179. Preisendörfer, P., Bitz, A. and Bezuidenhout, F. 2012a. Business start-ups and their prospects of success in South African townships. South African Review of Sociology, 43: 3-23. AHMR, Vol.3 No2, May-August 2017 – Special Issue 852 Preisendörfer, P., Bitz, A. and Bezuidenhout, F. 2012b. In search of black entrepreneurship: Why is there a lack of entrepreneurial activity among the black population in South Africa. Journal of Developmental Entrepreneurship, 17: 1-18. Preisendörfer, P., Perks, S. and Bezuidenhout, F. 2014a. Do South African townships lack entrepreneurial spirit? International Journal of Entrepreneurship and Small Business, 22: 159-178. Preisendörfer, P., Bitz, A. and Bezuidenhout, F. 2014b. Black entrepreneurship: A case study on entrepreneurial activities and ambitions in a South African township. Journal of Enterprising Communities, 8: 162-179. Radipere, N. 2012. An Analysis of Local and Immigrant Entrepreneurship in the South African Small Enterprise Sector (Gauteng Province). PhD Thesis, Pretoria: University of South Africa. Radipere, S. and Dhliwayo, S. 2014. An analysis of local and immigrant entrepreneurs in South Africa’s SME sector. Mediterranean Journal of Social Sciences, 5: 189-198. Rugunanan P. and Smit, R. 2011. Seeking refuge in South Africa: Challenges facing a group of Congolese and Burundian refugees. Development Southern Africa, 28: 705-718. Venter, R. 2012. Entrepreneurial values, hybridity and entrepreneurial capital: Insights from Johannesburg's informal sector. Development Southern Africa, 29: 225-239. Williams, C. 2007. Entrepreneurs operating in the informal economy: Necessity or opportunity driven? Journal of Small Business and Entrepreneurship, 20: 309-320. Williams, C. 2015. Explaining the informal economy: An exploratory Evaluation of competing perspectives. Industrial Relations, 70: 741-765. Williams, C. and Gurtoo, A. 2012. Evaluating competing theories of street entrepreneurship: Some lessons from a study of street vendors in Bangalore, India. International Entrepreneurship and Management Journal, 8: 391-409. AHMR, Vol.3 No2, May-August 2017 – Special Issue 853 Williams, C. and Youssef, Y. 2014. Is informal sector entrepreneurship necessity- or opportunity-driven? Some lessons from urban Brazil. Business and Management Research, 3: 41-53. Zwane, T. 2014. Spazas: Talking Shop is Good for Business. Mail & Guardian, November 6, 2014. AHMR, Vol.3 No2, May-August 2017 – Special Issue 854 Security Risk and Xenophobia in the Urban Informal Sector Sujata Ramachandran*, Jonathan Crush** and Godfrey Tawodzera*** Abstract Whenever there are violent attacks on refugee and migrant businesses in South Africa’s informal sector, politicians, officials and commissions of enquiry deny that xenophobia is a driving force or indeed exists at all in the county. A new strain of nativist research in South Africa does not deny the existence of xenophobia but argues that it is an insignificant factor in the violence. It is argued that because South African and non-South African enterprises are equally at risk, the reasons for the violence are internal to the sector itself. This paper critiques this position on the basis of the results of a survey of over 2,000 enterprises in the contrasting geographical sites of Cape Town and small town Limpopo. The survey results reported in this paper focus on security risks and the experience of victimisation and the experience of the two groups of enterprise operator are systematically compared. The findings show that while both cohorts experience many of the same security risks, refugee operators are significantly more exposed to most threats including verbal abuse, theft, unprovoked attacks and harassment by law enforcement agencies. Far from being irrelevant, xenophobia is an important additional risk for refugee entrepreneurs, as they themselves clearly recognise. Keywords Security risks, informal sector, xenophobia, refugee entrepreneurs, South African entrepreneurs. * Southern African Migration Programme, International Migration Research Centre, Balsillie School of International Affairs, 67 Erb St West, Waterloo, Ontario, Canada N2L6C2. ** International Migration Research Centre, Balsillie School of International Affairs, 67 Erb St West, Waterloo, Ontario, Canada N2L6C2. Email address: jcrush@balsillieschool.ca *** Department of Geography and Environmental Studies, University of Limpopo, P. Bag X1106, Sovenga. AHMR, Vol.3 No2, May-August 2017 – Special Issue 855 Introduction In early 2015, South Africa experienced a new wave of violent attacks on migrants, the culmination of several years of more localised but escalating collective violence targeting migrant-owned businesses in the country (Crush & Ramachandran, 2015a). After May 2008, when South Africa had witnessed a previous round of large-scale violence against migrants, a decisive shift had occurred in state discourses and management of xenophobia. The government’s stance towards xenophobia had moved from a lack of acknowledgement of its presence and policy neglect to public rejection and denial of its very existence in the country (Crush & Ramachandran, 2014). This shift framed the way in which government interpreted and responded to renewed violence in 2015. This response had three key elements: first, denial of the existence of xenophobia and as an explanation for the violence; second, blaming migrants for their own victimisation; and third, attributing ongoing violence against migrant-owned businesses as the work of criminal elements (Bekker, 2015; Crush & Ramachandran, 2015a; Misago, 2016). The major formal government response to the international and broader African outrage over the violence was the establishment of an Inter-Ministerial Committee on Migration (IMC) housed in the Presidency. As many as fifteen government ministers sat on the IMC, which was expanded in March 2017 to include all provincial premiers and the South African Local Government Association (SALGA). The stated brief of the IMC was to address “all the underlying causes of the tensions between communities and the foreign nationals” (Williams, 2015), which appeared to indicate that the problem of xenophobia was finally being taken seriously at the highest levels of government. The reality proved to be very different. In April 2015, the IMC launched Operation Fiela (“sweep” in Sesotho), which was officially described as “a multidisciplinary interdepartmental operation aimed at eliminating criminality and general lawlessness from our communities […] so that the people of South Africa can be and feel safe” (Gov.za, 2015). As a response to xenophobic violence, the actions of the IMC seemed grossly inappropriate and were roundly condemned by international and local human rights organisations. Lawyers for Human Rights, for example, characterised AHMR, Vol.3 No2, May-August 2017 – Special Issue 856 Operation Fiela as “state-sponsored xenophobia” and “institutional xenophobia” that blurred stark differences between “criminals” and migrants, while deepening the divide between citizens and foreigners by bolstering negative perceptions, instead of correcting them” (Jordaan, 2015; LHR, 2015). A central component of Operation Fiela was a massive police and army drive to harass migrants and migrant-owned businesses (Nicolson, 2015a). The IMC made its conclusions about the 2015 anti-migrant violence known in Parliament. According to the Chair, Minister Jeff Radebe, the primary cause was uncontrolled migration and “increased competition arising from the socio-economic circumstances in South Africa.” Foreign nationals were placing a strain on government services and “dominating trade in certain sectors such as consumable goods in informal settlements which has had a negative impact on unemployed and low skilled South Africans.” These unsubstantiated findings – essentially blaming the violence on migrants and the poor enforcement of migration controls – were compounded by the Chair’s observation at a press conference that “as the Inter-Ministerial Committee, we’ve concluded that South Africans are not xenophobic” (Davis, 2015). The Parliamentary Ad Hoc Committee constituted to investigate the 2015 violence reached the same conclusions. In the course of deliberations to finalise their report, members of the Committee made highly pejorative statements about migrants (PMG, 2015). They indicated that the competitive advantage enjoyed by foreign-owned businesses was achieved through the use of “unfair, competitive trading practices.” They emphasised that the violence and looting directed at migrant businesses and properties was the direct result of “competition for trading spaces” and “overcrowded trading spaces” and said there was no “credible evidence” that migrant-run businesses created employment. At public consultations, the committee chair chastised the media for using the term ‘xenophobia’ to describe episodes of violence targeting migrant-operated businesses (Masinga, 2015). She reportedly characterised the violence as “criminals […] targeting shops to get food and had nothing to do with foreign nationals.” The Committee’s final report reached the conclusion that xenophobia did not exist, explaining the violence as the actions of criminals “who are often drug addicts” (PJC, 2015: 35; Nicolson, 2015b). http://ewn.co.za/SearchResultsPage?search=xenophobia AHMR, Vol.3 No2, May-August 2017 – Special Issue 857 Xenophobia denialism was also evident in the report of the KwaZulu-Natal Special Reference Group (SRG), an independent commission of enquiry appointed by the provincial government and headed by Judge Navi Pillay (formerly the UN High Commissioner for Human Rights) (SRG, 2015). The SPG argued that the immediate cause of the outbreak of violence was “deliberate efforts of select individuals, some of whom had interests in the informal trading sector, to drive away competition by foreign national-owned businesses […] These deliberate efforts sparked the outbreak of widespread incidents of criminality, violence and looting of properties owned by foreign nationals” (SRG, 2015: 9-10). Furthermore, “many of the perceptions of foreign national traders, although largely unfounded, contributed to heightened tensions” (SRG, 2015: x). The SRG studiously avoided labelling the violence as xenophobic or seeing xenophobia as a contributing or even motivating factor. At most, it conceded that “the violent attacks against foreign nationals were, in some measure, fuelled by dominant and negative perceptions that exist amongst locals and foreign nationals about one another” (SRG, 2015). However, it is hard to see how the attitudes of foreign nationals could be responsible for their own violent victimisation and none of the mob violence was perpetrated by migrants on South Africans. This entrenched and contradictory tendency – denying the existence of xenophobia while simultaneously blaming migrants in ways that border on the xenophobic – has been a defining characteristic of official responses to the escalating attacks nationwide on migrant and refugee-owned informal businesses (Crush & Ramachandran, 2015a). If mere criminality is the source of the plague of chronic violence against non-South African entrepreneurs, we might expect South Africans operating in the same areas to be equally affected. Some researchers have claimed that this is, in fact, the case (Charman & Piper, 2012; Charman et al., 2012; Piper & Charman, 2016). They argue that the attacks on informal businesses in South Africa are structural in nature, shaped by competition and other localised factors other than nationality or xenophobia. In a study of Delft in Cape Town, Charman and Piper (2012: 81) maintain that “despite a recent history of intense economic competition in the spaza market in which foreign shopkeepers have come to dominate, levels of violent crime against foreign shopkeepers […] are not significantly higher than AHMR, Vol.3 No2, May-August 2017 – Special Issue 858 against South African shopkeepers.” They conclude, as a consequence, that there is no need to invoke xenophobia to explain violence against non-South African informal enterprises, a conclusion that ironically echoes that of government officials and the commissions of enquiry. Rather, “some combination of criminality and economic competition seems to explain the violence” (Charman & Piper, 2012: 89). In a larger study, Piper and Charman (2016: 332) examine patterns of violence in three cities and conclude that “it simply is not true that […] South African shopkeepers experience less violent crime than foreign shopkeepers” and therefore that “the chance of being violently targeted is less about nationality, and more about whether you keep prices low and (presumably) profits high.” In this paper, we provide alternate evidence that contradicts the conclusion about the unimportance of nationality in explaining citizen and police violence in the informal sector. The paper draws on a study of over 2,000 informal sector businesses in Cape Town and Limpopo Province, half owned by refugees and half owned by South Africans. The methodology is described in Crush et al (2015). Security Risks and Vulnerabilities The sustainability of informal enterprises is shaped by the challenges they encounter and the manner in which they are able to effectively manage business risks. A sizeable body of research has shown that small enterprises in the South African informal economy face significant business obstacles, preventing them from maximising their potential (Abor & Quartey, 2010; Crush et al., 2015; Grant, 2013; Thompson, 2016; Willemse, 2011). These business risks include limited trading spaces; lack of access to loans from formal financial institutions; few technical, financial and business-related skills; excessive licensing or regulatory restrictions on business operations; lack of a well-defined policy framework for such operators; intense competition with other similar businesses; and lack of infrastructure such as adequate storage facilities (Callaghan & Venter, 2011; Gastrow & Amit, 2015; Ligthelm, 2011; Rogerson, 2016a, 2016b; Venter, 2012). AHMR, Vol.3 No2, May-August 2017 – Special Issue 859 In South Africa, business risks are compounded by security risks because of the extremely unpredictable and often dangerous operating environment. These security risks are of several main types. In many cities, the informal economy is regarded with suspicion and even outright hostility by municipalities, and is seen as a reservoir of crime and illegality (Rogerson, 2016a). The resulting oppressive regulatory environment is enforced by the South African Police Services (SAPS) and by municipal police who make regular raids, issue fines and confiscate goods (Rogerson, 2016b; Skinner, 2008). Harassment by police and enforcement officials is compounded by police conduct including demand for bribes and illegal confiscation of business inventory/stock. Informal businesses are regular targets of national (Operation Fiela), provincial (Operation Hardstick in Limpopo) and city-wide (Operation Clean Sweep in Johannesburg) police purges of the streets and large-scale seizure of stock. The courts have generally concluded that these operations are largely targeted at the foreign-owned businesses. A 2014 Supreme Court (2014: 25) judgment striking down Operation Hardstick in Limpopo, for example, left the Court with “the uneasy feeling that the stance adopted by the authorities in relation to the licensing of spaza shops and tuck- shops was in order to induce foreign nationals who were destitute to leave our shores.” The obverse of police misconduct is a failure to provide consistent protection when businesses are under threat or are victims of crime and other violence. Many informal businesses service the basic needs of low-income, crime-ridden communities. This means that, by definition, they are vulnerable to opportunistic and often violent crime in the form of theft, robbery and assault. There is also a clear pattern of escalating group or mob violence in many parts of the country that is increasingly directed at informal businesses (Crush & Ramachandran, 2015a). Nationwide mob violence and looting in May 2008 and early 2015 were the most high-profile examples but in the years in between and since there have been numerous more localised attacks. These assaults generally involve widespread looting, destruction and burning of property and physical assault and murder. There is considerable evidence that this form of violence is targeted almost exclusively at foreign-owned businesses and, therefore, cannot be as easily dismissed as non-xenophobic. AHMR, Vol.3 No2, May-August 2017 – Special Issue 860 The question to be addressed is whether these security risks – government purges, police misconduct, opportunistic crime and mob violence – affect South African and non-South African informal businesses with equal intensity. If we accept the argument of Charman and Piper (2012) that xenophobia is not a factor, then we would expect there to be no difference between the frequency and severity with which the two groups are affected by these risks. If, however, there is systematic evidence that these risks are felt or experienced more severely by non-South African migrant informal business owners, then xenophobia needs to be reintroduced as an explanatory factor. As part of the 2016 survey of 2,000 informal businesses in Cape Town and Limpopo, respondents were asked a series of questions about the frequency with which they had experienced various security-related risks. We rely here on self-reporting since police crime statistics are unavailable and unreliable (particularly since many incidents go unreported or unprosecuted when they are reported). In addition, we accept the argument of Charman and Piper (2012) that risk may be under-reported since many informal businesses are comparatively new and may not yet have been exposed to these risks. At the same time, it is important to note that under-reporting is less likely amongst South Africans since they have a longer history of business operations in Cape Town and Limpopo. Table 1 presents the aggregated results of the security risks question for the two groups of entrepreneurs. First, it is clear from the table that not every South African and refugee respondent has been affected by the stated risks. Indeed, the majority of both groups have not been affected to date. This is an important initial finding because it does suggest that most informal entrepreneurs are able to run their businesses without any significant threat or interference. This may be because of where they are located or the measures and precautions they take to protect themselves. Second, it is clear that South Africans are not completely immune from any of these risks. Nearly one-third had been affected by robbery of their stock and nearly 20% had had income stolen. The degree of vulnerability to other security risks was much lower but not non-existent. To this extent, therefore, Piper and Charman (2016) are correct that South Africans in the informal sector are also victims AHMR, Vol.3 No2, May-August 2017 – Special Issue 861 of crime. But there is no support for their contention that South Africans and non-South Africans are equally at risk or victimised. Table 1: Security Risks Facing Refugee and South African Entrepreneurs Refugees % Affected South Africans % Affected Prejudice against my nationality 48.0 2.2 Crime/theft of goods/stock 47.8 30.9 Crime/theft of money/income 38.1 18.5 Verbal insults against my business 34.2 7.9 Conflict with South African entrepreneurs 32.7 21.1 Conflict with refugee entrepreneurs 27.1 19.0 Physical attacks/assaults by South Africans 21.4 3.5 Harassment/demand of bribes by police 18.7 5.5 Confiscation of goods by police 14.7 6.4 Arrest/detention of entrepreneur/employees 8.5 1.4 Physical attacks/assaults by police 7.9 1.1 Prejudice against my gender 6.5 5.0 On every single count, the proportion of refugees who had been affected was higher, sometimes significantly so. For example, 47% of refugees cited prejudice against their nationality as a risk to their business (compared to only 2% of South Africans). Or again, 34% of refugees were affected by related verbal insults against their business, compared to only 8% of South Africans. 48% of refugees, compared with 31% of South Africans, had been affected by theft of their goods and stock. Similarly, 38% of refugees, compared with 19% AHMR, Vol.3 No2, May-August 2017 – Special Issue 862 of South Africans, had been affected by theft of their income. Refugees reported higher levels of conflict with South African competitors (33%) than South Africans did with refugees (19%) and with other South Africans (21%). Interestingly, refugees also reported higher levels of conflict with other refugee businesses (27%). The precise details and outcomes of such conflicts need further research, but suggest that we cannot assume that refugees are a homogenous group with identical interests. Table 2 statistically validates the descriptive comparisons that suggest that refugees are more likely than South Africans to be affected by the various security risks. The difference in the frequency distributions is statistically significant at an alpha of 0.01 according the Pearson’s Chi-Square (x2(2048)=490.678, df=3, p<.001) and the Fisher’s Exact Test (531.104, p<.001). South African entrepreneurs had lower odds of experiencing every potential risk on the list. Refugees were nearly three times as likely to be victims of theft of income and five times as likely to be subject to demands for bribes by police. The odds of a refugee entrepreneur being physically assaulted, experiencing prejudice and being arrested and detained were over five times higher than for South Africans. AHMR, Vol.3 No2, May-August 2017 – Special Issue 863 Table 2: Odds Ratio Calculations for Business Problems 95% Confidence interval Odds Ratio Lower Upper Pearso n Chi- square Df P-Value (2- sided) n Prejudice against my gender 0.782 0.539 1.136 1.671 1 0.196 2051 Conflict with refugee entrepreneurs** 0.644 0.524 0.792 17.489 1 <.001 2056 Conflict with South African entrepreneurs** 0.552 0.453 0.673 34.938 1 <.001 2054 Crime/theft of goods/stock** 0.492 0.411 0.589 60.363 1 <.001 2052 Crime/theft of money/income* * 0.373 0.305 0.456 95.509 1 <.001 2050 Harassment/de mands for bribes by police** 0.241 0.176 0.330 88.722 1 <.001 2046 Confiscation of goods by police** 0.403 0.298 0.545 36.629 1 <.001 2047 Arrest/detention of 0.179 0.105 0.303 50.513 1 <.001 2045 AHMR, Vol.3 No2, May-August 2017 – Special Issue 864 yourself/employ ees** Verbal insults against your business** 0.167 0.129 0.216 214.406 1 <.001 2050 Physical attacks/assaults by police** 0.158 0.089 0.282 50.517 1 <.001 2052 Physical attacks/assaults by other South Africans** 0.137 0.096 0.196 150.979 1 <.001 2047 Prejudice against my nationality** 0.025 0.017 0.039 577.723 1 <.001 2049 *p<.05 **p<.01 In theory, we might expect to see higher degrees of informal business security risks for both groups in large cities such as Cape Town compared to the much smaller towns of Limpopo (Table 3). In the case of refugees, and with the exception of prejudice and verbal insults and treatment by police, Limpopo was indeed safer than Cape Town. For South Africans, Cape Town was also a more dangerous place to run a business. For example, 56% of refugees and 31% of South Africans had experienced theft of goods in Cape Town. In Limpopo, the equivalent figures were 38% and 30%, a much lower spread than in Cape Town. In both locations, however, the risks are significantly higher for refugees than South Africans. Indeed, refugees in Limpopo were less secure than South Africans in both Limpopo and Cape Town. Theft of goods had affected 38% of refugees in Limpopo, compared with around 30% of South Africans in both Limpopo and Cape Town. Or again, 31% of refugees in Limpopo had experienced theft of money, compared with 12% of South Africans in Limpopo and 26% in Cape Town. Some 19% of Limpopo refugees AHMR, Vol.3 No2, May-August 2017 – Special Issue 865 had experienced physical assault or attack, compared with 2% of South Africans in Limpopo and 7% in Cape Town. Table 3: Security Risks Facing Refugee and South African Entrepreneurs by Location Cape Town Limpopo Refugees % Affected South Africans % Affected Refugees % Affected South Africans % Affected Prejudice against my nationality 47.2 3.6 47.6 1.1 Crime/theft of goods/stock 56.2 30.9 38.3 30.4 Crime/theft of money/income 43.8 25.7 31.3 11.8 Verbal insults against my business 32.1 9.0 35.3 6.9 Conflict with South African entrepreneurs 34.5 20.3 30.2 21.4 Conflict with refugee entrepreneurs 27.9 15.7 25.6 22.1 Physical attacks/assaults by South Africans 23.0 5.8 19.0 1.6 Harassment/demand of bribes by police 10.5 6.6 26.2 3.9 Confiscation of goods by police 10.1 8.0 18.8 4.9 Arrest/detention of entrepreneur/employees 7.5 2.0 9.1 1.2 Physical attacks/assaults by police 6.7 1.6 8.7 1.1 AHMR, Vol.3 No2, May-August 2017 – Special Issue 866 Prejudice against my gender 5.8 6.8 6.9 3.5 Another common belief is that security risks are higher in informal settlements than in other parts of the city, particularly as many of the reports of violence against informal businesses come from the latter areas and general crime levels are also much higher. To test this hypothesis, we focused only on Cape Town and divided refugees and South Africans into two groups according to their operation in either an informal or formal part of the city (Table 4). In the case of both refugees and South Africans, the risks are higher in informal settlements across almost all indicators. However, the difference in the degree of risk between refugees and South Africans is significantly greater in informal settlements than it is in formal areas of the city. The only indicator where formal areas are riskier for both is the chance of having goods confiscated by the police. Since the police barely venture into large swathes of informal settlements, this is not surprising. Refugees are slightly more likely to experience theft of goods in the formal versus informal areas (57% versus 54%), but the difference is small and indicates that this is a major risk for the majority of businesses irrespective of location. Unsurprisingly, refugees were far more likely than South Africans to say that their business operations had been negatively affected by xenophobia: 38% versus 5% (Table 5). There are two possible reasons for South Africans being affected: first, when collective violence occurs at a particular localised settlement, it is possible that in the general chaos and mayhem, South African- owned businesses may also be caught up in the looting and vandalism. As one South African owner noted “when these attacks start, it becomes difficult for us to move and every business becomes a target. Xenophobia does not only affect foreigners, it affects everyone” (Interview, 3 March 2016). A second explanation is that there are cascading, spill-over effects on those South African small businesses with cooperative, dependent relationships and linkages with affected migrant-operated businesses (Peberdy, 2017). AHMR, Vol.3 No2, May-August 2017 – Special Issue 867 Table 4: Security Risks Facing Refugee and South African Entrepreneurs in Cape Town Formal Areas Informal Areas Refugees % Affected South Africans % Affected Refugees % Affected South Africans % Affected Prejudice against my nationality 44.9 1.7 53.8 6.2 Crime/theft of goods/stock 57.0 26.7 53.8 36.7 Crime/theft of money/income 41.4 19.5 50.8 34.3 Verbal insults against my business 30.1 6.8 37.9 11.9 Conflict with South African entrepreneurs 29.6 20.9 48.5 19.5 Conflict with refugee entrepreneurs 25.5 18.5 34.8 11.9 Physical attacks/assaults by South Africans 21.2 3.8 28.0 8.6 Harassment/demand of bribes by police 8.9 8.6 15.2 3.8 Confiscation of goods by police 10.2 8.9 9.8 6.7 Arrest/detention of entrepreneur/employees 5.1 1.7 14.4 2.4 AHMR, Vol.3 No2, May-August 2017 – Special Issue 868 Physical attacks/assaults by police 4.8 0.7 12.1 2.9 Prejudice against my gender 5.6 4.1 6.1 10.5 Table 5: Impact of Xenophobia on Business Operations Refugees % South Africans % A great deal 18.3 1.4 To some extent 19.3 3.5 Not very much 18.4 5.7 Not at all 43.9 89.4 Strategies of Self-Protection The dangerous and unpredictable environment in which informal entrepreneurs ply their trade in South African cities presents serious security challenges. It is clear from the previous section that while both groups are affected, refugees are much more vulnerable than South Africans to a range of risks. There is no a priori reason why this might be the case, other than the fact that refugees are targeted because their presence, which is viewed by citizens and officials as unwelcome and even illegitimate. This was certainly the view of most of the refugees interviewed for this study who consistently identified the manifestations of xenophobia as the major security problem they faced: Some [customers] swear at me, my customers sometimes steal from me and when you catch them, they tell [you] harshly that you are a foreigner. And that you need to go back to your country. You are always faced with difficulties when you are a foreigner and as such you need to be patient and know how to deal with different kinds of people. There is too much disrespect here from South Africans because even someone who is way AHMR, Vol.3 No2, May-August 2017 – Special Issue 869 younger than you, they can swear and say nasty things to you if you are a foreigner. And they tell straight that South Africa is their country (Interview with Somali refugee, 12 March 2016). If you are a foreigner, you are always affected by xenophobia. There is no way that you can live here and not be affected. Xenophobia starts from your customer. Some customers are very rude and if you respond, they will talk to you in their own language and scold you and then tell you to go back [to your country]. They have bad words for foreigners. Many times, my business was robbed when I was in Johannesburg. It was because I was a foreigner because they rarely stole from locals. Sometimes criminals would come to you and ask you to give them money and they would just ask you the foreigner. Why not the local people? That is xenophobia (Interview with Ethiopian refugee, 19 March 2016). Xenophobia affect[s] us all. We know who we are. We are foreigners and that doesn’t change. Nothing changes the reality. We live under alert anytime no matter the set up in which we are operating in. We always know that the same people we are dealing with can anytime become a danger to us. It is difficult to trust any person in South Africa. The person who is with you here today, when there is a protest and foreigners are being attacked, he will be the first to attack you. There is no safety. I have not been attacked but I have seen other people being attacked and it is serious. It kills your business and it can also kill you (Interview with Congolese refugee, 25 February 2016). Xenophobia is the most critical problem. I have been directly affected and have been caught up in the troubles. People have harassed me a lot, just talking like they want to kill you or to burn you or other such things. But that was when I was in Durban. Here [Cape Town] I have not been harassed. But there are many people who have been victims. They have been harassed and their goods destroyed, especially when there are strikes. The people just target anything that they can get. They are very cruel and they do not care what the owner will do to survive (Interview with Congolese refugee, 19 February 2016). AHMR, Vol.3 No2, May-August 2017 – Special Issue 870 Xenophobia affects everyone who is a foreigner. When people loot your shop is that not xenophobia? When they chase you away from operating in an area because you are a foreigner that is xenophobia. There is xenophobia here, everywhere in this country. I have friends in other parts of the country, it is xenophobia where they live. I think South Africa is the only country with such xenophobia. I have been affected many times. When I was in Gugulethu, we were robbed. That was xenophobia because they were robbing foreign-owned shops. Here I have been affected once during a strike and they took some things from the shop. So, xenophobia is everywhere here. The community leaders do not protect us during the strikes. Some of the leaders are at the forefront of looting when strikes occur, so how can they help? The government must protect us from xenophobia and crime. The police need to do their work better because right now they are not (Interview with Somali refugee, 7 March 2016). Inadequate police protection and failure to respond when refugee businesses are under attack deepens exposure to security risks. A Congolese refugee said that the only recourse available was to “run away” as the “police here in such instances, they don’t protect us, but instead abuse us” (Interview with Congolese refugee, Cape Town, 5 March 2016). Others displayed similar distrust of the police because of perceptions of bias: The police are not very helpful. If you have a case against a South African, they will always side with the South African. So, it’s a waste of time to report a case against a South African (Interview with Congolese refugee, 24 February 2016). How accurate are these perceptions of South African hostility towards refugee businesses and business-owners? A 2010 SAMP national survey of South African citizens found that only 20% were in favour of making it easier for migrants to establish small businesses and for migrant traders to buy and sell. Only 25% felt that refugees should be allowed to work in South Africa. A similar proportion said they would take part in actions to prevent migrants from operating a business in their neighbourhood, 15% said they would combine with others to force migrants to leave and 11% said they were prepared to use violence against them. Over 55% agreed with the proposition AHMR, Vol.3 No2, May-August 2017 – Special Issue 871 that the reason why migrants were victims of violence was because they did not belong in South Africa. Only 36% said that refugees should always enjoy police protection and 25% that they should never enjoy protection (Crush et al., 2013: 32-38). A number of studies have suggested that to lower the risks of victimisation, migrants adopt various measures to protect themselves and their employees (Gastrow, 2013; Hikam, 2011; Gastrow & Amit, 2015; Smit & Rugunanan, 2014). This survey sought to establish how common some of these strategies actually are and whether they are also adopted by South Africans (Table 6). One of the most common strategies is risk-sharing by partnering with other businesses. Nearly one-third of refugees and 17% of South Africans adopt risk- sharing through partnership. Sleeping on business premises (often a modified container) is a strategy pursued by both groups but, again, by more refugees (19% versus 9% of South Africans). There have been several high-profile shootings of robbers by refugees under attack but this survey found that only 6% keep weapons for self-protection. Other strategies (pursued by less than 10% of refugees and 5% of South Africans) include paying security guards and paying protection money to the police or community leaders. Around 5% of both groups purchase insurance. Table 7 analyses if the differences between the refugees and South Africans are statistically significant. With the exception of paying for insurance, refugees were far more likely than South Africans to adopt strategies of self-protection. Refugees were five times as likely to pay for protection and twice as likely to sleep on their business premises and to partner with others to distribute risk. AHMR, Vol.3 No2, May-August 2017 – Special Issue 872 Table 6: Protection Strategies Used by Informal Sector Entrepreneurs Refugees % South Africans % I partner with other businesses to distribute risks 31.0 17.4 I sleep on my business premises 18.8 8.7 I pay for security guards 7.3 1.9 I keep weapons for self-protection 5.8 4.0 I pay the police for protection 5.5 1.0 I purchase insurance 5.1 5.5 I pay community leaders for protection 2.5 0.6 Various other strategies emerged in the course of the in-depth interviews although it is not known how common these are. For example, some refugees said that they like to hire South Africans to assist in communication with customers and also because it reduces their vulnerability to violence. In addition to paying protection money to police and community leaders, refugees in one part of Cape Town pay regular protection money to the local taxi association. The taxi association then uses this reality to extort money from South Africans in the area as well. Others make sure that they do not keep all their stock on the business premises out of fear that they will be cleaned out during looting or confiscation of goods by the police. Still others are only open for business when they know that the police are no longer patrolling. AHMR, Vol.3 No2, May-August 2017 – Special Issue 873 Table 7: Odds Ratio Calculations of Business Strategies 95% C.I. for O.R. P-Values Odds Ratio Lower Upper Chi- Square Fisher's Exact Test I purchase insurance 1.078 0.732 1.587 0.703 0.768 I keep weapons for self- protection* 0.648 0.430 0.976 0.037 0.039 I partner with other businesses to distribute risks** 0.470 0.382 0.579 <.001 <.001 I sleep on my business premises** 0.411 0.315 0.537 <.001 <.001 I pay for security guards** 0.245 0.148 0.405 <.001 <.001 I pay community leaders for protection** 0.222 0.091 0.543 <.001 <.001 I pay the police for protection** 0.180 0.094 0.347 <.001 <.001 Conclusion In this paper, we have undertaken a comparative risks assessment and vulnerabilities analysis for refugee and South African entrepreneurs operating small business ventures in the informal economies of Cape Town city and various towns of Limpopo Province. Our results show that while both groups are exposed to several risks concurrently, refugee enterprises are far more vulnerable and overexposed. The social and structural insecurity experienced by refugee entrepreneurs is unambiguous from several key findings. Despite AHMR, Vol.3 No2, May-August 2017 – Special Issue 874 operating in the same localised environment and under similar conditions, this group encounters a more challenging set of hurdles and on a more frequent basis. The general act of operating small businesses in the informal economic sector does make business owners of all kinds vulnerable, but this alone cannot explain the greater vulnerabilities of the refugee cohort. Instead, xenophobia and refugee business owners’ status as ‘outsiders’ adds another layer of risk for these operators. Limited access to police protection and mistreatment by them only exacerbates this insecurity. What is also evident from our research and other recent studies is that the majority of refugee operators have not, to date, been affected by a range of potential risks. In part, this may be because of the mitigation strategies they adopt. As refugee and migrant communities grow in South Africa, the emergence of some individuals who are successfully able to mitigate common risks and build their enterprises is to be expected. However, rather than treating these achievements with suspicion and negativity, as the official mind tends to do, greater attempts need to be made to harness these productive capacities for the growth of local informal, entrepreneurial economies. These abilities are not an abnormal development nor are they driven largely or entirely by unfair advantages or the use of illicit practices. Ultimately, comprehensive national and localised strategies are required to develop and support informal entrepreneurship and small business growth in South Africa. In this process, both citizen-operated and refugee enterprises must be crucial stakeholders, and not written-off as insignificant, unequal or illegal. Acknowledgements We wish to thank the UNHCR Geneva for funding the Refugee Economic Impacts SAMP research project on which this paper is based. References Abor, J. and Quartey, P. 2010. Issues in SME development in Ghana and South Africa. International Research Journal of Finance and Economics, 39: 218-228. Bekker, S. 2015. Violent xenophobic episodes in South Africa, 2008 and 2015. African Human Mobility Review, 1: 229-252. AHMR, Vol.3 No2, May-August 2017 – Special Issue 875 Callaghan, C. and Venter, R. 2011. An investigation of the entrepreneurial orientation, context and entrepreneurial performance of inner-city Johannesburg street traders. Southern African Business Review, 15: 28-48. Charman, A. and Piper, L. 2012. Xenophobia, criminality and violent entrepreneurship: Violence against Somali shopkeepers in Delft South, Cape Town, South Africa. South African Review of Sociology, 43: 81-105. Charman, A., Petersen, L. and Piper, L. 2012. From local survivalism to foreign entrepreneurship: The transformation of the spaza sector in Delft, Cape Town. Transformation, 78: 47-73. Crush, J. and Ramachandran, S. 2014. Xenophobic Violence in South Africa: Denialism, Minimalism, Realism. SAMP Migration Policy Series No. 66, Cape Town and Waterloo: Southern African Migration Programme. Crush, J. and Ramachandran, S. 2015a. Doing business with xenophobia. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Cape Town and Ottawa: SAMP and IDRC, pp. 25-59. Crush, J., Chikanda, A. and Skinner, C. (Eds.). 2015. Mean Streets: Migration, Xenophobia and Informality in South Africa. Cape Town and Ottawa: SAMP and IDRC. Crush, J., Ramachandran, S. and Pendleton, W. 2013. Soft Targets: Xenophobia, Public Violence and Changing Attitudes to Migrants in South Africa After May 2008. SAMP Migration Policy Series No. 64, Cape Town: Southern African Migration Programme. Davis, G. 2015. Attacks on Migrants Due to Their Socio-Economic Impact. Eyewitness News. November 10, 2015. Gastrow, V. 2013. Business robbery, the foreign trader and the small shop: How business robberies affect Somali traders in the Western Cape. SA Crime Quarterly, 43: 5-15. Gastrow, V. and Amit, R. 2015. The role of migrant traders in local economies: A case study of Somali spaza shops in Cape Town. In: Crush, J., Chikanda, A. and AHMR, Vol.3 No2, May-August 2017 – Special Issue 876 Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Cape Town and Ottawa: SAMP and IDRC. Gov.za. 2015. Operation Fiela. From . Grant, R. 2013. Gendered spaces of informal entrepreneurship in Soweto, South Africa. Urban Geography, 34: 86-108. Hikam, A. 2011. An Exploratory Study on the Somali Immigrants’ Involvement in the Informal Economy of Nelson Mandela Bay. MA Thesis. Port Elizabeth: Nelson Mandela Metropolitan University. Jordaan, N. 2015. Operation Fiela ‘Demoralizes and Dehumanizes’ Migrants. Sunday Times, July 22, 2015. LHR (Lawyers for Human Rights). 2015. Civil Society Organizations Address Media on the Ongoing Raids Targeting Foreign Nationals. From http://bit.ly/2eKJODO. Ligthelm, A. 2011. Survival analysis of small informal businesses in South Africa, 2007–2010. Eurasian Business Review, 1: 160-179. Masinga, L. 2015. Stop Using the Word Xenophobia. Independent Online, July 10, 2015. Misago, J-P. 2016. Responding to xenophobic violence in post-apartheid South Africa: Barking up the wrong tree. African Human Mobility Review, 2: 443-467. Nicolson, G. 2015a. Operation Fiela: Thousands of Arrests, Doubtful Impact. Daily Maverick, September 8, 2015. Nicolson, G. 2015b. Parliamentary Report on Xenophobic Violence Talks a Lot, Says Very Little. Daily Maverick, November 24, 2015. Peberdy, S. 2017. Competition or Cooperation? South African and Migrant Entrepreneurs in Johannesburg. SAMP Migration Policy Series No. 75, Waterloo and Cape Town: Southern African Migration Programme. PMG (Parliamentary Monitoring Group). 2015. Meeting to Adopt Ad Hoc Joint Committee on Probing Violence Against Foreign Nationals Report. Cape Town: South African Parliament. http://www.gov.za/operation-fiela AHMR, Vol.3 No2, May-August 2017 – Special Issue 877 Piper, L. and Charman, A. 2016. Xenophobia, price competition and violence in the spaza sector in South Africa. African Human Mobility Review, 2: 332-361. PJC (Parliamentary Joint Committee). 2015. Report of the Ad Hoc Joint Committee on Probing Violence Against Foreign Nationals. Cape Town: South African Parliament. Rogerson, C. 2016a. Progressive rhetoric, ambiguous policy pathways: Street trading in inner-city Johannesburg, South Africa. Local Economy, 31: 204-218. Rogerson, C. 2016b. South Africa’s informal economy: Reframing debates in national policy. Local Economy, 31: 172-86. Skinner, C. 2008. The struggle for the streets: Processes of exclusion and inclusion of street traders in Durban, South Africa. Development Southern Africa, 25: 227-242. Smit, R. and Rugunanan, P. 2014. From precarious lives to precarious work: The dilemma facing refugees in Gauteng, South Africa. South African Review of Sociology, 45: 4-26. SRG (Special Reference Group). 2015. Report of the Special Reference Group on Migration and Community Integration in KwaZulu-Natal. Pietermaritzburg: Provincial Government of KwaZulu-Natal. From . Supreme Court. 2014. Somali Association of South Africa v Limpopo Department of Economic Development, Environment and Tourism (48/2014). ZASCA 143 (September 26, 2014). Thompson, D. 2016. Risky business and geography of refugee capitalism in the Somali migrant economy of Gauteng, South Africa. Journal of Ethnic and Migration Studies, 42: 120-35. Willemse, L. 2011. Opportunities and constraints facing informal traders: Evidence from four South African cities. Town and Regional Planning, 59: 8-16. Williams, P. 2015. ‘Minister Jeff Radebe: Inter-Ministerial Committee on Migration. From http://bit.ly/2ux1VBb. http://journals.sagepub.com/doi/abs/10.1177/0269094215621724 http://journals.sagepub.com/doi/abs/10.1177/0269094215621724 AHMR, Vol.3 No2, May-August 2017 – Special Issue 878 Venter, R. 2012. Entrepreneurial values, hybridity and entrepreneurial capital: Insights from Johannesburg's informal sector. Development Southern Africa, 29: 225-239. AHMR, Vol.3 No2, May-August 2017 – Special Issue 879 The Owners of Xenophobia: Zimbabwean Informal Enterprise and Xenophobic Violence in South Africa Jonathan Crush*, Godfrey Tawodzera**, Abel Chikanda*** and Daniel Tevera**** Abstract This paper is a contribution to our understanding of the intertwined economic and political crises in Zimbabwe and the crisis of xenophobia in South Africa. There have been few studies to date specifically examining the impact of xenophobic violence on Zimbabweans trying to make a living in the South African informal economy. The paper first provides a picture of Zimbabwean migrant entrepreneurship using survey data from a 2015 study of migrants in the informal economy. All of the Zimbabwean entrepreneurs interviewed in depth for the study in 2016 had either witnessed or been the victims of xenophobic violence or both. The interviews focused on the experience and impact of xenophobic violence on personal safety and business operations. The migrant accounts clearly demonstrate that they see xenophobia as a key driver of the hostility, looting and violence that they experience. The paper argues that the deep-rooted crisis in Zimbabwe, which has driven many to South Africa in the first place, makes return home in the face of xenophobia a non-viable option. Zimbabweans are forced to adopt a number of self-protection strategies, none of which ultimately provide insurance against future attack. Keywords Zimbabwean Informal Enterprise and Xenophobic Violence. * International Migration Research Centre, Balsillie School of International Affairs, 67 Erb St West, Waterloo, Ontario, Canada N2L6C2. Email address: jcrush@balsillieschool.ca ** Department of Geography and Environmental Studies, University of Limpopo, P. Bag X1106, Sovenga. Email address: godfreyltawodzera@yahoo.com *** Department of Geography & Atmospheric Science, and African & African-American Studies, University of Kansas, Lawrence, Kansas State, USA 66045. Email address: achikanda@ku.edu **** Department of Geography & Environmental Studies, University of Western Cape, Robert Sobukwe Road, Bellville 7535. Email: dtevera@uwc.ac.za mailto:jcrush@balsillieschool.ca mailto:godfreyltawodzera@yahoo.com mailto:achikanda@ku.edu mailto:dtevera@uwc.ac.za AHMR, Vol.3 No2, May-August 2017 – Special Issue 880 “South Africans are very xenophobic – they are the owners of xenophobia” (Zimbabwean migrant, Cape Town, 2016). Introduction The above arresting image of South Africans as the “owners of xenophobia” is confirmed by numerous surveys suggesting that levels of xenophobia in South Africa are unprecedented globally (Crush et al., 2013). South Africa’s crisis of xenophobia is defined by the discrimination and intolerance to which migrants are exposed on a daily basis. According to Misago et al. (2015: 17), xenophobia manifests in “a broad spectrum of behaviours including discriminatory, stereotyping and dehumanizing remarks; discriminatory policies and practices by government and private officials such as exclusion from public services to which target groups are entitled; selective enforcement of by-laws by local authorities; assault and harassment by state agents particularly the police and immigration officials; as well as public threats and violence […]that often results in massive loss of lives and livelihoods.” The nub of the crisis of xenophobia in South Africa is when feelings of hostility and intolerance manifest as extreme xenophobia, which Crush and Ramachandran (2015a) define as “a heightened form of xenophobia in which hostility and opposition to those perceived as outsiders and foreigners is strongly embedded and expressed through aggressive acts directed at migrants and refugees [and] recurrent episodes of violence.” Xenophobic violence represents ‘tipping points’ or intense moments in the general ongoing crisis of xenophobia. Southern African Migration Project’s (SAMP) national surveys have consistently found that a significant minority of South African citizens are willing to resort to violence to rid their communities of migrants (Crush, 2008; Crush et al., 2013). The deadliest examples of extreme xenophobia in South Africa to date were high-profile and widespread violence against migrants and refugees in May 2008 and March 2015. The nature and impacts of the 2008 crisis are now well-documented, although there remain differences of opinion about its causes (Bekker, 2015; Cabane, 2015; Desai, 2015; Everatt, 2011; Hassim et al., 2008; Hayem, 2013; Landau, 2012; Steinberg, 2012). AHMR, Vol.3 No2, May-August 2017 – Special Issue 881 For all its strengths, the literature on May 2008 tends to treat the victims of xenophobic violence in an undifferentiated fashion, leading to the assumption that all migrants – irrespective of national origin, legal status, length of time in the country and livelihood activity – were equally at risk. Yet, attitudinal surveys clearly show that South Africans differentiate between migrants of different national origin and that Zimbabweans are amongst the most disliked (Crush et al., 2013). Discussions of May 2008 also do not differentiate sufficiently between the types of targets that were attacked. For example, many African migrants and refugees operated small businesses in the informal economy of affected urban areas and these enterprises came under sustained attack during the pogrom. In 2015, one of the explicit targets of the xenophobic attacks was informal businesses run by migrants and refugees. Violent attacks on migrant and refugee entrepreneurs and their businesses have not been confined to acute episodes of extreme xenophobia such as those in May 2008 and March 2015 (Charman & Piper, 2012; Crush et al., 2015; Crush & Ramachandran, 2015a, 2015b; Tevera, 2013). Ongoing acts of extreme xenophobia have increasingly manifested in the form of collective violence targeted at migrant and refugee-owned businesses. The frequency and ferocity of such attacks have increased over time and cannot simply be written off, as the state seeks to do, as ‘mere criminality.’ Chronic extreme xenophobia has prompted various responses and remedial actions by migrants and refugees including paying protection money, beefing up business security, arming in self-defence, avoiding neighbourhoods known to be particularly dangerous and moving away from the major cities to smaller urban centres (Crush et al., 2015a). Zimbabweans are not the only small business owners who have become victims of extreme xenophobia in South Africa; attacks on migrants and refugees from other countries are also well documented (Gastrow, 2013; Gastrow & Amit, 2015; Piper & Charman, 2016). However, there have been few studies to date specifically examining the impact of xenophobic violence on Zimbabweans trying to make a living in the South African informal economy (Duri, 2016; Hungwe, 2014; Sibanda & Sibanda, 2014). AHMR, Vol.3 No2, May-August 2017 – Special Issue 882 This paper is based on two sources of data. First, SAMP’s Growing Informal Cities Project surveyed over 1,000 randomly selected migrant-owned informal sector enterprises in Cape Town and Johannesburg in 2015 (Peberdy, 2016; Tawodzera et al., 2016). The survey sample included 304 Zimbabwean-owned enterprises. For the purposes of this paper, we extracted this data from the larger database. Second, 50 in-depth interviews were conducted in 2016 with Zimbabwean informal business-owners in Cape Town, Johannesburg and Polokwane as part of SAMP participation in the Migrants in Countries in Crisis (MICIC) project. This paper is a contribution to our understanding of the intertwined economic and political crises in Zimbabwe and the crisis of xenophobia in South Africa. It also aims to contribute to the more general literature on migrants in countries in crisis in a situation of intersectionality where migrants are forced to navigate a state of crisis in both the country of origin and the country of destination. The term “migration in countries in crisis” is usually taken to refer to the plight of migrants caught up in an unexpected crisis situation in a “host country” (Koser, 2014; Martin et al., 2014; Weerasinghe & Taylor, 2015). Although emergencies affecting migrants are becoming increasingly complex and multifaceted, the intersection of crisis situations in countries of origin and destination have been given insufficient attention (IOM, 2012; McAdam, 2014). A dual or multiple crisis situation spanning origin and destination presents new, and not easily resolved, challenges for the management of crisis and the safety of migrants (Hendow et al., 2016; Perching, 2016). Crisis-Driven Migration Betts and Kaytaz (2009: 2) label the exodus from Zimbabwe an example of “survival migration,” which they define as refugees and “people who are forced to cross an international border to flee state failure, severe environmental distress, or widespread livelihood collapse.” Under conditions of survival migration, the traditional distinction between refugees and economic migrants breaks down (Betts, 2013). The argument that all Zimbabwean migrants should be defined as “survival migrants” requires closer scrutiny. For example, it is based in part on the view that conditions in Zimbabwe are so dire that out-migration for survival is the only option. However, this does not AHMR, Vol.3 No2, May-August 2017 – Special Issue 883 explain why the majority of Zimbabweans have not left nor the role of migration in reducing pressures for further out-migration through remittances (Crush & Tevera, 2010). The argument that all Zimbabweans are “survival migrants” also runs the risk of homogenising migrant flows and downplaying the heterogeneity of migration movement out of the country. The idea that all migrants from Zimbabwe are “survival migrants” also seems to rest on the admittedly desperate situation of migrants in squalid transit shelters in the border town of Musina and at overcrowded safe havens such as churches (Betts & Kaytaz, 2009; Kuljian, 2013). The idea of “survival migration” fits this sub-set of Zimbabwean migrants but certainly does not encompass them all. Far from being the desperate and destitute people conveyed by images of “survival migration,” many Zimbabwean migrants to South Africa exhibit considerable ingenuity, industry and energy. A recent survey of Zimbabwean migrants in Cape Town and Johannesburg found that over 60% of migrants who had come to South Africa in the previous decade were formally employed and only 18% were unemployed (Crush et al., 2015b). At the same time, increasing numbers were doing more menial jobs including 25% in manual work, 13% in the service industry and 8% in domestic work. A longitudinal study of day labourers in Tshwane demonstrates the increase in Zimbabweans seeking casual work, which rose from 7% in 2004 to 33% in 2007 to 45% of workseekers in 2015 (Blaauw et al., 2016). The extent of participation by Zimbabwean migrants in the South African informal sector is unknown. SAMP’s 2005 national survey of migrant-sending households in Zimbabwe found that 21% of working migrants outside the country were in the informal economy (Crush & Tevera, 2010: 12). A 2007 survey of migrants in Johannesburg found that 19% were working as hawkers or artisans (Makina, 2010). SAMP’s 2010 survey of recent Zimbabwean migrants in Johannesburg and Cape Town found that 27% were working or deriving income from the informal economy (Crush et al., 2015a). Crush and Tawodzera’s (2017) survey of poorer Zimbabwean households in South Africa found that 36% of household members in employment were working in the AHMR, Vol.3 No2, May-August 2017 – Special Issue 884 informal economy. While indicative, these studies suggest that somewhere between 20-30% of Zimbabwean migrants in major South African cities could be involved in the informal economy. They also suggest that the importance of informal sector employment to Zimbabweans has increased over time. The 2015 SAMP survey of migrant enterprises found a distinct gender bias in both cities with 60% of Zimbabwe entrepreneurs in Cape Town and 65% in Johannesburg being male. This was a marked contrast to the business of informal cross-border trading between Zimbabwe and South Africa which is dominated by female Zimbabweans (Chikanda & Tawodzera, 2017). The number of migrant entrepreneurs who arrived in South Africa peaked in the years between 2005 and 2010, at the height of the economic crisis in Zimbabwe, and appears to have fallen since (Figure 1). Less than 2% had migrated to South Africa before 1994. Nearly 18% of the Johannesburg migrant entrepreneurs had moved there before 2000 compared to only 2% of those in Cape Town. Over time, Cape Town has become an increasingly attractive destination. As many as 88% of the migrants in Cape Town arrived in the city after 2005 (compared to 52% of those who moved to Johannesburg). Figure 1: Comparison of Year of Migration and Year of Establishing Informal Business. AHMR, Vol.3 No2, May-August 2017 – Special Issue 885 Only 5% of the survey respondents had experience working in the Zimbabwean informal economy prior to migrating to South Africa. Those with prior experience had generally been involved in informal cross-border trading and were therefore familiar with South Africa. One male migrant described the transition as follows: I used to come here as a trader from the early 2000s. I stopped in 2005 and came here to South Africa to live. My business is about making and selling electric jugs and brooms. I used to come here and sell them and go back home. There are some reasons why I came to stay. One is that the economic situation was getting bad. [Zimbabwe] was no longer the same. I was selling things and not making much money. I wanted to build a house in Zimbabwe and I was failing to do so. The cost of living was high. I had just married and things were tough. Then there was the issue of politics. My wife was harassed when I was in Johannesburg buying goods. They came and searched our house and they found nothing. They wanted evidence that I was a sell-out, but they did not find anything. My wife was pregnant so I saw that they could injure her if they came back next time. That is when I moved to South Africa (Johannesburg Interview No. 4). In this case, economic hardship and political harassment were additional factors in the decision to move to South Africa. The overall reasons for migration to South Africa were clearly related to the ongoing economic crisis in Zimbabwe. Over 80% agreed with the proposition that they had come to South Africa in order to provide for family back home. As many as 73% said they had come to South Africa to look for employment (Figure 2). There was a marked difference here between the Johannesburg and Cape Town respondents (with 82% and 59% in agreement, respectively) which may reflect differences in the perception (and reality) of labour market access in the two cities. AHMR, Vol.3 No2, May-August 2017 – Special Issue 886 Figure 2: Reasons for Migrating to South Africa The reverse was true with regard to starting a business as a reason for migration, with 56% in Cape Town and 35% in Johannesburg in agreement. In general, this suggests that Johannesburg is seen as a place where it is easier to obtain formal sector employment and Cape Town is a more amenable location for starting an informal business. Unemployment was a significant driver of migration, with 39% of the sample reporting that they were unemployed prior to leaving Zimbabwe. Again, there was a marked difference between the entrepreneurs in the two cities: only 20% of the Cape Town respondents were unemployed prior to leaving compared with 51% of the Johannesburg respondents. The high proportion who said that they had migrated as refugees or asylum-seekers is a reflection of the fact that over 300,000 Zimbabweans applied for asylum-seeker permits between 2004 and 2010 in order to legalise their stay in South Africa (Amit & Kriger, 2014). At the same time, a proportion of this number left because of political persecution. Exactly how many is difficult to say given that South Africa has approved less than 3,000 of all Zimbabwean refugee claimants. The survey found that relatively few of the Zimbabwean entrepreneurs did not have papers permitting them to be in and/or legally work in the country (Table 1). Just over one-third of the migrants had asylum-seeker (Section 22) permits AHMR, Vol.3 No2, May-August 2017 – Special Issue 887 but only 5% had recognised refugee status (Section 24 permits). Both asylum- seekers and refugees have a legal right to work and earn. Around one-quarter had work permits, which the majority would have acquired under the Zimbabwe Dispensation Programme (ZDP) in 2010 and 2014 (Thebe, 2017). Around 10% of the migrants had visitor’s permits which are usually issued for 90 days at a time. Only 15% were undocumented and did not have permits to reside and/or work in South Africa. Table 1: Legal Status of Zimbabwean Entrepreneurs Total (%) Cape Town (%) Johannesburg (%) Asylum-seeker permit 35.9 43.2 31.2 Work permit 24.7 17.8 29.0 Visitor’s permit 9.5 8.5 10.2 Refugee permit 5.3 8.5 3.2 Permanent resident 4.6 3.4 5.4 Undocumented 14.5 16.1 13.4 Other 2.3 0.8 3.3 No answer 3.1 1.6 4.3 The majority of the surveyed Zimbabwean migrant enterprises were in the retail, trade and wholesale sector, followed by services and manufacturing, with slight differences between the two cities (Figure 3). As Figure 1 shows, most migrants did not immediately start an informal business on arrival but first needed to raise start-up capital. Migrants and refugees face severe obstacles in accessing loans from formal sources in South Africa as they require collateral (Tawodzera et al., 2015). Just over three-quarters of the migrant entrepreneurs in this survey relied on their personal savings to start their businesses (Table 2). There was slightly greater reliance on personal AHMR, Vol.3 No2, May-August 2017 – Special Issue 888 savings by entrepreneurs in Johannesburg (87%) than Cape Town (64%). More Zimbabwean migrants in Cape Town were able to access funds from relatives and non-relatives. Figure 3: Sectoral Breakdown of Zimbabwean Migrant Businesses Table 2: Primary Sources of Start-Up Capital Total (%) Cape Town (%) Johannesburg (%) Personal savings 78.0 64.4 86.6 Loan from relatives 9.2 13.6 6.5 Loan from non-relatives 6.6 15.3 1.1 Money lenders 2.0 1.7 2.2 Loan from informal financial institutions 2.0 1.7 2.2 Business credit (goods on terms) 0.7 0.0 1.1 AHMR, Vol.3 No2, May-August 2017 – Special Issue 889 Loan from micro-finance institution 0.3 0.8 0.0 Other source of capital 1.3 2.5 0.5 Experiencing Xenophobia All of the Zimbabwean entrepreneurs interviewed in depth had either witnessed or been the victims of xenophobic violence or both. The interviews focused on the experiences and impact of xenophobic violence on personal safety and business operations. Many of those interviewed had come to South Africa after the nationwide attacks on migrants in May 2008 but none were unaware of the violence or did not know some of the victims. Those who had been in the country at the time lost almost everything they had, but they did not return permanently to Zimbabwe, a primary objective of their attackers. Instead, they took refuge in shelters and churches and re-started their businesses once the worst of the violence was over. Three accounts in widely separated parts of the country (Alexandra Park in Johannesburg, Imizamo Yethu in Cape Town and Mankweng in Polokwane) show both the destructive nature of the 2008 xenophobic violence and the responses of the migrants: During the xenophobic attacks of 2008 I was there. My musika was destroyed. People came marching and asking foreigners to leave […] They came and destroyed the musika. It was made up of cardboard and corrugated iron sheets. They destroyed it. The cardboard was burnt and the corrugated sheets were taken and some of them thrown all over. I lost a lot of money there. Maybe ZAR 3,000. I had a lot of goods and I was also selling beans, groundnuts and even matemba and fish. I lost everything. I was only able to carry a few things and fled. Otherwise they would have killed me as well. What could they do? The people that start the violence are the ones that can even kill you. Many people died in Alex Park. They died. I actually saw a person who had been stoned to death and he was lying there for a day without the police getting him (Johannesburg Interview No. 13). AHMR, Vol.3 No2, May-August 2017 – Special Issue 890 I had just closed my spaza and had not even reached home when I saw people singing and getting up here. They were coming from the direction of the police station coming uphill. We had already heard of xenophobia and so I knew it was happening here. I wanted to go back and get some things from the spaza but I was too late because they were moving fast. I just had time to change direction and ended up in Hout Bay. There were other Zimbabweans who had also run away and were there. I joined them and we went to Wynberg and stayed there at the police station. There were many of us. Like me, most people had nothing because they never had time to go home and get clothes or blankets. I called someone in Rosebank and they told us that they were staying at a Methodist church there so that is where we went. We spent three weeks there. My spaza shop was looted. I never got anything back, not even a single sweet. They took everything so I had to start from scratch (Cape Town Interview No. 20). I was living in Mankweng with two other ladies from Zimbabwe. We were renting a room in Zone 2. We had been living there for some time and we knew most of the people there. But when xenophobia erupted it was as if we had never lived there. We saw some of the people that we knew actually looting things belonging to foreigners. We were confronted by a group of young men – and they demanded money otherwise they would kill us. It was like a dream. We could not believe it. We were robbed there, close to the road, where everyone could see. They just took what they wanted and went away singing. I lost my bag, my wallet and my friend also lost everything. I was scared that we could be killed or raped. Even now I cannot believe that I survived. We went to the highway, the N1 and hiked to Musina and then home to Zimbabwe. I only came back after a month when things had calmed down. I stopped doing business for over a month. I had no money to start over. I had to borrow some money and it took time to recover. Some of my customers moved with my money and I never recovered the money. I had to start from scratch and it was difficult (Polokwane Interview No. 3). Most of the respondents recounted incidents of violence that had personally affected them since 2008. These accounts revealed a number of features AHMR, Vol.3 No2, May-August 2017 – Special Issue 891 common to all xenophobic attacks on migrant businesses. First, much of the violence seems to the migrants to be spontaneous and occurs without warning. In practice, this is rarely the case as many attacks are preceded by community meetings from which migrants are excluded (Landau, 2012). Therefore, they have little notice or ability to take evasive action. As one victim of violence in the informal settlement of Diepsloot near Johannesburg in 2013 observed: They just occur haphazardly. We cannot always tell what happens next so it is difficult to do anything and to think of a way to respond. It just happens when you are least aware of the problems that are about to erupt. Sometimes we are caught up with all our wares and they are destroyed and stolen and so it is difficult to do anything (Johannesburg Interview No. 10). The journalist, Anton Harber’s (2011) portrait of Diepsloot paints a picture of a volatile settlement in which vigilante justice and attacks on foreign-owned businesses are common. In 2008, for example, “they showed no discrimination in targeting men, women and children, and destroyed, looted and burnt down their businesses and houses” (Harber, 2011: 123). Second, the perpetrators of xenophobic violence are often from the same community and are even personally known to the victims: The people that robbed us are in this community and we know them. They are the community members here. Some of the people here do not like us foreigners. They pretend when you deal with them to like us. But they do not like us (Johannesburg Interview No. 2). The fact that migrant entrepreneurs are able to provide goods, including food, at competitive prices and offer credit to consumers is clearly insufficient to protect them when violence erupts. In one part of Khayelitsha, there is reportedly little violence as long as migrant business owners pay protection money to the powerful local taxi association. In many other areas, the respondents reported that community leaders are either ineffective in dealing with the violence or, in some cases, actively foment hostility and instigate attacks. AHMR, Vol.3 No2, May-August 2017 – Special Issue 892 Third, the looting of stock on the premises is a constant feature in the narratives of the migrant business-owners. As one observed: There are hard core thieves who rob people and also jobless people around who are now taking advantage of these xenophobic attacks and robbing people to get money because they have nothing to do with their lives (Johannesburg Interview No. 1). However, the respondents consistently maintained that robbery per se was not the prime motive for the attacks. As one respondent noted: They target shops, the owners as well as the goods inside. They only target foreign owned shops. There is more to that (than robbery), they want us to leave their country because they hate our businesses here and they say we are finishing their jobs (Johannesburg Interview No. 3). Others pointed out that South African business owners in the same vicinity are left alone during crowd violence, that attacks often involve vicious physical assaults against the person, and that they are usually accompanied by vituperative xenophobic language: People were being beaten up and they were dying. A group of South Africans moved around this whole squatter camp terrorising all foreigners and they used to move with someone who knew where all foreigners stayed. These people moved with knobkerries, metal sticks, sjamboks and any sort of weapon you can think of for distraction. If you were a foreigner and did not have a passport they would beat you up (Johannesburg Interview No. 3). The violence was there for two days or so and I thought it was over. I went to service a car in Heideveld. When I was coming back I passed through my friend’s place and he accompanied me half way. When he had gone, and I was in Sisulu Street down there, they attacked me with a plank and something like a rubber. They hit me all over and even stomped on me. It was xenophobia. They told me that they would kill me and that I was a foreigner and not wanted here. I cried and asked for them to leave me and they continued. No-one intervened. It was past 8pm and there were still people moving about. A few other guys joined in. I AHMR, Vol.3 No2, May-August 2017 – Special Issue 893 was saved by a car that passed, when its lights flashed at them they ran away. They told me that next time they would kill me (Cape Town Interview No. 3). I was robbed in broad daylight here in Masiphumelele. It was not a real robbery, it was a gang just saying foreigners must leave. I was about to park my car when the group of men descended on me. They asked for my ID and when I said let me go and get it, they pounced on me and started pushing me. My neighbours just looked on. I asked what I had done but they were just singing derogatory songs. It was pure xenophobia. Most of the locals here joined and wanted to chase me out. Even my neighbours were caught up in that chaos and were told to go. You see, when there is a small thing that happens, it ends up being that foreigners must be chased away. Is that not xenophobia? Many times here I have been insulted only because I am a foreigner. You ferry someone’s goods and they pay you little and the next time you want your balance they start some story that you are a thief or so on and the others join in. Is that not xenophobia? Why do they not do that to South Africans? Why only to foreigners? These people have xenophobia in their blood (Cape Town Interview No. 13). Fourth, many of the accounts describe how an anti-government service delivery protest or march can quickly disintegrate into mob violence and looting of shops and stores owned by migrants. The connection between the two events is not immediately obvious but, according to the respondents, the looting is never indiscriminate but only targets migrants. The reason, according to some, is that they become scapegoats for the government’s failure to deliver services South Africans are not friendly. They say this is their country and they do what they want to us, hurting us. These locals ask services from their government and if they are not given them they demonstrate and if their concerns are not heard they put their frustrations on foreigners. Most of them are uneducated so they think we are the cause of their problem and when they see you in business they think you are taking over their business. They target foreigners in business. They start with businesses AHMR, Vol.3 No2, May-August 2017 – Special Issue 894 and sometimes when their concerns are not heard they even start attacking those not in business and foreigners in their homes (Cape Town Interview No. 11). Fifth, there was some evidence of “violent entrepreneurship” involving attacks orchestrated by South African competitors. One Zimbabwean entrepreneur in Polokwane, for example, described how he had established a business selling and repairing cell phones. He said that his South African competitors reported him to the police for dealing in stolen phones but his records showed that all his transactions were legitimate. According to him they had tried several times to get him arrested. The reason? “They even tell me to my face that they want me out of this place because I am a foreigner. How can they fail to make business when I as a foreigner is doing well? That is their quarrel. Some have even organised thieves to rob me and I have been robbed twice.” Sixth, xenophobic violence is gender-indiscriminate in that both male and female migrants recounted equally harrowing stories. Lefko-Everett (2010) has argued that one of the most common strategies adopted by Zimbabwean women migrants travelling to South Africa as cross-border traders is to travel and stay in groups as a means of protection. Zimbabwean women living in South Africa and selling on the streets are generally unable to benefit from group protection. One woman in Johannesburg described her experience and helpless situation as follows: They were calling me names and some were telling me to go back to Zimbabwe saying I would die that night. Some of the foreigners who were there and had been trying to support me saw that the situation was getting serious and just disappeared. I lost most of my goods that day as people just started taking them. The lady who was selling close to me also lost her products as people just took and went. It was terrible. No-one was on our side. They just did not care that we were females. They just harassed us. I even thought of going back home that day. What stopped me is the thought of going back to look at my kids without anything. And there was nothing that I would do in Zimbabwe. Here we live with xenophobia every day. We see it happening and there is nothing that we can do (Johannesburg Interview No. 10). AHMR, Vol.3 No2, May-August 2017 – Special Issue 895 Finally, the respondents differed on whether Zimbabweans were particular targets. Most said that all foreign-owned businesses were targeted, not simply Zimbabweans. A number commented that the type of business made a difference, with food and grocery shops being especially vulnerable. However, virtually all agreed on one point: the purpose of the attacks was not simply to steal certain desirable goods but to clean them out and destroy their business premises and operations so that they could not continue to operate. There were numerous examples of entrepreneurs who had lost all of their stock and also had their premises vandalised and wrecked, even when they were operating from containers, which are generally considered to be the best form of protection: They broke and took away everything as if they don’t want one to be in business. If they wanted goods only, they would have just broken in and taken stuff only but they destroyed, breaking windows and even removing them and most people are not yet back on their feet (Cape Town Interview No. 8). In 2011, the business was attacked by local people. The shop was attacked by the mob. They looted everything and left me with almost nothing. I had goods worth over R 15,000 in here. Everything except some few bottles of cooking oil and cigarettes remained. It almost destroyed my business. I was left with very little. I had not saved much so it took me some time to be on my feet again. I had to borrow some money from friends because I needed to restock. I cannot afford to stock much as I am not sure what happens tomorrow. These days we no longer put everything here. Some of the stock is at home so that if the steal here, I will have some of my stock at home to start again. I just replenish what is in short supply here (Cape Town Interview No. 10). We had just brought stuff from Zimbabweans on a Sunday. They were worth about R 10,000 and included nyimo, mbambaira, nzungu, matemba and we had also just stocked the local products. We had bought a lot of crates like onions for about ZAR 15,000. All these products were in the container and the container was destroyed. They upended it and spilled all the products that were inside to the ground. Some of the AHMR, Vol.3 No2, May-August 2017 – Special Issue 896 products were burnt, taken and we were left with nothing. And because we had just stocked we didn’t have any money at home so we had to start all over from scratch (Cape Town Interview No. 1). Responses to Xenophobic Violence The pervasive view amongst South African politicians that xenophobia does not even exist in the country seems particularly odious given the experiences of Zimbabwean and other migrants. The term “xenophobia” itself was used by all the respondents to describe the harassment and physical abuse they experience and some even referred to the widespread violence in 2008 and 2015 as “the xenophobia.” However, they were also asked if they thought South Africans were xenophobic and, if so, why. No-one answered the question in the negative. A selection of responses clearly indicate that for Zimbabweans, South Africans are, indeed, the “owners of xenophobia”: I can say that three-quarters of them show their hatred towards us foreign nationals. They don’t like us. Xenophobia is a South African thing. It happens more than anywhere in the world I think. Everything they do shows it. They do not like us. They speak to us like we are not like them. They look down upon us. They are like that whether they are Christians or not. The children learn it from their parents. They call us makwerekwere. Do you know even small kids can call you makwerekwere? Is that not xenophobic? (Johannesburg Interview No. 2). If you want to see how they hate us, just have a disagreement and they will tell you bad things, telling you that you will die. What kind of a person wants to see another person dying? Life is sacred, but here in South Africa no one seems to care about that. They would rather you die so that they can get what you have. This is the only society where people kill each other over very simple disagreements (Johannesburg Interview No. 10). You can see it almost every day in the train and other places when you pass they call you derogatory names like makwerekwere. We can see it every day in our daily life and we live with it. It does not only happen to people doing business, but it happens to any foreigner no matter whom. If you AHMR, Vol.3 No2, May-August 2017 – Special Issue 897 can’t speak their language, you already are a kwerekwere and you are in trouble (Cape Town Interview No. 4). The way they see us, they see us as if we are lesser than them. They say bad things about us, like we are thieves and we are ugly and we do not bath, such things. But they know most of these things are not true but they like saying them anyway (Cape Town Interview No. 10). Is there a country in the world where foreigners are killed and burnt like here? No. South Africa is a place like no other place. It is a country with people that do not care about other people. Look at the way they kill foreigners. The way they chase foreigners and steal their goods and injure them. That is not done by normal people. South Africans are xenophobic. They do not fear evil spirits from the dead. They just kill and the next hour they are busy braaiing and singing and eating amagwinya. They are not normal people (Cape Town Interview No. 12). The language and practices of xenophobia cow the victims into silence and a sense of helplessness, short of returning to Zimbabwe, which is not seen as a viable option. As one respondent said: “Here we live with xenophobia every day. We see it happening and there is nothing that we can do” (Johannesburg Interview No. 10). And another: “I remain silent because I am Zimbabwean and I can’t go against what they say. But they have to realise that we are the same we have the same skin as black people but we just keep quiet even as they insult us” (Johannesburg Interview No. 12). The interviews provide important insights into how migrant entrepreneurs themselves respond to the threat and reality of xenophobic violence. From the responses of some of the migrants, it appears that trying to “fit in” and integrate by learning local languages, dress codes and cultural practices is one way to try and pre-empt attacks (Hungwe, 2012, 2013). However, these strategies are no guarantee of protection when mob violence breaks out: I was robbed during the day. There was a strike and I was coming from the shops. I was not here the previous day and so I did not know that there was a strike. When they saw me coming the mob ran to me. I was AHMR, Vol.3 No2, May-August 2017 – Special Issue 898 beaten and robbed. They knew I was a foreigner. I can speak three local languages and I spoke in isiZulu but they knew me, some of them and they said he is a Zimbabwean and they attacked me. If I was a local I was not going to be attacked. I had ZAR 1,800. All was taken. That was my money that I had collected from my customers. They robbed me because I was a Zimbabwean, a foreigner (Johannesburg Interview No. 2). A number of the respondents observed that unlike some migrant groups, such as Somalis and Ethiopians, Zimbabweans are not inclined to band together to form associations or groups to lobby for and secure protection for their members. Some did suggest that there was safety in numbers and that by doing business in areas where there were many other migrant businesses, the chances of being attacked were considerably reduced. One respondent explained the attraction of running a business in the Johannesburg CBD as follows: “You will find that incidences of xenophobic attacks are very rare in Joburg central where they are a lot of foreigners. Also, Park Station is a strategic location which supplies the whole of South Africa so our protection as foreigners is better” (Johannesburg Interview No. 13). The downside of operating in safer spaces is that business competition is extremely fierce. Most were aware that a great deal of the xenophobic violence was confined to low-income areas, particularly informal settlements. While it was possible for some to avoid doing business in these areas, and instead operating in areas of the city where attacks were less frequent, this was not a feasible option for all. Many Zimbabwean migrants to South Africa do not have the financial resources to afford accommodation nor the means to run a business anywhere other than informal settlements. A number of the respondents noted that the unpredictability of the attacks made it difficult to plan in advance. Some said that they made sure that they did not keep all of their stock at the place of business, storing some of it at home or in rented containers. All tried to minimise the amount of cash they kept on the premises, although not many Zimbabwean entrepreneurs have access to formal banking facilities. One noted that as soon as he had made some money, he immediately remitted it to Zimbabwe “so that even if I am attacked, there is nothing much that they can take from me. It is better if my AHMR, Vol.3 No2, May-August 2017 – Special Issue 899 family can have that money” (Polokwane Interview No. 6). Another said that he was planning to relocate once he had sufficient capital saved up: Surely experience is the best teacher but I think you plan when you have money, so I am thinking of saving a lot of money and looking for safer business locations like in town. I am thinking so because in 2008 they also attacked my business. They just broke and took away all my stuff, now they have burnt the structure down. So I re-constructed and started again so I am now thinking how I am to keep myself and my stuff safe (Cape Town Interview No. 15). Various reactive strategies were mentioned when their businesses were attacked. These included temporarily ceasing business operations, staying indoors at home and moving to stay with friends or relatives in other parts of the city “until the dust settled,” as one put it. Others said that the best strategy was simply to flee the area (or as one graphically put it “you run with your life”), if possible taking some valuable item with them which they could later sell and restart the business with. None of the respondents said that xenophobic attacks would put them permanently out of business. On the contrary, most said they would simply raise the capital and start up again. The logical implication of the determination to stay in business is that xenophobic violence has failed in its two main aims: to drive migrant entrepreneurs out of business and to drive them out of the country and back to Zimbabwe. The respondents were asked if they would return to Zimbabwe as a result of xenophobic attacks and the general consensus was that they would not. A significant number noted that they had settled in South Africa with their families and did not want to return. Many more made reference to the fact that the crisis in Zimbabwe meant that there was nothing for them to return to, even if they wished to do so: There is nothing in Zimbabwe. I am not going back. I am trying to make my life here. My wife is here and my child is here. I am not going back there. Zimbabwe is a country I love. It’s just that at the moment things are tough and there is really nothing to do when you return back home (Johannesburg Interview No. 11). AHMR, Vol.3 No2, May-August 2017 – Special Issue 900 While the hardships which I face in South Africa are many they are still better than the hardships I endured back in Zimbabwe. In the event of future attacks, I could try and survive because at least I will be doing something (Johannesburg Interview No. 13). I could never go back because there are no means of surviving. I could simply have to look for an alternative way to survive while in South Africa. Even if they attack me I will look for another means to survive as long as I am not dead (Johannesburg Interview No. 19). I am not going back. There is nothing to do in Zimbabwe especially because we left a long time ago. What will we do there? So we stay here because this is where our life is. We are establishing here and so if you leave you have to start again. I am not going back. When xenophobia starts we simply move to areas that are safe and return when it is quiet (Cape Town Interview No. 12). Perceptions of Government Inaction All of the respondents were asked about the response of the South African and Zimbabwean governments to xenophobic violence. The responses ranged from the outright cynical to the totally dismissive. Not a single respondent said they had been helped by either government and none were prepared to defend their response to xenophobic violence. Most were extremely critical of both governments. The general consensus was not that the governments did not do enough but rather that they did nothing at all. In the case of the Zimbabwean Government, the prevailing sentiment was captured by one Johannesburg respondent who said: The Zimbabwean Government does nothing. I have never heard them comment or say anything about these attacks. They do not help us at all. They do not send anyone to come and see how we are living and even provide us with assistance. There is no government that helps us (Johannesburg Interview No. 2). Explanations for why the government “does nothing” ranged from sheer lack of interest in what happened to Zimbabweans outside the country, a lack of AHMR, Vol.3 No2, May-August 2017 – Special Issue 901 resources to do anything to help and a desire to see Zimbabweans return home instead of staying in South Africa. Much harsher criticism was reserved for the South African Government’s practice of “doing nothing”: The South African government does not do anything. At least nothing that I know of. They are just silent. We just see the police, but they come too late and do not do anything. They do not arrest anyone even though you report. They are just moving about, but really doing nothing. I sometimes think that even the police hate us the foreigners. Would they do the same and not help if foreigners attack local people? No, they would arrest us. So the police do not help us and would rather see us gone. Even the community leaders do not do anything (Johannesburg Interview No. 2). Some of the people in government are fuelling xenophobia. They are also xenophobic because they say a lot of things that are not true. Like we are the ones who are causing problems here. They had problems here before we came. They are very corrupt but they are the ones that tell people that foreigners are the cause of the problems. People listen to the government. They keep saying that foreigners are bad. What do you expect the people to do? The people follow their government (Cape Town Interview No. 5). They are the ones that cause it so they do not care. The one that occurred in Durban it was the King who incited people. Now he is saying he did not do it but we all saw him on TV. The government does not care for us. They care about their people only. If it were foreigners doing violence against the locals, we would all be in jail (Cape Town Interview No. 6). Some felt that xenophobic violence was tolerated by government because it supposedly achieved the desired effect of getting “foreigners” to leave the country. A recurring theme was police inaction during episodes of mass xenophobic violence. Some felt that the police were extremely slow to respond. As one respondent noted “the police usually come late when everything has been done and people have been killed or their goods stolen” (Cape Town Interview AHMR, Vol.3 No2, May-August 2017 – Special Issue 902 No. 5). Others commented on how the police offered no protection even when they were present: Both the City of Cape Town and the police are not protecting us at all. Like on the day that the people were demonstrating, the police were there. They were just walking. After they were passing, the people started taking our things. There was no one to protect us and no one to stop those people. So, I don’t know what they are doing. I think they just put on uniform and walk around. When there is trouble they don’t come to protect us (Cape Town Interview No. 4). The police just stand at the robots. Or they run away. There is poor enforcement because their response is very slow. Containers were being opened and things taken while the police stared. They are either scared of the people or because it’s their own people so they can’t stop them. There were three police vehicles, but they just stood while people’s containers were being opened. Only foreign containers were broken and they knew whose container it is. No containers for local Xhosas (South Africans) were broken into and destroyed (Cape Town Interview No. 12). One respondent felt that the reason for inaction was that “South Africans do not fear police” and compared the police behaviour with that in Zimbabwe: They throw stones at the police. Have you ever seen people throwing stones at the police in Zimbabwe? No, they do not do that. Here they just do what they want. So they attack foreigners even if the police are there. Unless the police are using teargas or throwing water. But they rarely do that. But you can run to a police station if you are close and seek refuge. There are other areas where the local people even attacked police stations – attacked foreigners in police stations (Cape Town Interview No. 7). There was also a pervasive view that there was little point in reporting theft or assault to the police because nothing was ever done, based on past experiences of police inaction. Dockets may have been opened but the perpetrators were rarely arrested and brought to book and stolen goods were rarely, if ever, recovered: AHMR, Vol.3 No2, May-August 2017 – Special Issue 903 It was the mob that took the things and what would I tell the police? Besides there were many people whose goods were destroyed that I never bothered. The police do not help much. It is useless to report to the police. The police here do not care. Especially if you are a foreigner. They will just tell you it is a mob. They cannot arrest a mob (Johannesburg Interview No. 2). The argument that the police were not particularly concerned by what happened to “foreigners” was very common. One respondent claimed that even if a perpetrator was arrested, “as soon as they have gone around the corner they will ask for a bribe and release the person. As soon as the person is released they will either come and shoot you or permanently injure you” (Johannesburg Interview No. 18). Another said that they had reported a robbery to the police and even named the assailants but little was done: They took down my details and the details of the things I lost. I listed all of them and went with them to the police station. I was told that they would call me when they have made progress and that was that. I went back but there was no progress. The officer who was dealing with the issue kept telling me there were no suspects and that there was nothing they could do. I even gave them some names of the suspects because I had seen some of them, but the police officer did not even take them down. He insisted that there needed to be a witness for him to put those people as witnesses. I thought he should have at least questioned them or gone to their homes and searched. Neither was done (Polokwane Interview No. 4). Apart from the failure to protect, in a xenophobic environment in which migrants are extremely vulnerable, there is always the possibility that the police themselves might seek to take advantage of the situation for their own personal gain. This was certainly the view of many of the respondents who described persistent police harassment, and even theft, during business hours: They know that we are not South Africans. Sometimes the metro and police can just come and take your products. During winter they came and took socks and hats. Once you just try to confront them, they tell you AHMR, Vol.3 No2, May-August 2017 – Special Issue 904 that this is not your country, go back to your country. Tomorrow the same thing can happen again. The police officer will just come and say they lost the gloves and take another pair. If that day they are in the mood of arresting people, they will arrest you despite the fact that they took your things before. Some are those who arrest you and ask for a certain amount of money like ZAR 200 even if you don’t have it. Maybe that day you only made ZAR 50 and if you try to explain that you don’t have the money, they threaten to take all your stock. If the stock value is more than ZAR 200 and I don’t have it, I am forced to ask from other people. If they assist me, I give them and they go and if not, they take all my stock (Johannesburg Interview No. 1). Confiscation of stock appears to be relatively common and the owners are forced to pay large fines to retrieve their goods. In many cases, the fines are so heavy that they simply abandon the goods, borrow money and begin again. Simply to be allowed to operate in an area for a day or to avoid impounding of goods may require payment of a bribe of up to ZAR 200. Mobile vendors play a continuous cat and mouse game with the police, ready to pack up their goods and disappear at the first sign of a police car. In sum, police protection cannot be counted on during episodes of mob violence and there is also very little redress when individuals report crimes against their businesses or themselves to the police. Fear of reprisals from those they report or identify is also a very real disincentive to getting the police involved. As a result, there is a certain fatalism and resignation to the inevitability of losing goods and property in general or isolated attacks. Conclusion Crush and Ramachandran (2015a) argue that there are three main policy and scholarly responses to violence against migrants in general, and migrant entrepreneurs in particular: xenophobia denialism (the official position of the South African government since 2008 and supported by some researchers who argue that South Africans are equally as vulnerable to violence as migrants); xenophobia minimalism (whose proponents suggest that xenophobia may exist but it is an epiphenomenon and that the real causes lie elsewhere) and xenophobia realism (which argues that xenophobia is not only AHMR, Vol.3 No2, May-August 2017 – Special Issue 905 widespread and real but can take a violent form in specific places and under certain circumstances). This paper revisits these arguments from the perspective of a group of migrants themselves, that is Zimbabweans running businesses in the informal economy. The migrants clearly have no difficulty in naming what happens to them as xenophobic. Nor do they hesitate, on the basis of first-hand experience, to name South Africans as the “owners of xenophobia.” Their accounts clearly demonstrate that they see xenophobia as a key driver of the hostility, looting and violence that they experience. We suggest in this paper that xenophobic violence has several key and common characteristics that constantly put Zimbabwean informal enterprise owners at risk of losing their and their property. We also argue that the deep-rooted crisis in Zimbabwe, which has driven many to South Africa in the first place, makes return home in the face of xenophobia a non-viable option. Instead, Zimbabweans are forced to adopt a number of self-protection strategies, none of which are ultimately an insurance against attack. Acknowledgements We would like to acknowledge the financial support of the International Development Research Centre (IDRC) and the EU-funded Migrants in Countries in Crisis (MICIC) Project. References Amit, R. and Kriger, N. 2014. Making migrants' il-legible': The policies and practices of documentation in post-apartheid South Africa. Kronos, 40: 269- 290. Bekker, S. 2015. Violent xenophobic episodes in South Africa, 2008 and 2015. African Human Mobility Review, 1: 229-252. Betts, A. 2013. Survival Migration: Failed Governance and the Crisis of Displacement. Ithaca, NY: Cornell University Press. Betts, A. and Kaytaz, E. 2009. National and International Responses to the Zimbabwean Exodus: Implications for the Refugee Protection Regime. UNHCR Research Paper No. 175, Geneva: UNHCR. AHMR, Vol.3 No2, May-August 2017 – Special Issue 906 Blaauw, D., Pretorius, A. and Schenck, R. 2016. Day labourers and the role of foreign migrants: For better or for worse? Econ 3x3, May. Cabane, L. 2015. Protecting the ‘most vulnerable’? The management of a disaster and the making/unmaking of victims after the 2008 xenophobic violence in South Africa. International Journal of Conflict and Violence, 9: 56- 71. Charman, A. and Piper, L. 2012. Xenophobia, criminality and violent entrepreneurship: Violence against Somali shopkeepers in Delft South, Cape Town, South Africa. South African Review of Sociology, 43: 81-105. Chikanda, A. and Tawodzera, G. 2017. Informal Entrepreneurship and Informal Cross-Border Trade Between Zimbabwe and South Africa. SAMP Migration Policy Series No. 74, Cape Town: SAMP. Crush, J. 2008. The Perfect Storm: The Realities of Xenophobia in Contemporary South Africa. SAMP Migration Policy Series No. 50, Cape Town: SAMP. Crush, J. and Ramachandran, S. 2015a. Migration myths and extreme xenophobia in South Africa. In: Arcarazo, D. and Wiesbrock, A. (Eds.). Global Migration: Old Assumptions, New Dynamics. Santa Barbara, CA: Praeger. Crush, J. and Ramachandran, S. 2015b. Doing business with xenophobia. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa: IDRC, pp. 25-59. Crush, J. and Tawodzera, G. 2017. South-South migration and urban food security: Zimbabwean migrants in South African cities, International Migration, 55(4): 88-102. Crush, J. and Tevera, D. (Eds.). 2010. Zimbabwe’s Exodus: Crisis, Migration, Survival. Ottawa: IDRC and Cape Town: SAMP. Crush, J., Chikanda, A. and Skinner, C. (Eds.). 2015a. Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa: IDRC. AHMR, Vol.3 No2, May-August 2017 – Special Issue 907 Crush, J., Chikanda, A. and Tawodzera, G. 2015b. The third wave: Mixed migration from Zimbabwe to South Africa. Canadian Journal of African Studies, 49: 363-382. Crush, J., Ramachandran, S. and Pendleton, W. 2013. Soft Targets: Xenophobia, Public Violence and Changing Attitudes to Migrants in South Africa After May 2008. SAMP Migration Policy Series No. 64, Cape Town: SAMP. Desai, A. 2015. Migrants and violence in South Africa: The April 2015 xenophobic attacks in Durban. Oriental Anthropologists, 15: 247-259. Duri, F. 2016. From victims to agents: Zimbabweans and xenophobic violence in post-apartheid South Africa. In: Mawere, M. and Marongwe, N. (Eds.). Myths of Peace and Democracy? Bamenda: RPCIG. Everatt, D. 2011. Xenophobia, state and society in South Africa, 2008-2010. Politikon, 38: 7-36. Gastrow, V. 2013. Business robbery, the foreign trader and the small shop: How business robberies affect Somali traders in the Western Cape. SA Crime Quarterly, 43: 5-16. Gastrow, V. and Amit, R. 2016. The role of migrant traders in local economies: A case study of Somali spaza shops in Cape Town. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa: IDRC,. Harber, A. 2011. Diepsloot. Jeppestown: Jonathan Ball. Hassim, S., Kupe, T. and Worby, E. (Eds.). 2008. Go Home or Die Here: Violence, Xenophobia and the Reinvention of Difference in South Africa. Johannesburg: Wits University Press. Hayem, J. 2013. From May 2008 to 2011: Xenophobic violence and national subjectivity in South Africa. Journal of Southern African Studies, 39: 77-97. Hendow, M., Pailey, R. and Bravi, A. 2016. Migrants in Countries in Crisis: A Comparative Study of Six Crisis Situations. Report for International Centre for Migration Policy Development, Vienna: ICMPD https://scholar.google.ca/citations?view_op=view_citation&hl=en&user=GxruDvQAAAAJ&sortby=pubdate&citation_for_view=GxruDvQAAAAJ:dhFuZR0502QC https://scholar.google.ca/citations?view_op=view_citation&hl=en&user=GxruDvQAAAAJ&sortby=pubdate&citation_for_view=GxruDvQAAAAJ:dhFuZR0502QC http://www.tandfonline.com/doi/abs/10.1080/03057070.2013.767538 http://www.tandfonline.com/doi/abs/10.1080/03057070.2013.767538 AHMR, Vol.3 No2, May-August 2017 – Special Issue 908 Hungwe, C. 2012. The migration experience and multiple identities of Zimbabwean migrants in South Africa. Journal of Social Research, 1: 132-138. Hungwe, C. 2013. Surviving Social Exclusion: Zimbabwean Migrants in Johannesburg, South Africa. PhD Thesis. Pretoria: University of South Africa. Hungwe, C. 2014. Zimbabwean migrant entrepreneurs in Kempton Park and Tembisa, Johannesburg: Challenges and opportunities, Journal of Enterprising Culture, 22. IOM 2012. Moving to Safety: Migration Consequences of Complex Crises. International Organization for Migration Report. Geneva: IOM. Koser, K. 2014. Non-citizens caught up in situations of conflict, violence and disaster. In: Martin, S., Weerasinghe, S. and Taylor, A. (Eds.). Humanitarian Crises and Migration: Causes, Consequences and Responses. London: Routledge. Kuljian, C. 2013. Sanctuary. Auckland Park: Jacana. Landau, L. (Ed.). 2012. Exorcising the Demons Within: Xenophobia, Violence and Statecraft in Contemporary South Africa. Johannesburg: Wits University Press. Lefko-Everett, K. 2010. The voices of migrant Zimbabwean women in South Africa. In: Crush, J. and Tevera, D. (Eds.). Zimbabwe’s Exodus: Crisis, Migration, Survival. Ottawa: IDRC. McAdam, J. 2014. Conceptualizing ‘crisis migration’: A theoretical perspective. In: Martin, S., Weerasinghe, S. and Taylor, A. (Eds.). Humanitarian Crises and Migration: Causes, Consequences and Responses. London: Routledge. Makina, D. 2010. Zimbabwe in Johannesburg. In: Crush, J. and Tevera, D. (Eds.). Zimbabwe’s Exodus: Crisis, Migration, Survival. Ottawa: IDRC. Martin, S., Weerasinghe, S. and Taylor, A. (Eds.). 2014. Humanitarian Crises and Migration: Causes, Consequences and Responses. London: Routledge. Misago, J-P., Freemantle, I. and Landau, L. 2015. Protection from Xenophobia: An Evaluation of UNHCR's Regional Office for Southern Africa's Xenophobia Related Programmes. Pretoria: UNHCR AHMR, Vol.3 No2, May-August 2017 – Special Issue 909 Peberdy, S. 2016. International Migrants in Johannesburg’s Informal Economy. SAMP Migration Policy Series No. 71, Cape Town: SAMP. Perching, B. 2016. Actor and Stakeholder Involvement in Crisis Mitigation. MICIC Research Brief. Vienna: ICMPD. Piper, L. and Charman, A. 2016. Xenophobia, price competition and violence in the spaza sector in South Africa. African Human Mobility Review, 2: 332-361. Sibanda, O. and Sibanda. F. 2014. 'Land of opportunity and despair': Zimbabwean migrants in Johannesburg. Journal of Social Development in Africa, 29: 55. Steinberg, J. 2012. Security and disappointment: Policing, freedom and xenophobia in South Africa. British Journal of Criminology, 52: 345-360. Tawodzera, G., Chikanda, A., Crush, J and Tengeh, R. 2015. International Migrants and Refugees in Cape Town’s Informal Economy. SAMP Migration Policy Series No. 70, Cape Town: SAMP. Tevera, D. 2013. African migrants, xenophobia and urban violence in post- apartheid South Africa. Alternation, 7: 9-26. Thebe, V. 2017. ‘Two steps forward, one step back’: Zimbabwean migration and South Africa’s Regularising Programme (the ZDP). Journal of International Migration and Integration 18: 612-633. Weerasinghe, S. and Taylor, A. 2015. On the margins: Noncitizens caught in countries experiencing violence, conflict and disaster. Journal on Migration and Human Security, 3: 26-57. http://samponline.org/wp-content/uploads/2016/10/Acrobat71.pdf https://bjc.oxfordjournals.org/content/52/2/345.full https://bjc.oxfordjournals.org/content/52/2/345.full AHMR, Vol.3 No2, May-August 2017 – Special Issue 910 Linking Harare and Johannesburg through Informal Cross-Border Entrepreneurship Godfrey Tawodzera* and Abel Chikanda** Abstract Zimbabwe has witnessed a rapid expansion of informal cross border trading (ICBT) with neighbouring countries over the past two and a half decades. That expansion has largely been due to a persistent decline in the economy since the introduction of the Economic Structural Adjustment Programme (ESAP) in the 1990s, which led to the closure of many industries, increased unemployment and forced many people into the informal sector. This 2014 study sought to provide a current picture of ICBT in Zimbabwe by interviewing 514 informal entrepreneurs involved in ICBT between Harare (Zimbabwe) and Johannesburg (South Africa). The sample profile revealed that ICBT in Zimbabwe is dominated by females and young adults and that traders are fairly educated. The study results demonstrate the important role played by ICBT in the survival of households in the country. The traders make important contributions to the Zimbabwean economy through business establishment, providing goods that are unavailable in the country, availing goods cheaper than the formal retail sector and benefiting the fiscus through import duties. In South Africa, ICBT benefits wholesalers from whom traders purchase their goods and supports the South African transport and hospitality industries. The report concludes that ICBT in Zimbabwe has become more than a survivalist strategy. It is contributing to the economy by generating jobs and reducing unemployment. Therefore, there is a need for policies that encourage rather than restrict the operation of informal trade. Keywords Cross-border traders, informal economy, international trade, Harare, Johannesburg. * Department of Geography & Environmental Studies, University of Limpopo, P. Bag X1106, Sovenga ** Department of Geography & Atmospheric Science and African & African-American Studies, University of Kansas, Lawrence, Kansas State, USA 66045. AHMR, Vol.3 No2, May-August 2017 – Special Issue 911 Introduction Across the Southern African Development Community (SADC) region, informal cross-border trade (ICBT) is a well-established entrepreneurial activity accounting for 30-40% of intra-regional trade (Makombe, 2011; Mijere, 2009; Peberdy et al., 2015). For Zimbabwe, ICBT not only expands the market for Zimbabwean goods (such as handicrafts) but also imports a wide variety of perishables and non-perishables for the Zimbabwean consumers. ICBT is not a new phenomenon in Zimbabwe, dating back to the 1990s when the country started experiencing serious economic hardships (Gaidzanwa, 1998; Kurebwa, 2015). The first decade after Zimbabwe’s independence in 1980 was a period of relative growth and stability: the real GDP growth rate averaged 7% per annum, average annual inflation was in the single digits, budget deficit as a percentage of GDP was generally around 5% per annum and export growth averaged around 15% per year (Malaba, 2006). However, government expenditure rose dramatically over the same period to over 50% of GDP by 1990, largely due to increased spending on health, social services and infrastructure (Brown, 2000). Consequently, there was relatively less spending on capital projects to grow the economy. The result was increasing unemployment, which reached 26% by 1990, up from just 8% at independence (Robinson, 1991). National government debt reached 71% of GDP in 1989 (Munjoma, 1999) and by 1997, per capita incomes were lower than in 1990 (Mupedziswa & Gumbo, 2001). A decade after the country’s independence, and against this backdrop of a contracting economy, the Government of Zimbabwe launched a series of economic reforms. The first phase, launched in 1990, was the International Monetary Fund (IMF) and the World Bank-funded Economic Structural Adjustment Programme (ESAP) (Mupedziswa & Gumbo, 2001). This was succeeded by the Zimbabwe Programme for Economic and Social Transformation (ZIMPREST) in 1997 (Government of Zimbabwe, 1997). The austerity measures imposed on the economy by these reforms (such as the removal of subsidies on food and agricultural inputs, the removal of controls on prices, labour and wages and the operation of social services at cost recovery levels) led to a host of negative impacts that included the closure of AHMR, Vol.3 No2, May-August 2017 – Special Issue 912 many factories, massive retrenchments, declining real wages and skyrocketing consumer prices. The resultant decline in the formal economy led to a growth in the informal sector (Kamete, 2017; Tibaijuka, 2005). ICBT as a household livelihood strategy also began to escalate, as thousands of jobless workers had no choice but to enter the informal economy. Despite South Africa imposing tough visa restrictions on Zimbabweans after 1995, cross-border trading continued to be an important source of livelihood for many Zimbabweans. Just as South Africa removed visa restrictions in 2005, the Government of Zimbabwe launched an assault on all forms of urban informality through Operation Murambatsvina (Chirisa, 2007; Potts, 2006; Tibaijuka, 2005). Though this operation temporarily interrupted the trade, ICTB quickly re- established itself as informal entrepreneurs had few viable alternatives. In 2007, at the height of Zimbabwe’s economic crisis, SAMP conducted a border monitoring study of informal traders at four major border posts and found that the trade was not only increasing and diversifying, but more men were also becoming involved (Tevera & Tawodzera, 2007). As Zimbabwe’s shops emptied and the purchasing power of the Zimbabwean dollar evaporated during the economic meltdown of 2008, ICBT traders played a critical role in ensuring that essential supplies, such as foodstuffs, remained accessible to many households. The volume of informal trade at the Zimbabwean borders with South Africa and Botswana reached record proportions during the crisis (Ama et al., 2013; Campbell & Crush, 2014; Kachere, 2011). In 2009, the Zimbabwean economy stabilised after the inauguration of the Government of National Unity. Stabilisation was short-lived as the country’s economic situation has deteriorated once again. However, the period of economic stability following the crisis of 2007-8 does raise the question of whether and what impact it had on ICBT. On the one hand, many goods that were formerly unavailable in the country returned to the shops. On the other, formal sector unemployment remained stubbornly high and many households continued to rely on the informal economy for their livelihoods (Tawodzera, 2014). In June 2016, the Government of Zimbabwe, through Statutory Instrument 64 of 2016, banned the importation of a wide variety of goods, including various processed and tinned foods, bottled water, dairy products, AHMR, Vol.3 No2, May-August 2017 – Special Issue 913 household furniture, fertilizer, cotton fabric and some building materials. According to the Minister of Industry and Commerce, the move was designed “to support our local industry […] buying locally manufactured goods; whether it involves purchasing machinery worth millions or even just a T-shirt; begins a cycle in which you re-invest money into the local economy, instead of spending it on an imported product and sending the money outbound” (Pindula News, 2016). The announcement triggered widespread protests by informal cross-border traders in Harare and at the Beitbridge border post between Zimbabwe and South Africa. The ostensible reason was that the livelihoods of many in the informal economy, who made a living importing goods from South Africa and Botswana, were under threat by the state (Kalaba, 2016; Scoones, 2016). A recent study has estimated that there are as many as 3.5 million Micro Small and Medium (MSME) enterprises in Zimbabwe (Finmark Trust, 2014). Of these, 2 million (or 71%) are individual entrepreneurs, 800,000 have employees and nearly half reported that the business was their only source of income (Finmark Trust, 2014). Additionally, 85% of these businesses were not registered and therefore operated in the informal economy, hence underscoring the importance of the informal sector to the livelihoods and sustenance of many Zimbabweans. Against this backdrop, the Growing Informal Cities project set out to examine the nature and role of contemporary ICBT between Zimbabwe and South Africa by interviewing a sample of ICBT entrepreneurs in Harare in 2014. In this paper, we discuss the findings of this survey and demonstrate that despite many challenges and obstacles, ICBT continues to thrive. Research Methodology The research on which this paper is based employed both qualitative and quantitative techniques to obtain a profile of Zimbabwean cross-border traders and insights into their motivation, operations, challenges and opportunities. A questionnaire was administered to a total of 534 ICBT traders in Harare (Table 1). Only traders who were trading in the informal sector (not registered for tax purposes); had been in operation for at least a year to allow for a retrospective analysis of the start-up, problems and opportunities; and AHMR, Vol.3 No2, May-August 2017 – Special Issue 914 were conducting business primarily between Harare and Johannesburg were interviewed. Because it is impossible to draw a representative sample of cross- border traders, the study adopted a snowball sampling strategy. This involved identifying a small group of cross-border traders in multiple locations across Harare. These traders were then used as initial sampling units and were asked to identify other traders to be included in the sample. Each interviewed cross- border trader was asked to identify at least three other traders, from which one was randomly selected for interview. The process was repeated until the desired sample size was achieved. Table 1: Sampled Areas of Harare No. % Harare CBD 263 49.6 Mbare and Mupedzanhamo 110 20.6 Chitungwiza, Makoni and St Mary’s 36 6.7 Highfield and Machipisa 33 6.1 Avondale 20 3.7 Kuwadzana 17 3.2 Warren Park 1 15 2.8 Epworth 12 2.2 Glen View 3, 7 and 8 12 2.2 Mabvuku – Tafara 12 2.2 Other 4 0.7 Total 534 100.0 In addition to the questionnaire survey, 24 in-depth interviews were conducted with selected cross-border traders in Harare using a semi- structured interview schedule. The interviews allowed for further probing on AHMR, Vol.3 No2, May-August 2017 – Special Issue 915 the origins and contemporary context of ICBT as well as other key issues surrounding the trade. Two focus group discussions were also held in Harare and focused on understanding the nature of the trade, relationships with local authorities and the police and the challenges faced in the traders’ everyday lives. Three in-depth interviews were also conducted with traders’ organisations in the city that were able to supply information relevant to cross-border trading, particularly issues relating to space, licensing and how traders organise. Because of the complexity of ICBT, it is possible that some trading activities were not captured in the survey. Those operating home- based businesses, and those selling to middlemen, may be under-represented, for example. These limitations notwithstanding, the size of the sample from across the city is likely to give a generally accurate picture of Harare’s cross- border trade. Profile of ICBT Entrepreneurs ICBT from Zimbabwe has traditionally been dominated by female traders with limited involvement of males (Muzvidziwa, 2001, 2015). In Southern Africa as a whole, women make up an estimated 70-80% of informal cross-border traders (UNIFEM, 2010). The majority of traders interviewed in this study were females (68%) with males constituting only 32%. The sample was also relatively young, with a mean age of 33 years. The dominance of young people in cross-border trading between Zimbabwe and South Africa is explained by the arduous nature of the trade, which involves a great deal of travelling and spending considerable time waiting at border posts and the places where goods are sourced. Cross-border trade in the 1990s was associated with the less educated and unskilled (Muzvidziwa, 2015). However, this Harare sample was better educated with only 1% of the sample having not completed primary school, while two-thirds had a high school diploma, 17% had some secondary level education, 11% had a college certificate and 3% had at least an undergraduate degree. The entry of better-educated people into ICBT is certainly a result of continuing high unemployment levels in Zimbabwe. In terms of marital status, 17% of this sample had never been married, 66% were married and 16% were divorced, separated or widowed. AHMR, Vol.3 No2, May-August 2017 – Special Issue 916 The largest number of traders were living in nuclear households (43%), while 28% were from extended households (Table 2). A significant minority (17%) were from female-centred households while only 5% were male-centred. What this suggests is that female ICBT traders came from a variety of different household structures, which would have meant different domestic roles and responsibilities and different abilities to trade full-time. In general, the array of people involved in ICBT indicates that it is no longer the preserve of a few. An economic activity once completely dominated by low-income households and a relatively uneducated population has become a viable livelihood alternative for others reeling under economic pressure. Although still dominated by women, male traders are involved in increasing numbers as ICBT has increasingly become a business for all. Table 2: Household Structure of ICBT Traders No. % Female-centred 97 18.3 Male-centred 26 4.9 Nuclear 228 42.9 Extended 145 27.3 Single person household 35 6.6 Total 531 100.0 Only 1% of the ICBT businesses surveyed were established before 1990. This is primarily because the economy of the country was still strong and most workers were being absorbed in the formal sector. Around 11% were established in the 1990s, the decade of the Economic Structural Adjustment Programme (ESAP), while another 15% started cross-border trading between 2000 and 2005. By far the largest proportion of ICBT traders began operations during and after the economic crisis of 2008, with nearly three-quarters starting up between 2006 and 2014. The rate of start-up seems to have fallen slightly after 2010, most probably as a result of the stabilisation of the AHMR, Vol.3 No2, May-August 2017 – Special Issue 917 economy after the establishment of the GNU as well as the more stable political environment. Because most food became available in the country through formal sector imports, the incentive to continue ICBT, especially involving food, was no longer as compelling. Because high rates of unemployment are thought to be one of the main reasons pushing people into the informal economy and ICBT, the survey sought to find out what occupation the traders were engaged in before starting their enterprises. Around one-third were unemployed (Table 3). Another 22% were already working in the informal economy either running a business that did not involve ICBT or being employed by someone else (6%). This suggests that the majority of traders went straight into ICBT, rather than expanding into ICBT from an existing enterprise. Another 36% of the traders were employed in a formal sector job immediately prior to establishing an ICBT business. Table 3: Occupation Before Starting ICBT Business No. % Unemployed Unemployed/job seeker 178 33.4 Informal Economy Operated own informal sector business in Zimbabwe only 89 16.7 Employed by someone in the informal economy 32 6.0 Employed Office worker 40 7.5 Manual worker (skilled) 35 6.6 Manual worker (unskilled) 35 6.6 Professional (e.g. lawyer, doctor, academic, engineer) 15 2.8 Teacher 10 1.9 AHMR, Vol.3 No2, May-August 2017 – Special Issue 918 Health worker 9 1.7 Employer/manager 6 1.1 Police/military/security 4 0.8 Domestic worker 3 0.6 Agricultural worker 2 0.4 Other occupation 32 6.0 Other Businessman/woman formal sector (self-employed) 7 1.3 Scholar/student 36 6.8 Total 533 100.0 Around 30% of the sample indicated that they had additional sources of income while 70% survived solely on the ICBT business. The majority of those with alternative income-generating sources indicated that they were involved in non-ICBT businesses (55%), while 11% were deriving income from part- time or casual work (Table 4). According to one trader, having more than one business was a necessity given the unpredictable nature of the informal trade: I sell clothes and shoes, but I also have another income from selling foodstuffs. Clothes do not sell faster than food so when I am in need of faster cash I know that I will always get some from trading food. However, the profit from food is not much so it just takes care of my immediate needs. When I am looking for serious money, I know I will get it from clothes and shoes even if they are bought less frequently (Interview No. 4, Harare, 27 September 2014). AHMR, Vol.3 No2, May-August 2017 – Special Issue 919 Table 4: Other Income Generating Activities No. % Another non-ICBT business 85 55.9 Part-time/casual work 17 11.2 Another ICBT business 14 9.2 Formal employment in the private sector 7 4.6 Formal employment in the public sector 3 2.0 Rentals 3 2.0 Other work 23 15.1 Total 152 100.0 Entrepreneurial Motivation There is a general assumption in the literature on the informal economy in Zimbabwe that people participate because they have no choice. This would imply that most participants in ICBT are necessity-driven entrepreneurs, pushed to participate by the need to survive. However, ICBT is not the easiest or most obvious form of informal activity and may attract those with different, more opportunity-driven entrepreneurial motivations. Therefore, the study examined the reasons why traders started an ICBT business. Respondents were asked to rate 27 factors as motivations for starting their ICBT business on a scale from 1 (no importance) to 5 (extremely important). The responses confirm the importance of financial survival in pushing people into cross- border trading (Table 5). Easily the most important factors were the need to make more money just to survive (a mean score of 4.9) and the need to give their families greater financial security (4.6). This suggests that ICBT is largely a necessity-driven survival strategy to generate income. Employment-related factors, including unemployment and having an undesirable job, were significantly less important. AHMR, Vol.3 No2, May-August 2017 – Special Issue 920 Table 5: Entrepreneurial Motivation of ICBT Entrepreneurs Mean Score Survivalist factors I needed more money just to survive 4.9 I wanted to give my family greater financial security 4.6 I was unemployed and unable to find a job 3.0 I had a job but it did not pay enough 2.6 I wanted to provide employment for members of my family 2.3 I had a job but it did not suit my qualifications and experience 1.2 Entrepreneurial orientation I have always wanted to run my own business 4.2 I wanted more control over my own time/to be my own boss 3.9 I like to challenge myself 2.9 I wanted to do something new and challenging 2.7 I have the right personality to be involved in cross-border trade 2.6 I wanted to compete with others and be the best 2.5 I like to learn new skills 2.5 I enjoy taking risks 2.4 Market opportunities I wanted to expand my business 3.2 I had a good idea for a product to sell to people in Zimbabwe 3.1 I started cross-border trading to grow my business 3.1 There were shortages of certain goods in my home country 3.0 AHMR, Vol.3 No2, May-August 2017 – Special Issue 921 I had a good idea for a product to sell in South Africa 1.2 I had a good idea for a product to sell to people from my home country 1.2 Building and utilising social capital I wanted to increase my status in the community 3.2 I wanted to contribute to the development of my home country 2.7 Support and help in starting my business was available from family members 2.5 I wanted to provide employment for other people 2.1 My family members have always been involved in cross-border trade 2.1 Support and help in starting my business was available from other traders 1.9 I decided to go into business in partnership with others 1.5 The second set of factors relates more to self-assessment of personality traits usually associated with entrepreneurship. A desire to own one’s own business did rate relatively highly (at 4.2), as did the related desire to be one’s own boss (3.9). However, the other personality-related factors had low ratings, which suggest that the traders did not see themselves as natural entrepreneurs. They were generally more alert to the market opportunities provided by participation in ICBT, with a number of factors scoring around 3.0 on average including awareness of shortages in Zimbabwe, having a good idea for a product to sell, and wanting to expand an existing business. Financing ICBT The amount of start-up capital required to engage in ICBT has been a deterrent for low-income individuals and households in the past (Muzvidziwa, 2007). In this survey, almost three-quarters of the sample had minimal start-up capital AHMR, Vol.3 No2, May-August 2017 – Special Issue 922 of ZAR 5,000 or less, which suggests that the entry threshold may have declined (Figure 1). Only 8% had started their business with a capital outlay of more than ZAR 10,000. According to some respondents, a basic ICBT business can be established with as little as ZAR 2,000. As one commented: Cross-border business does not need a lot of money to start. You only need to be clever and buy things that sell faster and are wanted by the people. Then you can increase your stock as time goes on. I started my business with ZAR 1,500 buying and selling rice when the country was in dire need of food. I would buy rice from Musina and sell in Harare and do two trips a week. I used trucks which are cheaper and did not sleep in South Africa, so I reduced my costs (Interview No. 9, Harare, 28 September 2014). Most of the traders acquired their start-up capital from multiple personal sources. About 58% used their personal savings, 38% borrowed money from relatives, while 2% borrowed from non-relatives and 1% from informal money lenders. Very few (1.2%) started their ICBT business with money from formal sources such as banks and micro-finance institutions. While start-up capital is important for the business to take off, most businesses require a constant injection of cash during their life-span, especially in the initial stages (Abdulsaleh & Worthington, 2013). In this survey, almost one-third (27%) of the traders reported borrowing additional money for their business in the year prior to the study (Table 6). AHMR, Vol.3 No2, May-August 2017 – Special Issue 923 Figure 1: Capital used to start an ICBT business Table 6: Sources of Operating Loans No. % Loan from relatives 77 53.2 Micro-finance institution 21 15.0 Bank 17 12.0 Informal financial institution 9 6.3 Money lenders 9 6.3 Loan from other business owners 6 4.2 Business credit (goods on terms) 1 0.7 Other sources 9 6.3 Total 144 100.0 AHMR, Vol.3 No2, May-August 2017 – Special Issue 924 The majority reported that they had borrowed from relatives. According to the traders, relatives are the best source of loans because they do not place stringent conditions on the borrower and are likely to be understanding should they default on the loan or delay the repayment: Of course I have borrowed a number of times to finance my business, but I only borrow from my close relatives. They are very understanding when you can’t pay them back or delay the repayment. If you borrow from other people, they give you all sorts of problems in that they can even come and take your goods as surety, or even sell your goods (Interview No. 10, Harare, 29 September 2014). A small number of traders had obtained additional loans from micro-finance institutions (15%) and banks (12%). With the introduction of the US dollar (USD) as one of the trading currencies in the country in 2009, a number of micro-finance institutions were established in Harare. These organisations lend money on a short-term basis, ranging from days to months, but rarely for periods exceeding a year. Their interest rates are generally higher than those offered by the banks but they are less stringent on surety requirements, accepting such assets as cars, fridges, stoves, TVs and washing machines as surety. Banks, on the other hand, generally prefer immovable properties such as houses. In the absence of such property, a borrower is expected to show evidence of employment and ability to repay through submitting pay-slips. One trader compared the two in this way: I borrow many times when my stocks are low and I need to boost my business. I usually borrow from micro-finance houses where I can just leave the registration book of my car and get it when I repay them. Banks require many documents which I do not have. I also do not own a house for surety. So I prefer the microfinances as they are simple. I can apply in the morning and get the money before the end of the day (Interview No. 7, Harare, 28 September 2014). Other sources of loans include informal financial institutions and money lenders. Stokvels (money clubs) lend money to members at reasonable rates AHMR, Vol.3 No2, May-August 2017 – Special Issue 925 and the repayment period is usually mutually agreed upon with the borrower who is also a member. Money lenders, however, tend to lend at punitive rates and can use brutal coercive methods to enforce repayment in case of a default. As a result, most ICBT traders avoid doing business with them. A few traders (4%) said they preferred to borrow from each other before going to banks and money lenders. This is because the terms of repayment from fellow traders are very favourable and there is a mutual understanding of the challenges of the trade. How much do the traders borrow from these various sources? Loans from fellow traders were ZAR 16,667 on average, compared to ZAR 12,187 from banks, ZAR 11,214 from micro-finance institutions, ZAR 7,444 from informal financial institutions, ZAR 6,335 from relatives and ZAR 3,527 from money- lenders. A key determinant of where money is borrowed is the interest charged on the loan. The highest interest rates are charged by money lenders (averaging 23%), followed by stokvels (18%), micro-finance institutions (17%) and banks (12%). Almost all traders (95%) said they use loans to purchase goods for resale, while others use the funds to meet transportation costs (57%) and for rental of the market stalls (10%). In Harare, some traders have no stalls of their own and rent from others on a daily, weekly or monthly basis. Most ICBT traders use public transport (buses and haulage trucks) to ferry their goods from South Africa and incur significant costs. Traders who take arts and curio products to South Africa to sell use the money to transport their goods as far afield as Cape Town and Port Elizabeth. What is clear is that loans are used for basic business expenses and not for other household needs. Many traders (51%) indicated that the borrowed money significantly increased the volume of their sales (Figure 2). Others reported an increase in the profitability and competitiveness of their business (23%), and diversification of the products they traded (8%). However, not all traders had positive experiences. As one recounted: I borrowed ZAR 20,000 from a micro-finance in March so that I could buy more stock. However, that loan has caused me more problems than any real benefits because from that time everything has been AHMR, Vol.3 No2, May-August 2017 – Special Issue 926 going downhill. The interest rate was such that I was expected to pay back the ZAR 20,000 plus ZAR 5,000 interest in two months. When I bought more stock from Johannesburg, I paid too much duty at the border and when I resold the goods in Harare, I failed to raise the initial ZAR 20,000. The micro-finance then took my car and gave me seven days to pay, which I failed to do and my car was sold (Interview No. 3, Harare, 27 September 2014). Figure 2: Impact of loans on ICBT business activity Less than 20% of the ICBT traders had applied for a bank loan to finance their business at start-up or for ongoing operations. When asked about the reasons for low patronage of banks, around one-quarter indicated that they had no interest in getting a bank loan, 14% that banks asked for too much collateral, while 13% were dissuaded by the high interest rates, and 12% said the application process was too complicated for them to understand (Table 7). Only half of those who did apply for a bank loan were successful. Most were rejected because they had insufficient collateral. According to a trader whose application had been rejected, the documentation required by the bank was also especially onerous: AHMR, Vol.3 No2, May-August 2017 – Special Issue 927 I have applied for a loan and every time my application is rejected I am given a different reason. The first time I was told that I did not have sufficient collateral even though I had submitted title deeds for a plot. The second time I was told that the amount of money I needed was too much and could not be covered by the plot. I eventually just gave up. But it is frustrating after having wasted so much time and money looking for documents and printing, and travelling to and from the bank (Interview No. 5, Harare, 28 September 2014). Table 7: Main Reasons for Not Applying for a Bank Loan No. % Not interested in getting a loan 101 24.5 Guarantee/collateral asked for is too much 59 14.3 Interest rates are too high 54 13.1 Application procedures too complicated 49 11.9 Did not think it would be approved 47 11.4 Don't have a bank account 46 11.2 No need for a loan – had sufficient capital 33 8.0 Other reason 23 5.6 Total 412 100.0 Buying in Johannesburg, Selling in Harare ICBT traders from Harare are highly mobile, spending only 1.79 days in South Africa on average per visit, although a few stay for several weeks, particularly if they are taking goods to sell. The traders travel relatively frequently to South Africa, with 67% making at least one trip per month and 82% travelling more than four times per year. The vast majority do not have their own means of transport to South Africa and rely on bus services (99%), trucks (3.9%), their own vehicles (1.1%) and taxis (0.6%). Only a few traders (0.2%) travel by AHMR, Vol.3 No2, May-August 2017 – Special Issue 928 plane. Some ICBT traders use their services of informal transporters (omalayisha) as it saves them the trouble of having to transport their own goods through the border. The omalayisha are well-known to the customs officials on both sides of the border, which enables them to carry goods without paying customs duties (Thebe, 2015). Chinese malls in Johannesburg are an important source of goods for traders (Laribee, 2008). In Johannesburg, the majority of Zimbabwean ICBT traders purchase their goods at outlets such as China Mall (26%), other Chinese-run malls (25%) and Oriental Plaza (5%) (Figure 3). Some buy from wholesalers (14%) while others patronise independent small shops/retailers (13%) or small shops/retailers in malls (8%). A much smaller number purchase goods from factories, supermarkets, the informal sector and direct from farmers. Figure 3: Source of goods in Johannesburg The kinds of goods purchased in South Africa in 2015 for resale in Zimbabwe varied widely (Table 8). By far the most common items were new clothing and footwear, with 88% of all traders buying these items. Other goods included accessories such as bags and suitcases (bought by 27%), bedding materials AHMR, Vol.3 No2, May-August 2017 – Special Issue 929 (20%), electronics (12%) and household products (8%). Foodstuffs, which were carried by three-quarters of ICBT traders in 2008, were not at all popular any longer in 2015 (Peberdy et al., 2015). This is a reflection of post-2009 developments in Zimbabwe that led to a more abundant supply of food products in the country (Tawodzera, 2014). On a typical trip, the traders spend an average of ZAR 8,328 on a variety of products for resale, with a minimum spend of ZAR 1,000 and a maximum of ZAR 150,000. Only a small number of the traders take products from Zimbabwe to sell in South Africa. These include arts and crafts, fabrics, bedding, household products and new clothing and footwear. Table 8: Goods Purchased in South Africa No. % New clothing and footwear 467 87.6 Accessories (bags, sunglasses, etc.) 141 26.4 Bedding 104 19.5 Electronics 64 12.0 Household products 41 7.7 Hardware/tools 31 5.8 Toiletries and cosmetics 30 5.6 Cell phones/accessories 24 4.5 CDs/DVDs 15 3.5 Cooking oil 11 2.1 Pre-owned clothing and footwear 9 1.7 Furniture 5 0.9 Stationery 5 0.9 Rice and pasta 4 0.8 AHMR, Vol.3 No2, May-August 2017 – Special Issue 930 Tea/coffee 4 0.8 Sugar 4 0.8 Beds and mattresses 4 0.8 Milk (fresh/sour) 3 0.6 Spare parts and raw materials 2 0.4 Eggs 2 0.4 Snacks 2 0.4 Car parts 2 0.4 Confectionary 1 0.2 Mealie meal 1 0.2 Bread 1 0.2 Tinned/canned fruits and vegetables 1 0.2 Plastic goods 1 0.2 Toys, sports equipment 1 0.2 Note: multiple response question Although the overwhelming majority of the traders sell their goods in the country’s capital, where the survey was conducted, more than 10% also sell goods in other major cities such as Bulawayo and Masvingo, and in smaller urban centres such as Gokwe and Mutoko. As many as 86% sell their products in their own stall in an informal market while 24% sell through friends, family and other personal networks, 16% from their houses, 10% from their own shop in the informal sector, 9% on the street, 7% door-to-door and 2.5% in offices and other workplaces. AHMR, Vol.3 No2, May-August 2017 – Special Issue 931 Economic Contributions An important objective of this research was to assess the importance of ICBT trade to the economies of both Zimbabwe and South Africa. This is particularly important in the case of Zimbabwe where the informal economy is viewed as a problem by government and a hindrance to the success of formal businesses (Ngundu, 2010; Rogerson, 2016). First, it is clear from the wide variety of goods imported by the ICBT traders that there is a ready market in Zimbabwe for their products. This means, in turn, that they are servicing the needs of Zimbabwean consumers either by providing goods that are unavailable in the country or at a cheaper rate than supplied by local producers and the formal retail sector. Given high rates of unemployment and the fact that many households are cash-strapped, the ICBT traders play a critical role in servicing the basic consumer needs of Zimbabweans. Second, the ICBT enterprises contribute to the economy through business establishment, growth and the strategic deployment of their profits. On average, the surveyed traders were recording total sales worth ZAR 12,000 per month. Sales ranged from as low as ZAR 1,000 per month for the small traders to as high as ZAR 130,000 per month for those dealing in large quantities of goods, especially those trading in hardware and building materials. The reported profits averaged ZAR 4,765 per month but were also highly variable, ranging from a minimum of ZAR 200 per month to a maximum of ZAR 45,000 per month. Traders with high sales turnover generally reported higher monthly profits. The size of these profits indicates that ICBT is not just a survivalist enterprise, but rather a viable business operation. More than half (59%) of the ICBT operators reported that their income had increased since embarking on ICBT while a further 24% reported that their income was variable, sometimes increasing and at other times decreasing. Only 15% indicated that their income had decreased since starting an ICBT business, buttressing the important role that ICBT plays in the livelihoods of many Zimbabweans. Third, given that the income of the majority of the traders had improved significantly through ICBT, it is important to know whether or not they use their profits for the businesses or themselves. Over 90% reported spending AHMR, Vol.3 No2, May-August 2017 – Special Issue 932 their profits on family needs, including paying rent for accommodation, buying food for the family and other obligatory family expenses (Table 9). Nearly three-quarters said that they reinvest their profits in the business, presumably to buy more stock. Table 9: Use of Business Profits No. % Spent on family needs in Zimbabwe 489 91.6 Re-investment in business 391 73.2 Education 230 43.1 Spent on self 210 39.3 Personal savings 152 28.5 Spent on family needs outside Zimbabwe 3 0.6 Retirement fund 3 0.6 Re-investment in business outside Zimbabwe 4 0.7 Note: Multiple response question Fourth, as well as stimulating the economy through the reinvestment of profits inside and outside the business, ICBT traders contribute to Zimbabwe’s economy through job creation. Most traders personally procure goods in South Africa and hire people to sell them in Harare. A total of 37% of the traders employ people in their businesses. Of the 308 people employed in the enterprises surveyed, 236 (or 77%) were in paid positions. Even when members of the family are involved in the business, they tend to be compensated for their time. Of the 128 family members employed, 71% were in paid positions. As many as 61% of the paid jobs went to non-family members. In sum, employment generated by the ICBT traders is shared fairly evenly between family and non-family members and there is a general preference for female labour in the business hires. AHMR, Vol.3 No2, May-August 2017 – Special Issue 933 Finally, other beneficiaries include the Zimbabwean bus companies and the general Zimbabwean fiscus, which benefits from duties levied at the border on returning traders bringing in products whose value is above the duty free limit. On their last trip to South Africa, the ICBT traders paid ZAR 431 on average in import customs duties, which amounts to ZAR 228,529 levied on these traders alone. Given that this sample is only a small proportion of the total number of ICBT traders and that each trader crosses several times a year, the total duties levied undoubtedly run into ZAR millions. In terms of contributions to the South African economy, the ICBT traders contribute, firstly, through expenditures while in the country. The largest beneficiaries are South African wholesalers and others from whom the traders purchase their goods. The traders also support the South African transport and hospitality industries. Overall, traders spent an average of ZAR 8,039 on their last trip to South Africa. The spending on particular items included ZAR 6,737 on goods (or ZAR 3.5 million in total), ZAR 683 on transportation (ZAR 361,000), ZAR 42 (ZAR 22,000) on accommodation and ZAR 141 on other costs (Figure 4). Figure 4: Amounts spent on last trip to South Africa. AHMR, Vol.3 No2, May-August 2017 – Special Issue 934 Second, South Africa benefits from the value added tax (VAT) paid by the traders. Visitors to South Africa who buy goods for use in their home country are eligible to apply for a VAT refund when they exit the country. Visitors are supposed to request a tax invoice when they buy the goods in South Africa and present it at the border for VAT refund. However, only 22% of the cross-border traders said they claim VAT back at the border. Over one-third (37%) did not even know that they could do so. Another third said that the process of claiming VAT at the border wasted too much time. The delays are exacerbated by the fact that ownership levels of motor vehicles is very low and reliance on public transport high, which means that they only have limited time to wait. Nearly 10% noted that the receipts which they get from South African retailers are invalid for the purpose of claiming VAT. Receipts from Chinese-owned shops, in particular, generally do not meet the requirements for claiming VAT. Others do not claim VAT simply because they do not want to declare their goods and feel that claiming exposes their goods to customs scrutiny. There is clearly a serious lack of information on the procedures for claiming VAT refunds at the border. Unclaimed or unreimbursed VAT benefits the South African Revenue Services (SARS). Business Problems and Challenges Cross-border traders confront numerous challenges in the course of their business activities. These are of two main types: (a) those related to customs and immigration at the border and (b) those related to their daily business operations in both South Africa and Zimbabwe. By far the most common customs and immigration problems relate to the queues and delays at the border. Almost all of the traders indicated that they often experienced delays (Table 10). ICBT has become a way of life for a significant number of Zimbabweans and this has put pressure on immigration officials operating at the Beitbridge and Musina border posts. Another common problem experienced by the traders relates to the high duties levied at the border (86% said they experience this problem). In his mid-term fiscal policy review statement in September 2014, Zimbabwean Finance Minister Patrick Chinamasa increased duty on various finished AHMR, Vol.3 No2, May-August 2017 – Special Issue 935 products including meat and dairy products, beverages, vegetables, soap and furniture. He also removed foodstuffs, beverages and washing preparations from the duty free facility. While the stated rationale for these measures was to ‘protect local industry’ the local companies that once produced these goods closed down at the height of the economic crisis in 2007-8. Consequently, imposing higher duties does not deter traders from bringing these goods into the country but simply burdens the general population that relies on these goods. Table 10: Problems Experienced by Zimbabwean Traders at Border Yes (%) No (%) Long queues/congestion/delays at border post 96.4 3.6 Duties paid are too high 85.5 14.5 Restrictions on import/export of goods (type & volume) 77.6 22.4 Transport problems/poor road networks/transport prices high 72.1 27.9 Too much corruption at border post 66.8 33.2 Days allowed in South Africa are too few 42.0 58.0 Unwarranted confiscation/detention of goods 41.6 58.4 Verbal harassment by South African police, army, border officials 36.3 63.7 Verbal harassment by Zimbabwean police, army, border officials 18.9 81.1 Physical harassment/assault by South African police, army, border officials 18.0 82.0 Physical harassment/assault by other people 16.8 83.2 AHMR, Vol.3 No2, May-August 2017 – Special Issue 936 Physical harassment/assault by Zimbabwean police, army, border officials 9.6 90.4 Other problems cited by a significant number of traders included restrictions on the import or export of goods (78%) and corruption at the border (67%). The problem of corruption has also been highlighted elsewhere (Fitzmaurice, 2009; Kwanisai et al., 2014). The delays at the border have created an opportunity for touts who know that the traders want to cross the border quickly and will pay a small fee for officials to speed up the clearing processes for them. Some traders have also lost money to bogus agents masquerading as immigration and customs officials. Harassment, both verbal and physical, was highlighted as a problem by a small number of traders. For instance, verbal and physical harassment/assault by South African police, army and border officials is experienced by 36% and 18% respectively. In Zimbabwe, the incidence of this is lower at 19% and 10%. Perhaps the biggest business challenge confronting ICBTs is competition from other traders (experienced by 84%) (Table 11). Another significant problem is competition from large retailers or supermarkets (experienced by 60%). These outlets enjoy economies of scale as they buy goods in bulk from suppliers at low prices and are able to price their goods competitively. However, even the large supermarkets cannot outcompete the informal sector traders. For instance, beverages, foodstuffs and detergents are sold on pavements in Harare at cheaper prices than those charged by retail supermarkets. Given the evidence of pervasive animosity of South African informal entrepreneurs towards foreign-owned enterprises, it is notable that few of the ICBT traders from Zimbabwe experience serious problems: only 8% experience harassment by South African competitors (Crush & Ramachandran, 2015). Harassment by the South African municipal authorities is marginally more common (13%), however, as many as 39% had experienced harassment in Harare. Other significant problems experienced by the traders in their daily operations include confiscation of goods and AHMR, Vol.3 No2, May-August 2017 – Special Issue 937 difficulty getting merchandise back thereafter, and discrimination based on nationality. Table 11: Problems Related to Operating ICBT Business Yes % No % Business competition Competition from other traders 84.0 16.0 Competition from large retailers or supermarkets 60.0 40.0 Difficulty negotiating with other traders/conflicts among traders 49.6 50.4 Harassment by South African traders 8.4 91.6 Operational challenges Theft of money or goods 46.9 53.1 No relevant training in accounting, marketing, other business skills 30.5 69.5 Problems securing a selling site in Zimbabwe 29.3 70.7 Policy environment Harassment by authorities in Zimbabwe 39.4 60.6 Confiscation of goods/difficulty getting confiscated goods back 25.1 74.9 Harassment by authorities in South Africa 22.8 87.2 Discrimination Prejudice against my nationality 36.6 63.4 Prejudice against my gender 16.5 83.5 AHMR, Vol.3 No2, May-August 2017 – Special Issue 938 A related dimension of discrimination is the potential challenge posed by xenophobia in South Africa. As many as 78% said they had not been affected by xenophobia at all in their business operations. This suggests that Zimbabwean entrepreneurs on short-term visits are able to avoid the worst excesses of xenophobic violence, unlike their resident compatriots. Robbery is a more serious problem with 30% having been robbed in Zimbabwe and 28% in South Africa. This suggests that ICBT traders are at risk of victimisation in both Zimbabwe and South Africa, primarily because their business is cash- based. In June 2016, for example, a spate of robberies targeting buses carrying cross-border traders from Zimbabwe was reported in South Africa, bringing to the fore the dangers that traders face in their business (The Chronicle, 2016; The Herald, 2016). Conclusion International cross-border trade has become a widespread, entrenched and crucial component of the Zimbabwean economy and livelihoods in generals. Its importance has risen significantly amidst a prolonged decline in the formal economy. This paper has provided detailed insights into contemporary ICBT activities between Harare and Johannesburg, showing that Zimbabwean traders travel regularly to South Africa, with three-quarters of the sample reporting that they visit Johannesburg for business at least once per month. They reported making monthly profits of more than ZAR 4,000 per month, far exceeding the salaries of most people in formal employment in the country. This is despite the small initial capital outlay involved, averaging around ZAR 5,000. ICBT traders have been able to grow their businesses to such an extent that they are able to hire non-family members. ICBT has therefore become more than a survivalist strategy and should be seen as an important pillar of the Zimbabwean economy. The contribution of the informal economy in generating jobs and reducing unemployment needs to be acknowledged by policies that encourage rather than restrict the operation of informal trade. Traders make important contributions through business establishment, providing goods that are unavailable in the country, availing goods cheaper than the formal retail sector and benefiting the fiscus through import duties. AHMR, Vol.3 No2, May-August 2017 – Special Issue 939 In South Africa, ICBT benefits wholesalers and other outlets from which traders purchase their goods, South African transport and hospitality industries and the South African Treasury through unclaimed VAT. Official denunciation of the supposed damage done to South Africa by the non-South African informal sector enterprises rarely mentions ICBT; perhaps this is because the evidence of their benefit to the country is so overwhelming. Acknowledgements We would like to acknowledge the International Development Research Centre (IDRC) for its support of the SAMP Growing Informal Cities Project. References Abdulsaleh, A. and Worthington, A. 2013. Small and medium sized enterprises financing: A review of literature. International Journal of Business and Management, 8: 36-54. Ama, N., Mangadi, K., Okurut, F. and Ama, H. 2013. Profitability of the informal cross-border trade: A case study of four selected borders in Botswana. African Journal of Business Management, 7: 4221-4232. Brown, A. 2000. Cities for the Urban Poor in Zimbabwe: Urban Space as a Resource for Sustainable Development. East Kilbride: UK Government Department for International Development (DFID). Campbell, E. and Crush, J. 2014. ‘They don’t want foreigners’: Zimbabwean migration and xenophobia in Botswana. Crossings: Journal of Migration and Culture, 6: 159-80. Chirisa, I. 2007. Post-2005 Harare: A case of the informal sector and street vending resilience: What options do key actors have? Local Governance and Development Journal, 1: 53-64. Crush J. and Ramachandran, S. 2015. Doing business with xenophobia. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. http://www.africabib.org/query_a.php?pe=%21J00006284%21&SR=3 http://www.africabib.org/query_a.php?pe=%21J00006284%21&SR=3 AHMR, Vol.3 No2, May-August 2017 – Special Issue 940 Finmark Trust. 2014. FinScope MSME Survey Zimbabwe 2012. Harare: Finmark Trust. Fitzmaurice, M. 2009. Situational Analysis at Beitbridge Border Post between Zimbabwe and South Africa. Report by Transport Logistics Consultants, Port Elizabeth: Transport Logistics Consultants. Gaidzanwa, R. 1998. Cross-border trade in southern Africa: A gendered perspective. In: Sachikonye, L. (Ed.). Labour Markets and Migration Policy in Southern Africa. Harare: SAPES Trust. Government of Zimbabwe. 1997. Poverty Assessment Study Survey 1995. Main Report, Harare: Ministry of Public Service, Labour and Social Welfare. Kachere, W. 2011. Informal Cross Border Trading and Poverty Reduction in the Southern African Development Community: The Case of Zimbabwe. PhD Thesis. Alice: University of Fort Hare. Kalaba, M. 2016. Lessons to be Learnt from Zimbabwe’s Blunt Use of an Import Ban. The Conversation, July 27, 2016. Kamete, A. 2017. Governing enclaves of informality: Unscrambling the logic of the camp in urban Zimbabwe. Geoforum, 81: 76-86. Kurebwa, J. 2015. The vulnerability of female cross border traders to HIV and AIDS in Zimbabwe: The case of Beitbridge Border Post. IOSR Journal of Humanities and Social Science, 20: 65-69. Kwanisai, G., Mpofu, T., Vengesayi, S., Mutanga, C., Hurombo, G. and Mirimi, K. 2014. Borders as barriers to tourism: Tourists experiences at the Beitbridge Border Post. African Journal of Hospitality, Tourism and Leisure, 3: 1-13. Laribee, R. 2008. The China Shop phenomenon: Trade supply within the Chinese diaspora in South Africa. Africa Spectrum, 43: 353-370. Makombe, P. 2011. Informal Cross-Border Trade and SADC: The Search for Greater Recognition. Johannesburg: Open Society Initiative for Southern Africa. Malaba, J. 2006. Poverty Measurement and Gender: Zimbabwe’s Experience. Inter-Agency and Expert Group Meeting on the Development of Gender Statistics, United Nations, New York. AHMR, Vol.3 No2, May-August 2017 – Special Issue 941 Mijere, N. 2009. Informal Cross-Border Trade in the Southern African Development Community (SADC). Addis Ababa: OSSREA. Munjoma, T. 1999. Property Investment under an Economic Structural Adjustment Programme: The Case of the Harare Central Business District Office Development. PhD Thesis. Aberdeen: University of Aberdeen. Mupedziswa, R. and Gumbo, P. 2001. Women Informal Traders in Harare and the Struggle for Survival in an Environment of Economic Reforms. Research Report No. 117, Uppsala: Nordiska Afrikainstitutet. Muzvidziwa, V. 2001. Cross-border women traders: Multiple identities and responses to new challenges. Journal of Contemporary African Studies, 19: 67- 80. Muzvidziwa, V. 2007. Women without Borders: Informal Cross-Border Trade Among Women in the Southern African Development Community. Addis Ababa: OSSREA. Muzvidziwa, V. 2015. Gendered nature of informal cross-border trade in Zimbabwe. Journal of Social Development in Africa, 30: 131-146. Ngundu, K. 2010. The Impact of the Informal Economy on the Social and Economic Development of Women-headed Households in Chegutu Urban District. MA Thesis. Pretoria: University of Pretoria. Peberdy, S., Crush, J., Tevera, D., Campbell, E., Zindela, N., Raimundo, I., Green, T., Chikanda A. and Tawodzera, G. 2015. Transnational entrepreneurship and informal cross-border trade with South Africa. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. Pindula News. 2016. Zimbabwe Bans a Host of Goods from Being Imported. Pindula News, June 19, 2016. Potts, D. 2006. Restoring order? Operation Murambatsvina and the urban crisis in Zimbabwe. Journal of Southern African Studies, 32: 273-291. AHMR, Vol.3 No2, May-August 2017 – Special Issue 942 Robinson, P. 1991. What is structural adjustment? In: The Social Implications of Structural Adjustment Programmes in Africa. Proceedings of a Workshop held at University of Zimbabwe, Harare, pp. 54-58. Rogerson, C. 2016. Responding to informality in urban Africa: Street trading in Harare, Zimbabwe. Urban Forum, 27: 229-251. Scoones, I. 2016. Zimbabwe’s Riots: The Rise of the Informal Trader and a New Political Economy. The Conversation, July 14, 2016. Tawodzera, G. 2014. Household food insecurity and survival in Harare: 2008 and beyond. Urban Forum, 25: 207-216. Tevera, D. and Tawodzera, G. 2007. Cross-Border Trade: The Case of Beitbridge, Forbes, Chirundu and Nyamapanda Border Posts. Report for SAMP, Harare: University of Zimbabwe. The Chronicle. 2016. Another Zim Bus Robbed in SA. The Chronicle, June 25, 2016. The Herald. 2016. South Africa: Eagle Liner Bus Robbed, Burnt in South Africa. The Herald, June 21, 2016. Thebe, V. 2015. The malayisha industry and the transnational movement of remittances to Zimbabwe. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. Tibaijuka, A. 2005. Scope and Impact of ‘Operation Murambatsvina’ in Zimbabwe. Report of Fact-Finding Mission, New York: UN-Habitat Special Envoy on Human Settlements. UNIFEM, 2010. Women in Informal Cross Border Trade in Southern Africa. Johannesburg: UNIFEM. AHMR, Vol.3 No2, May-August 2017 – Special Issue 943 Informal Entrepreneurship and Cross-Border Trade between Mozambique and South Africa Abel Chikanda* and Ines Raimundo** Abstract Informal cross-border trading is an essential part of Maputo’s informal economy. This paper presents the results of a 2014 SAMP survey of informal entrepreneurs involved in cross-border trade between Johannesburg and Maputo. A questionnaire was administered to a sample of 403 informal traders in 7 markets in Maputo. The study showed that most of the entrepreneurs began their business activities as vendors and only later moved into cross-border trading. The overwhelming majority used their personal savings to start their business and they face significant barriers in accessing business loans from formal banking channels. The study demonstrates the importance of cross-border traders to both the Mozambican and South African economy. In South Africa, the cross-border traders make a significant contribution by buying local goods and utilising the services provided by the country’s travel and hospitality industry. In Mozambique, they supply affordable products to the country’s growing informal sector and play an important role in generating employment. Keywords Cross-border traders, informal economy, informal sector, migration, South Africa. Introduction Over the past decade, the Mozambican economy has experienced GDP growth rates above 6% (ADB, 2016). This has made Mozambique the fastest growing non-oil economy in Sub-Saharan Africa (Nucifora & da Silva, 2011). However, there has been limited formal employment generation (Jones & Tarp, 2013). Most of the country’s urban working population is still in the informal economy (Jenkins, 2013) and nearly two-thirds of the economically active population in Maputo City is involved in the informal economy in some way * Department of Geography & Atmospheric Science, African & African-American Studies, University of Kansas, Lawrence, Kansas State, USA 66045. Email address: achikanda@ku.edu. ** Faculty of Arts and Social Sciences, University of Eduardo Mondlane, Maputo, Mozambique. AHMR, Vol.3 No2, May-August 2017 – Special Issue 944 (Jenkins, 2012). A 2007 survey found that 70% of households were involved in informal economic activities, with a significantly higher participation rate by female-headed (86%) than male-headed (62%) households (Paulo et al., 2007). Maputo’s informal economy is firmly integrated into the national and local economies of neighbouring states such as Mozambique and Swaziland and the primary mechanism of integration is informal cross-border trade or ICBT. In the late 1990s and early 2000s, female ICBT traders became important players in the Mozambican economy, particularly in the supply of food as well as non-agricultural products such as electrical goods and building materials (Macamo, 1999). A SAMP border monitoring study in Mozambique in 2006- 2007 showed that women made up 71% of the total number of cross-border traders (Peberdy et al., 2015). A significant feature of the post-2000 cross- border trade from Mozambique was its diversity. A number of Mozambican traders (36.5%) said they also traded between Swaziland and Mozambique, buying goods in Swaziland to sell in Maputo (Peberdy, 2000a). Those travelling to South Africa to buy goods for resale tended to gravitate towards Johannesburg and Durban, while the rest travelled to towns in the neighbouring province of Mpumalanga (Peberdy, 2000b). Among Mozambicans, the expressions dumba nengue (trust your feet) and dumba kutsutsuma (run if you can) are used to describe the informal economy (War on Want, 2006). This terminology clearly reflects the tension that exists between informal traders and law enforcement agents. Compared with other cities in the region, however, Maputo has traditionally adopted a more tolerant approach to the informal economy (Kamete & Lindell, 2010; Rogerson, 2016). Policy interventions aim to discourage informality through registration and formalisation rather than by eradication and punishment (Dibben et al., 2015). Two main strategies have been pursued by the municipal government. First, formal urban markets have been established, existing informal markets have been upgraded and vendors now pay rent for stands. When Xikhelene market was upgraded, for example, all trading on the streets around the old market was eliminated (Ulset, 2010). Second, a simplified tax system was introduced that requires traders to pay business tax either as a lump sum or as a AHMR, Vol.3 No2, May-August 2017 – Special Issue 945 percentage of turnover (Dibben et al., 2015). This initiative has been hampered by low uptake and strong resistance from the informal traders. Informal trade is a major catalyst for involvement in informal economies globally (Desai, 2009). The ECA (2010: 143) has noted that “informal trade is as old as the informal economy. It is the main source of job creation in Africa, providing between 20 per cent and 75 per cent of total employment in most countries.” Informal trade also plays a vital role in linking informal economies in different Southern African cities. This requires a perspective on informality that takes into account the impacts of interaction between different urban informal economies across the region (Akinboade, 2005). The volume of cross border trade has been monitored at the Mozambique and South Africa border in the past but this paper provides the opportunity to engage in an in-depth analysis of different types of cross-border trade and how ICBT is integrated into Maputo’s informal economy (Dlela, 2006; Peberdy, 2007; Peberdy & Crush, 1998). There has been a tendency in the past to view informal traders as sole operators rather than micro-enterprises with the potential to grow significantly, to create jobs and to generate the capital to branch out into other sectors of the informal and formal economy. By viewing informal traders as entrepreneurs per se and their activities as a potentially strong promoter of growth and employment, we are in a position to move beyond the idea that ICBT traders are “survivalists” struggling to make ends meet, and rather towards an analysis of the potential for growing their businesses, the obstacles they face and the kinds of policy environment they require in order to realise their entrepreneurial ambitions (Lesser & Moisé-Leeman, 2009; Peberdy & Crush, 2001; Söderbaum, 2007). Research Methodology In the first phase of the research on which this paper is based, a survey questionnaire was administered to a sample of 403 informal traders in 7 Maputo markets. The sample was divided into 3 entrepreneurial categories (Table 1): AHMR, Vol.3 No2, May-August 2017 – Special Issue 946  Category A: 243 Mozambican traders who travel to and from Johannesburg as part of their business, buying goods in South Africa and selling them in Maputo.  Category B: 61 Mozambican traders who travel to and from Johannesburg, buying goods in South Africa and supplying other informal traders in Maputo. These traders supply vendors in the markets and small shops.  Category C: 99 Mozambican informal traders who do not travel, but purchase goods from cross-border traders for resale. In sum, 61% of the sample were cross-border traders who travel to and from South Africa as part of their business, 14% were trader intermediaries who travel to South Africa to buy goods and sell to other traders on return, and 29% did not cross international borders but obtained their goods from those who did. Table 1: Market Location and Types of Trader Markets Category A Category B Category C Total No. of Interviews Xipamanine 58 11 23 92 Malanga and Fajardo 48 13 18 79 Xiquelene 42 9 16 67 Estrela Vermelha 37 9 13 59 Mandela 23 6 11 40 Museu 22 6 10 38 Praça de Touros 13 7 8 28 Total 243 61 99 403 AHMR, Vol.3 No2, May-August 2017 – Special Issue 947 The sampling procedure tried to capture the range of products sold in different markets as markets are getting more specialised, primarily because of their geographical location and potential customer base. Most customers who buy at Museu Market, for example, come from the relatively wealthy areas of Museu, Polana and Sommerschield, while people who buy at Xiquelene mainly come from poor wards. Foodstuffs are the primary product at Xiquelene, while Museu's main products are alcohol and cigarettes. Praça de Touros market is situated in one of the busiest areas of Maputo and mainly caters to vehicle owners who need spare parts and also has a car-repair garage. Estrela Vermelha market, also known as Red Star Shopping Centre, sells a variety of goods including household furniture, alcohol and cigarettes. This market is located between Central and Alto Mae wards, and suburbs such as Mafalala, and serves people in both middle- and low-income brackets. Not all of the goods in the markets come from South Africa. In addition to locally sourced agricultural products, goods come from as far afield as Europe and South America. For instance, frozen chicken from Brazil is a common sight in many of the markets (de Oliveira et al., 2015). There is also a major trade in imported second-hand clothes. At the Xipamanine market in Maputo, for instance, hundreds of traders sell second-hand clothing from Australia, Europe and North America (Brooks, 2012, 2013; Ericsson & Brooks, 2014). Many of the clothes that reach Xipamanine are of low quality and, according to one researcher, neither improve the lives of the vendors nor the consumers, an outcome which he terms “clothing poverty” (Brooks, 2015). In the second phase of the research, four focus group discussions were conducted with traders from the following markets: Xiquelene, Zimpeto, Malanga and Museu. Key informant interviews were also conducted with different government, private sector and international organisation stakeholders with an interest in the informal economy. Among the issues covered in the interviews were the history of cross-border trade in the country, the role of cross-border trade in economic development, legislation governing the informal economy and the challenges faced by cross-border traders. AHMR, Vol.3 No2, May-August 2017 – Special Issue 948 Profile of ICBT Traders There is a general perception that cross-border trading in Mozambique is predominantly undertaken by women, while men tend to stay at home and sell products brought into the country by women (de Vletter & Polanda, 2001; Peberdy 2000a; Peberdy et al., 2015). However, some studies have shown greater participation of men in Mozambican informal cross-border trade over time (Macamo, 1999). In this study, 56% of those interviewed were men (Table 2). It is likely that the profile of the respondents would have weighted more heavily in favour of women if the study had been conducted at the border. Women tend to find it easier to be involved in cross-border trade and leave the marketing of the products to either their husbands or sons (Raimundo, 2010). The predominance of women in cross-border trade has been attributed to their long experience in crossing borders dating back to the early days of Mozambique’s 16-year civil war; their business acumen; their familiarity with managers of wholesale storehouses in Johannesburg; and the fact that they find it more difficult than men to access formal employment (Raimundo, 2010). Male respondents claimed that females are better equipped to deal with customs officials and have strategies for avoiding paying import duties; that they themselves viewed trading as more of a hobby or a way to generate extra income; and that they have greater access to jobs in the formal sector. Racially, the overwhelming majority of the respondents (99%) were Black, while a minority were of mixed race, Indian or Asian. The mean age of the sample was 37 years with 42% aged between 30 and 39 and 28% between 40 and 49. Around 10% were over the age of 50, and the oldest respondent was 78. Participation by young people in the trade was relatively limited with only 20% under the age of 30. Across all the categories, individuals aged between 20 and 49 made up more than 90% of the total participants. AHMR, Vol.3 No2, May-August 2017 – Special Issue 949 Table 2: Gender, Race and Age Profile of Traders Cross- border traders (%) Cross-border trader intermediaries (%) Non-cross- border traders (%) Total Gender (N=402) Male 35.1 6.5 14.9 56.5 Female 26.3 7.2 10.0 43.5 Total 61.4 13.7 24.9 100.0 Race (N=391) Black 98.3 98.2 100.0 98.7 Mixed race 0.4 1.8 0.0 0.5 Indian 0.4 0.0 0.0 0.3 Other Asian 0.4 0.0 0.0 0.3 Other race 0.4 0.0 0.0 0.3 Age (N=395) Below 20 0.0 0.0 1.0 0.3 20-29 18.9 16.7 28.6 21.0 30-39 42.8 44.4 36.7 41.5 40-49 29.6 29.6 23.5 28.1 50-59 6.2 7.4 9.2 7.1 60-69 2.1 1.9 0.0 1.5 70 and above 0.4 0.0 1.0 0.5 AHMR, Vol.3 No2, May-August 2017 – Special Issue 950 Involvement in the informal economy is usually associated with low levels of educational attainment (Amaral & Quintin, 2006; Wilson et al., 2006). As many as 75% of the respondents had less than high school qualifications, and fewer than 2% had completed an undergraduate degree (Figure 1). Interestingly, traders who do not travel to Johannesburg themselves but obtain goods from cross-border traders, had the highest level of education, with 28% having completed at least high school, compared to 22% of the cross-border traders who travel to Johannesburg as part of their business and 16% of the intermediaries who supply goods to non-cross-border traders in Maputo. Figure 1: Highest level of education of traders. In understanding how household structure affects participation in the business of informal trade, marital status needs to be considered. For example, are the traders independent operators looking out for themselves? Are they heads of households that depend on them for a livelihood? Or are they just contributing to household income? Given that this business can require cross- border traders to be away from the household for several days at a time, there was an expectation that it might be dominated by single, widowed and divorced people. However, the survey found that only 29% were single (which 0 5 10 15 20 25 30 35 N o f o rm a l sc h o o li n g S o m e p ri m a ry sc h o o li n g P ri m a ry (c o m p le te d ) S o m e s e co n d a ry S e co n d a ry / H ig h sc h o o l/ d ip lo m a U n d e rg ra d u a te d e g re e P o st g ra d u a te d e g re e O th e r P e r c e n ta g e AHMR, Vol.3 No2, May-August 2017 – Special Issue 951 tallies with the more mature age profile of the entrepreneurs) and 6% were divorced or widowed. Of the rest, 38% were married or in a common law relationship and a further 26% were co-habiting. More than half of the respondents (52%) came from nuclear households (defined as a family made up of husband/male partner and wife/female partner with or without children). About 16% came from male-centred households, where there is no wife/female partner in the household and 13% were from female-centred households. Another 11% were from extended households, while only 7% lived alone. An AFSUN survey in Maputo in 2008/2009 showed that the majority of residents in the poorer areas of the city came from extended households, even though female-centred and nuclear households were also common (Raimundo et al., 2014). This seems to suggest variation in patterns of participation by type of household. In total, 66% of households are male-headed nuclear and extended family households. These households constitute 63% of the traders’ households (although the reason why extended families are far less likely to participate than nuclear families is unclear). The major difference is in female-centred household participation where these households constitute 27% of all households but only supply 13% of traders, suggesting that these household heads are inhibited from participation because they cannot leave their children. Male-centred households appear not to have a problem, with 8% of the total number of households supplying 16% of the traders. Business Start-Up and Ownership There is an assumption in the literature that many informal entrepreneurs are pushed into participation by unemployment (Jones & Tarp, 2013; Sparks & Barnett, 2010). However, only 26% of the respondents were unemployed before they started their business. Another 19% had been students (Table 3). Around 30% had been employed, primarily in low-paying jobs such as domestic work (9%), agriculture (5%) and unskilled manual labour (3%). Less than 5% had occupied skilled or semi-skilled positions. The rest were already employed (8%) or self-employed (17%) in the informal economy in another enterprise. AHMR, Vol.3 No2, May-August 2017 – Special Issue 952 Table 3: Occupation Prior to Starting the Business No. % Unemployed/job seeker 105 26.3 Pupil/student 76 19.0 Operated informal sector business based only in Mozambique 68 17.0 Domestic worker 37 9.3 Employed by someone in the informal economy 30 7.5 Agricultural worker 19 4.8 Manual worker (unskilled) 12 3.0 Businessman/woman formal sector (self-employed) 11 2.8 Manual worker (skilled) 7 1.8 Office worker 6 1.5 Police/military/security 3 0.8 Teacher 3 0.8 Health worker 2 0.5 Professional (e.g. lawyer, doctor, academic, engineer) 1 0.3 Other occupation 19 4.8 Total 399 100.0 In contrast to ICBT traders from countries such as Zimbabwe, the majority of the Maputo City entrepreneurs had been involved in the informal economy for many years. As many as 43% had established informal businesses before 2000 and another 41% had done so between 2000 and 2010. Less than 20% had set up shop in the last five years, which may suggest that entry into a highly competitive business is becoming more difficult. The survey also showed that most of the entrepreneurs began their business activities as vendors and only later moved into cross-border trading. So, while over 40% had started their businesses before 2000, only 21% were engaged in ICBT at that time (Figure 2). Conversely, while 57% established their businesses after 2000, the proportion who started ICBT activities during this period was close to 80%. This indicates that most ICBT traders are post-2000 entrants to that market, a direct result of various economic factors and market opportunities. Factors of relevance include the devastating floods of 2000, which impacted many AHMR, Vol.3 No2, May-August 2017 – Special Issue 953 households; the lifting of South African visa restrictions in 2005; and the strengthening of the metical in comparison to the rand (Christie & Hanlon, 2001). Figure 2: Year of business start-up and entry into cross-border trade. The overwhelming majority of the respondents started their businesses from a very low base with three-quarters having less than ZAR 10,000 (approximately USD850) to invest (Figure 3). Of these, the majority had between ZAR 2,001 and ZAR 5,000 (USD170-USD425). Just 15% had invested ZAR 20,000 (USD 1,700) or more. This group was most likely to be cross- border traders (83%) rather than cross-border intermediaries or non-cross- border traders. The non-cross-border traders tended to have the lowest levels of start-up capital with nearly 90% investing ZAR 10,000 or less in their businesses. The sources of capital used to start the business varied but the majority (82%) had used personal savings. Other sources of start-up business capital included loans from relatives (used by 33%), bank loans (9%) and loans from informal financial sources (8%). Access to formal sources of business capital was limited, constituting a general reflection of the lack of support given to informal enterprises by formal financial institutions. 0,0 10,0 20,0 30,0 40,0 50,0 60,0 1971-1980 1981-1990 1991-2000 2001-2010 2011-2014 P e r c e n ta g e Year business started Year started ICBT activities AHMR, Vol.3 No2, May-August 2017 – Special Issue 954 Figure 3: Amount of money used to start business. One recent study showed that nearly 80% of Mozambicans have no access to any sort of banking or microfinance services (Finmark Trust, 2009). Only 12% of the sample had access to banking services; 10% to the informal microfinance sector and 1% to formal microfinance services. However, this study of informal entrepreneurs in Maputo found that although most could not get bank loans, as many as 44% had a bank account. Lack of access to business financing from formal sources is common among informal entrepreneurs. Benjamin and Mbaye (2012) have noted that informal entrepreneurs in Africa frequently fail to access business finance because of the onerous procedures required for loan applications as well as the collateral requirements. The situation is no different in Mozambique: 27% indicated that they cannot apply for bank loans because of the high interest rates and 24% do not have interest at all in applying for bank loans. In an effort to improve financial access to informal entrepreneurs, Moza Bank became the first private bank of Mozambique to give loans to informal traders through the ASSOTSI - Informal Cross-Border Trade Association. The loans are only available to ASSOTSI 0,0 5,0 10,0 15,0 20,0 25,0 30,0 R 5 0 0 o r le ss R 5 0 1 - R 1 ,0 0 0 R 1 ,0 0 1 - R 2 ,0 0 0 R 2 ,0 0 1 - R 5 ,0 0 0 R 5 ,0 0 1 - R 1 0 ,0 0 0 R 1 0 ,0 0 1 - R 2 0 ,0 0 0 R 2 0 ,0 0 1 - R 3 0 ,0 0 0 R 3 0 ,0 0 0 - R 4 0 ,0 0 0 R 4 0 ,0 0 1 - R 5 0 ,0 0 0 R 5 0 ,0 0 1 a n d a b o v e P e r c e n ta g e Cross border traders Cross border trader intermediaries Non-cross border traders AHMR, Vol.3 No2, May-August 2017 – Special Issue 955 members who can present a reference from ASSOTSI and a simplified tax return. The loans may be up to one-third of the annual business turnover shown in the tax return, to a maximum loan of MZN100,000 (Baxter & Allwright, 2015). The programme officer of the Competitiveness and Private Sector Development Project noted that several government funding schemes are available to small and medium-scale enterprises, but not to businesses in the informal economy: The Government of Mozambique through the Ministry of Commerce and Industry has established the Enterprise Competitiveness and Private Sector Development Project, which funds small and middle enterprises for competitiveness. The State does not recognise unlicensed businesses which are run by cross-border traders. To be funded one needs to be licensed. The informal economy is not eligible for funds (Interview, 25 November 2015). Just over one-third of the respondents were aware of the scheme but only 2% had applied, suggesting that they are aware of the ineligibility of informal enterprises for government assistance. Business Strategies The research sought to examine the ICBT linkages and flows between Maputo and Johannesburg. Johannesburg was the primary place, and beneficiary, of the purchase of goods by Mozambican traders. On average, the cross-border traders spend 1.52 days in South Africa on each trip. As many as 54% travel to Johannesburg at least once a week, which translates to nearly 80 days per year spent in South Africa. A further 34% travel there at least once a month, which translates to 18 days per year in South Africa (Figure 4). As many as 8% of the respondents indicated that they also buy goods in other places in Gauteng, such as Pretoria and Randfontein. More than one-third of the respondents purchased goods from other places in South Africa, especially from towns close to the Mozambican border such as Nelspruit, Malelane and Komatipoort. The traders are therefore able to conduct their business activities in South Africa and return to Mozambique within the same day. A small number of respondents (around 12%) also travel to other countries to conduct their AHMR, Vol.3 No2, May-August 2017 – Special Issue 956 business, including Swaziland, China and Dubai. As many as 27% of those who travel to these other countries for business are non-Mozambican nationals. Figure 4: Frequency of trips to Johannesburg Most of the traders sell the goods from South Africa in Maputo, the capital, while a small number (less than 5%) also sell their goods in other cities such as Xai Xai and Beira. The goods are sold mainly in their own shops in the informal sector (38%) or in their own stalls in an informal market (24%). However, there is also evidence of informal-formal sector linkages with 9% selling in their own shops in the formal sector, 8% selling to retailers, 3% to wholesalers and 1% to restaurant owners. A total of 424 other people were employed directly in the businesses interviewed (Figure 5). Around half (51%) provide employment to others, or an average of 2.1 jobs per business. A significant proportion of the traders employ more than one person: 27% of those providing employment had two employees, 10% had three employees and 5% had four or more employees. There was a major difference in the employment practices between those who travel to South Africa and those who do not (58% versus 31% providing jobs). This confirms that many cross-border traders prefer to focus on their cross- border activities and employ others to sell the goods on their behalf in Maputo. Non-family members made up 69% of the paid employees and the rest were 0 10 20 30 40 50 60 At least once every week At least once every month More than 4 times per year 2 - 4 times per year At least once per year Other P e r c e n ta g e AHMR, Vol.3 No2, May-August 2017 – Special Issue 957 family members, with a total of eight employees below the age of 18. Some 71% of the employees were men, confirming that there is an explicit focus on male employment in the businesses supported by ICBT. This seems to support our earlier observation that men prefer not to cross borders but are employed by women in the sale of goods in Mozambique. Figure 5: Employment by traders The involvement of children in the informal economy is a controversial issue (ILO, 2004; Thorsen, 2012). Some regard their involvement as an essential part of household survival strategy (Becker, 2004). Others view it as child labour that limits the proper development of children and should therefore be eliminated (Burra, 2005). The survey found that 16% of enterprises involve children in their business in different capacities. Participation was higher among children of cross-border traders (20%) than non-cross-border traders (7%). Among those who involve their children, 53% said their children help them to sell goods, 43% sometimes ask their children to look after the stall and 5% involve them in other ways. Participation in cross-border trading is sometimes viewed as a part-time activity to supplement income from other sources. The overwhelming majority of respondents (90%) have no other income-generating activity and rely solely on their ICBT-supported business for survival. Discussions of informal entrepreneurship have focused on whether the sector can create viable enterprises that can provide decent incomes to the participants (ILO, 0 50 100 150 200 250 300 Family members - over 18 years Family members - under 18 years Other people from your country People from other countries N u m b e r o f e m p lo y e e s AHMR, Vol.3 No2, May-August 2017 – Special Issue 958 2009; Sparks & Barnett, 2010). In other words, can cross-border traders generate incomes comparable to formal sector jobs? The traders generate an average of ZAR 21,838 per month (USD 1,850) in total sales and a profit of ZAR 7,087 per month (USD 600) (Table 4). Cross-border traders who travel to Johannesburg as part of their business activities are likely to generate more monthly sales and profit than non-cross-border traders or cross-border trader intermediaries. Table 4: Total Monthly Sales and Profit Mean monthly sales (ZAR) Mean monthly profit (ZAR) Cross-border traders 25,639.89 8,169.68 Cross-border trader intermediaries 11,098.04 6,101.39 Non-cross-border traders 16,416.67 5,095.15 Total average 21,838.28 7,086.82 Note: 1 ZAR = 0.085 USD at the time of the survey Cross-border entrepreneurial activity is certainly financially beneficial to the participants. Nearly two-thirds (64%) said that their income status had increased compared to the period before they started their business and only 6% said it had decreased. Another 26% said that their income was variable, fluctuating according to market conditions. The greatest improvement was reported by the cross-border traders. The profits generated from informal business play an important role in meeting personal (79%) and family (77%) needs. One-quarter of respondents were investing the proceeds in education of family members, more than the proportion re-investing income in the business itself (only 19%). One-third said that profits were being saved for future use. Just under 10% send money outside of Mozambique as remittances either to support the needs of their family members or for investment in business. AHMR, Vol.3 No2, May-August 2017 – Special Issue 959 Contributions to the South African Economy Who benefits most in South Africa from the purchasing behaviour of Mozambican cross-border entrepreneurs? This section examines the activities of those cross-border traders who travel to South Africa and seeks to identify the South African beneficiaries of Mozambican ICBT (Peberdy et al., 2015). First, ICBT between South Africa and Mozambique contributes to South Africa’s massive trade surplus by exporting South African goods and importing far less from Mozambique. Only 5% of the traders sell products from Mozambique in South Africa (including cigarettes, fabric/capulana, fresh fruit and vegetables and alcohol). Most of these products are sold through personal networks, but they are also sold to wholesalers and informal vendors. Second, a wide variety of outlets in South Africa benefit from patronage by Mozambican traders, including wholesalers, supermarkets, small retailers (formal and informal), factories, farms and fresh produce markets. Easily the most important beneficiaries of Mozambican patronage are South African wholesalers (used by 48%). Other important sources of goods for the traders include the Johannesburg Fresh Produce Market (21%), supermarkets (16%), manufacturers (16%) and small shops or retailers either in a mall (15%) or outside a mall (9%) (Table 5). Chinese shops are also popular with the cross- border traders, with 19% buying goods from the China Mall and 11% buying goods from other Chinese-run malls. AHMR, Vol.3 No2, May-August 2017 – Special Issue 960 Table 5: Source of Goods in South Africa % Wholesalers 48.4 City Deep/Johannesburg Fresh Produce Market 20.7 China Mall 18.4 Supermarkets 16.4 Manufacturer/factories 15.8 Small shops/retailers (in a mall) 14.8 Other Chinese-run mall 10.5 Small shops/retailers (not in a mall) 8.9 Direct from farmers 5.3 Other fresh produce markets 4.9 Informal sector producer/retailer 3.6 Oriental Plaza 1.0 Other place 2.3 N=304 Note: Multiple response question The goods bought in South Africa can be grouped into five main categories: food and beverages, household/home goods, personal goods, electrical goods and miscellaneous. In the first category of food and beverages, the most common items were cooking oil (purchased by 22% on their last visit), eggs (20%), alcohol (20%), mealie meal (18%) and fresh fruit and vegetables (18%). The most popular household/home goods were household products (26%) and bedding material such as blankets and duvets (8%). Personal goods were dominated by new clothing and footwear (19%), while electronics and cellphones and phone accessories were the most popular electrical goods bought in South Africa. A third party benefitting from the presence of the Mozambican traders is the South African Treasury. Most of the prices that the traders pay for their goods include VAT, although VAT refunds can be claimed as these goods are not AHMR, Vol.3 No2, May-August 2017 – Special Issue 961 consumed within South Africa. VAT refunds ensure that the traders are not double taxed through paying VAT in South Africa and customs duty at the border. However, on their last trip to South Africa, only 55% of traders had claimed the VAT to which they were entitled. Of the 45% who did not claim VAT, nearly half said they did not know the procedure and 36% said that the procedure takes too long. In focus group discussions, it emerged that one of the reasons was that bus drivers did not want to spend time at the border while customs officials searched for goods. This makes it extremely difficult for the traders using public transport to submit VAT refund claims. Those using taxis or their own vehicles are more likely to claim VAT refunds. While it is advantageous for South Africa if people do not claim these refunds when they leave the country, it is fundamentally unfair to the traders. The Mozambican Government and civil society should certainly launch a campaign aimed at educating the cross-border traders on their rights and the procedures to claim VAT refunds. In addition, the Mozambican Government needs to work with its South African counterparts to clear the bureaucratic bottlenecks related to claiming VAT refunds. Fourth, ICBT traders spent money on transportation, accommodation and food when in South Africa. About 37% usually stay in paid accommodation, including rented rooms, hotels, guesthouses and B&Bs (Table 6). Those who do not pay for accommodation sleep at the bus or train station, on the street, in an automobile or stay with friends and family. Public transport is the most common way for traders to travel to and from Johannesburg, including buses (used by 43%), trucks (15%), taxis (11%) and trains (1%). Others use private transport including their own vehicles (13%), individually-rented vehicles (11%) and vehicles rented with others (6%), but still pay for petrol and other costs such as parking in South Africa. AHMR, Vol.3 No2, May-August 2017 – Special Issue 962 Table 6: Accommodation in South Africa No. % Paid accommodation Rent/rent shared room 36 15.2 Hotel (shared room) 34 14.3 Bed and breakfast/guest house/lodge (own room) 13 5.5 Hotel (own room) 3 1.3 Bed and breakfast/guest house/lodge (shared room) 2 0.8 Non-paying accommodation Bus or train station 51 21.5 Stay with friends and family 40 16.9 Street 35 14.8 In car or truck 14 5.9 Stay with partner 9 3.8 Total 237 100.0 In terms of the monetary spend in South Africa, the cross-border traders reported spending an average of ZAR 14,287 on goods, ZAR 1,714 on customs duties, ZAR 1,441 on transport, ZAR 218 on accommodation and ZAR 258 on other expenses on their last trip to South Africa. In total, a trader travelling from Maputo to Johannesburg thus spent approximately ZAR 17,900 per trip on business-related costs. This translates to nearly ZAR 5.4 million per trip for the entire sample, most of which directly benefits the South African economy (Table 7). The financial contribution of cross-border traders to the South African economy is clearly significant. AHMR, Vol.3 No2, May-August 2017 – Special Issue 963 Table 7: Contribution to Local and Regional Economy Buying goods (ZAR) Paying duty at the border (ZAR) Transport (ZAR) Accom- modatio n (ZAR) Addition al costs (ZAR) Total (ZAR) Amount per individual R14,287.21 R1,714.38 R1,441.21 R218.28 R258.49 R17,920 Total amount for sample R4,343,311 R521,171 438,127 R66,357 R78,580 R5,447,549 Note: 1 ZAR = 0.085 USD at the time of the survey Business Challenges In general, the ICBT traders do not have problems with their documentation and immigration status when in South Africa. The introduction of the visa exemption for Mozambicans certainly played a significant role in reducing undocumented migration from Mozambique. Most significantly, only 2% travel to South Africa with no official documentation. As such, the vast majority of the Mozambican ICBT traders enter the country using legal channels. The most important border or customs-related problem cited by the cross- border traders was corruption, experienced often or sometimes by 86% of the respondents (Table 8). Corruption leads to a reduction in government customs revenue collection and may also result in a reduction of the profit margins of the cross-border traders, which ultimately reduces prospects for business expansion. Others claimed that the duties that they pay at the border are too high (85% often/sometimes), while long queues, congestion and delays at the border are experienced often/sometimes by 82% of the cross-border traders. Clearly, the cross-border traders are dissatisfied with the time it takes them to clear customs at the border and there is evidence to suggest that paying bribes can help speed up the process. Some of the study participants noted that they AHMR, Vol.3 No2, May-August 2017 – Special Issue 964 use the magaigai or intermediaries with experience in dealing with customs officials in an attempt to avoid paying the high duties. Table 8: Problems Related to Customs and Immigration Yes (%) No (%) Too much corruption at border post 86.0 14.0 Duties paid are too high 85.3 14.7 Long queues/congestion/delays at border post 82.3 17.7 Restrictions on import/export of goods (type and volume) 66.7 33.3 Unwarranted confiscation/detention of goods 65.3 34.7 Verbal harassment by South African police, army, border officials 19.7 80.3 Verbal harassment by Mozambican police, army, border officials 19.3 80.7 Physical harassment/beating/violation of human rights by South African police, army, border officials 12.0 88.0 Physical harassment/beating/violation of human rights by Mozambican police, army, border officials 11.0 89.0 Physical harassment/beating by other people 8.3 91.7 Other respondents noted that they experience restrictions on the types and volumes of goods that they can either import or export (experienced often/sometimes by 67% of respondents). During the focus group discussions, one trader noted that “we can import spare vehicles, vegetables, groceries, fruit, meat, furniture, alcohol. The only limitation is related with duties and some do not know the Common Customs Tariff, which is a heavy book to be read.” The Common Customs Tariff is composed of 97 categories of products and sub products. The president of the Mukhero Association commented on the Common Customs Tariff as follows: Mukheristas get advice from the association about their rights and obligations and some were trained. All the time that the Government updates duties or other related issues the associations are informed. As a matter of fact, before any increase or changes on Common Customs Tariff, AHMR, Vol.3 No2, May-August 2017 – Special Issue 965 the Government calls the associations to inform them, otherwise we block the streets. When the Government of South Africa introduced a new regulation that every crosser should prove they had an amount of ZAR 3,000, we felt that was a decision that had to be discussed between the two governments. We were concerned about security and not every trader carries that amount of money. South African should understand that we bring money to them and we are no longer poor as it was in the past during civil war.8 Others cited the unwarranted confiscation/detention of goods at the border (56% often/sometimes). A focus group participant noted that they are sometimes forced to leave their goods at the border if they fail to reach a reasonable agreement with customs officials. Participants noted that it was not worthwhile trying to get goods back because of the high cost of import duty, especially on goods such as cigarettes and alcohol. The customs department is said to auction confiscated goods. Even though only a small number of respondents reported cases of harassment and abuse, these are significant enough to warrant attention. Verbal harassment by police, army and border officials had been experienced often/sometimes by the cross-border traders on both the South African (16%) and Mozambican (16%) sides of the border. Physical harassment/beating/violation of human rights by police, army and border officials had also been experienced often/sometimes on both the South African (11%) and Mozambican (10%) sides of the border. The president of the Mukhero Association described the border as a site of struggle between traders and customs officials: It is a titanic fight as both officers and traders are strong. Traders have their way of fighting and avoiding customs, while customs use their power as official authority, but we do have our own ways of counteracting them. However, this fight ended by the time the cross-border traders realised that the only way of this fight was to meet with the authorities, as the government did in Rome with RENAMO. Conversation is the only way to 8 Interview, 3 September 2014. AHMR, Vol.3 No2, May-August 2017 – Special Issue 966 avoid death of people, because the authorities were using fire weapons as well as some traders also had weapons.9 The challenges faced by the traders in their daily operations can be divided into two categories: those experienced while conducting business operations in Mozambique and those encountered when travelling to South Africa for business. In Mozambique, the most common problems related to competition from other traders (65.5%), competition from large retailers or supermarkets (45%), insecurity or problems securing a selling site (43%), conflicts with other traders (35%) and confiscation of merchandise (34%) (Table 9). Table 9: Challenges in Daily Operations of Business Yes (%) No (%) In Mozambique (N=400) Competition from other traders 65.5 34.5 Competition from large retailers or supermarkets 45.0 55.0 Insecurity of selling site/problems securing a selling site when needed 42.8 57.3 Difficulty negotiating with other traders/conflicts among traders 34.5 65.5 Confiscation of goods/difficulty getting confiscated goods back 33.8 66.3 Theft of money or goods 31.5 68.5 Harassment by authorities in Mozambique 18.0 82.0 In South Africa (N=300) Difficulty finding an affordable and safe place to stay 21.7 78.3 Insecurity of selling site/problems securing a selling site 18.0 82.0 Harassment by authorities in South Africa 12.7 87.3 Prejudice against my nationality 12.7 87.3 Harassment by South African traders 9.7 90.3 9 Ibid. AHMR, Vol.3 No2, May-August 2017 – Special Issue 967 Theft of goods is experienced by the traders on their way home from South Africa. This is related to renewed conflict in Mozambique where armed opposition party members are said to attack travellers in order to loot their cash and goods. One of the focus group discussion participants noted that: Bandits assault the Mukheristas taxis. They use mats with nails to punch the tyres. Then they steal their money. Some of Mukheristas do carry a lot of money, sometimes more than R 10, 000 in one trip. Last year one of our mates was on her way to Johannesburg when the taxi she was travelling in was ambushed near Machado. She lost more than R 40,000 cash in the robbery. There were 10 traders in that taxi.10 In South Africa, the biggest challenges relate to the difficulty of finding an affordable and safe place to stay in Johannesburg (22%), insecure trading sites (18%), prejudice against their nationality (13%), harassment by the police or municipal authorities (13%) and harassment by South African traders (10%). The study also sought to understand the cross-border traders’ experiences of xenophobia in South Africa. Nearly one in five of the respondents (19%) noted that their business had been affected by xenophobia a great deal or to some extent (Figure 6). However, as many as 63% said that xenophobia had not affected their business at all. While these findings are encouraging, given the widespread xenophobic attitudes and attacks on informal entrepreneurs, it is likely that Mozambicans are able to avoid the worst forms of victimisation by having a legal right to be in the country, by not remaining long and by not competing directly with South African informal businesses (Crush & Ramachandran, 2015). 10 Focus Group Participant, Zimpeto market, 5 December 2014. AHMR, Vol.3 No2, May-August 2017 – Special Issue 968 Figure 6: Extent to which xenophobia has affected business operations Finally, the respondents were asked about their treatment while conducting their business activities in both Mozambique and South Africa. Across all but one of the measures, the traders experienced more problems in Mozambique than in South Africa. They have had their business goods looted more often in Mozambique than in South Africa (47% versus 19%). They have been robbed more in Mozambique than in South Africa (39% versus 15%) and they have also been assaulted more (19% versus 12%). In addition, harassment by local authorities was more frequent in Mozambique (11%) than in South Africa (5%), as were incidents of unlawful arrest (6% and 1%, respectively). Conclusion Cross-border trading has become a way of life for many in Mozambique, geographically encompassing every part of the country and also involving migrants from other countries residing in Mozambique. Cross-border trade in Mozambique is primarily done by women with men mainly involved in the sale of the products brought back from South Africa. The traders are clearly playing a key role in supplying commodities that are in scarce supply in Mozambique. 0,0 10,0 20,0 30,0 40,0 50,0 60,0 70,0 A great deal To some extent Not very much Not at all P e r c e n ta g e AHMR, Vol.3 No2, May-August 2017 – Special Issue 969 Even though the sector is an important part of the Mozambican economy, little support is granted to the traders by local and municipal authorities or the private sector. Access to finance remains a major obstacle to business success as neither the government nor private banks provide loans to the traders. This paper has demonstrated the specific roles played by the cross-border traders in the economies of both Mozambique and South Africa. It has shown that cross-border traders contribute to the South African economy through buying goods, as well as paying for accommodation and transport costs. The cross-border traders are directly contributing to the retail, hospitality and transport sectors in South Africa, thereby creating and sustaining jobs in those sectors. In Mozambique, the traders pay import duty for the goods bought in South Africa and they play a significant role in reducing poverty and unemployment in the country. Therefore, the policy environment should encourage rather than discourage the operation of ICBT. A change in the attitude of government towards cross-border traders is required as they do contribute to poverty alleviation, and there is a definite need for a forum that involves government and municipal officials and the traders. Therefore, there is scope to include the informal traders in Mozambique’s poverty alleviation strategy. Although they are regarded as informal, they pay taxes to the local authorities for access to trading sites. They also buy goods in South Africa, some of which are sold to formal retailers, thereby blurring the formal-informal boundary. The informal tag becomes a hindrance when considering the functioning of the Mozambican economy. The traders need to be seen as an essential component of the Mozambican (and South African) economy because, in their absence, both would be poorer than they are today. Acknowledgements We would like to acknowledge the International Development Research Centre (IDRC) for its support of the SAMP Growing Informal Cities Project. References African Development Bank (ADB). 2016. African Economic Outlook 2016. Abidjan: ADB. AHMR, Vol.3 No2, May-August 2017 – Special Issue 970 Akinboade, O. 2005. A review of women, poverty, and informal trade issues in East and Southern Africa. International Social Science Journal, 57: 255-275. Amaral, P. and Quintin, E. 2006. A competitive model of the informal sector. Journal of Monetary Economics, 53: 1541-1553. Baxter, M. and Allwright, L. 2015. Opportunities to Improve Financial Inclusion in Mozambique: Building on Investments and Economic Activities Associated with the Extractives Sector. Report by OzMozis for FSDMoç (Financial Sector Deepening - Mozambique), Maputo: OzMozis Lda. Becker, K. 2004. The Informal Economy: Fact Finding Study. Stockholm: SIDA. Benjamin, N. and Mbaye, A. 2012. The Informal Sector in Francophone Africa: Firm Size, Productivity and Institutions. Washington DC: World Bank. Brooks, A. 2012. Riches from rags or persistent poverty? The working lives of secondhand clothing vendors in Maputo, Mozambique. Textile: The Journal of Cloth & Culture, 10: 222-237. Brooks, A. 2013. Stretching global production networks: The international second-hand clothing trade. Geoforum, 44: 10-22. Brooks, A. 2015. Clothing Poverty: The Hidden World of Fast Fashion and Second-Hand Clothes. London: Zed Books. Burra, N. 2005. Crusading for children in India’s informal economy. Economic and Political Weekly, 40: 5199-5208. Christie, F. and Hanlon, J. 2001. Mozambique and the Great Flood of 2000. London and Oxford: IAI and James Currey. Crush, J. and Ramachandran, S. 2015. Doing business with xenophobia. In: Crush, J. Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. de Oliveira, C., Pivoto, D., Spanhol, C. and Corte, V. 2015. Developments and competitiveness of Mozambican chicken meat industry. RAIMED - Revista de Administração IMED, 5: 205-216. AHMR, Vol.3 No2, May-August 2017 – Special Issue 971 de Vletter, F. and Polana, E. 2001. Female Itinerant Maize Traders in Southern Mozambique: A Study of a Higher-End Informal Sector Activity and its Potential for Poverty Reduction. ILO/SAMAT Discussion Paper No. 17. Geneva: International Labour Organization. Desai, M. 2009. Women cross-border traders: Rethinking global trade. Development, 52: 377-86. Dibben, P., Wood, G. and Williams, C. 2015. Towards and against formalization: regulation and change in informal work in Mozambique. International Labour Review, 154: 373-392. Dlela, D. 2006. Informal Cross Border Trade: The Case for Zimbabwe. Occasional Paper No. 52, Johannesburg: Institute for Global Dialogue. ECA (Economic Commission for Africa). 2010. Assessing Regional Integration in Africa IV: Enhancing Intra-African Trade. Addis Ababa: Economic Commission for Africa. Ericsson, A. and Brooks, A. 2014. African second-hand clothes: Mima-te and the development of sustainable fashion. In: Fletcher, K. and Tham, M. (Eds.). Routledge Handbook of Sustainability and Fashion. New York: Routledge. Finmark Trust, 2009. FinScope Mozambique Survey, 2009: Survey Report. Johannesburg: Finmark Trust. ILO (International Labour Organization). 2004. Child Labour and the Urban Informal Sector in Uganda. Kampala: ILO and Government of Uganda. ILO. 2009. Global Employment Trends – Update. Geneva: ILO. Jenkins, P. 2012. Home Space: Context Report. Copenhagen: Danish Research Council. Jenkins, P. 2013. Urbanization, Urbanism and Urbanity: Home Spaces and House Cultures in an African City. New York: Palgrave Macmillan. Jones, S. and Tarp, F. 2013. Jobs and Welfare in Mozambique. WIDER Working Paper No. 2013/045, Helsinki: UNU-WIDER. AHMR, Vol.3 No2, May-August 2017 – Special Issue 972 Kamete, A. and Lindell, I. 2010. The politics of ‘non-planning’ interventions in African cities: Unravelling the international and local dimensions in Harare and Maputo. Journal of Southern African Studies, 36: 889-912. Lesser, C. and Moisé-Leeman, E. 2009. Informal Cross-Border Trade and Trade Facilitation Reform in Sub-Saharan Africa. OECD Trade Policy Working Papers No 86, Paris, OECD. Macamo, J. 1999. Estimates of Unrecorded Cross-Border Trade between Mozambique and her Neighbors: Implications for Food Security. USAID Technical Paper No. 88, Washington, DC: U.S. Agency for International Development (USAID). Nucifora, A. and da Silva, L. 2011. Rapid growth and economic transformation in Mozambique 1993-2009. In: Chuhan-Pole, P. and Angwafo, M. (Eds.). Yes, Africa Can: Success Stories from a Dynamic Continent. Washington, DC: World Bank. Paulo, M., Rosário, C. and Tvedten, I. 2007. Xiculongo: Social Relations of Urban Poverty in Maputo, Mozambique. Chr. Michelsen Institute (CMI) Report No 13, Chr. Bergen: Chr. Michelsen Institute. Peberdy, S. 2000a. Border crossings: Small entrepreneurs and informal sector cross border trade between South Africa and Mozambique. Tjidschrift voor Economische en Sociale Geographie, 91: 361-378. Peberdy, S. 2000b. Mobile entrepreneurship: informal sector cross-border trade and street trade in South Africa. Development Southern Africa, 17: 201- 219. Peberdy, S. 2007. Monitoring Small Scale Cross Border Trade in Southern Africa. SAMP Report for Regional Trade Facilitation Programme. Cape Town: SAMP. Peberdy, S. and Crush, J. 1998. Trading Places: Cross Border Traders and the South African Informal Sector. SAMP Migration Policy Series No 6, Cape Town: SAMP. AHMR, Vol.3 No2, May-August 2017 – Special Issue 973 Peberdy, S. and Crush, J. 2001. Invisible trade, invisible travellers: The Maputo Corridor Spatial Development Initiative and informal cross border trading. South African Geographical Journal, 83: 115-123. Peberdy, S., Crush, J., Tevera, D., Campbell, E., Zindela, N., Raimundo, I., Green, T., Chikanda, A. and Tawodzera, G. 2015. Transnational entrepreneurship and informal cross border trade with South Africa. In: Crush, J., Chikanda, A. and Skinner, C. (Eds.). Mean Streets: Migration, Xenophobia and Informality in South Africa. Ottawa and Cape Town: IDRC and SAMP. Raimundo, I. 2010. Gender, Choice and Migration: Household Dynamics and Urbanisation in Mozambique. Saarbrucken, Germany: Verlag Dr. Muller Aktiengesellschaft & Co. KG. Raimundo, I., Crush, J. and Pendleton, W. 2014. The State of Food Insecurity in Maputo, Mozambique. AFSUN Urban Food Security Series No. 20, Cape Town: AFSUN. Rogerson, C. 2016. Policy responses to informality in urban Africa: The example of Maputo, Mozambique. GeoJournal, doi: 10.1007/s10708-016- 9735-x. Söderbaum, F. 2007. Blocking human potential: How formal policies block the informal economy in the Maputo Corridor. In: Guha-Khasnobis, B., Kanbur, R. and Ostrom, E. (Eds.). Linking the Formal and Informal Economy: Concepts and Policies. Oxford: Oxford University Press. Sparks, D. and Barnett, S. 2010. The informal sector in Sub-Saharan Africa: Out of the shadows to foster sustainable employment and equity? International Business & Economics Research Journal, 9: 1-11. Thorsen, D. 2012. Children Working in the Urban Informal Economy: Evidence from West and Central Africa. Briefing Paper No. 3, Dakar: UNICEF. Ulset, A. 2010. Formalization of Informal Marketplaces: A Case Study of the Xikhelene Market, Maputo, Mozambique. MA Thesis. Oslo: University of Oslo. War on Want. 2006. Forces for Change Informal Economy Organisations in Africa. London: War on Want. AHMR, Vol.3 No2, May-August 2017 – Special Issue 974 Wilson, D., Velis, C. and Cheeseman, C. 2006. Role of informal sector recycling in waste management in developing countries. Habitat International, 30: 797- 808. AHMR, Vol.3 No2, May-August 2017 – Special Issue 975