International Migration of Nurses and Human Resources for Health Policy: The Case of South Africa. Chizuko Sato

Working Paper Series Studies on Multicultural Societies No.11 International Migration of Nurses and Human Resources for Health Policy: The Case of South Africa Chizuko Sato Afrasian Research Centre, Ryukoku University Phase 2

Mission of the Afrasian Research Centre Today's globalised world has witnessed astonishing political and economic growth in the regions of Asia and Africa. Such progress has been accompanied, however, with a high frequency of various types of conflicts and disputes. The Afrasian Research Centre aims to build on the achievements of its predecessor, the Afrasian Centre for Peace and Development Studies (ACPDS), by applying its great tradition of research towards Asia with the goal of building a new foundation for interdisciplinary research into multicultural societies in the fields of Immigration Studies, International Relations and Communication Theory. In addition, we seek to clarify the processes through which conflicts are resolved, reconciliation is achieved and multicultural societies are established. Building on the expertise and networks that have been accumulated in Ryukoku University in the past (listed below), we will organise research projects to tackle new and emerging issues in the age of globalisation. We aim to disseminate the results of our research internationally, through academic publications and engagement in public discourse. A Tradition of Religious and Cultural Studies Expertise in Participatory Research/ Inter-Civic Relation Studies Expertise in Asian and Africa Studies Expertise in Communication and Education Studies New Approaches to the Understanding of Other Cultures in Japan Domestic and International Networks with Major Research Institutes

Afrasian Research Centre, Ryukoku University International Migration of Nurses and Human Resources for Health Policy: The Case of South Africa Chizuko Sato* Working Paper Series Studies on Multicultural Societies No.11 2012

International Migration of Nurses and Human Resources for Health Policy: The Case of South Africa Chizuko Sato * Introduction Since the mid-1990s, there has been a rapid increase in the international migration of physicians, nurses and other healthcare workers. Against this background, the movement of people across borders has become intensified in general due to the advancement of globalization, while the demand for medical and healthcare workers has increased due to declining birthrates and aging populations in developed countries. Although there is some international migration of medical and healthcare workers between developed countries such as from the United Kingdom (U.K.) to North America or Australia, or between EU member states, the majority of such worker movements are from Asian and African countries to Europe, North America and East Asia. Japan also began to accept Indonesian candidates for nurses and certified care workers in August 2008, following the enforcement of the Economic Partnership Agreement (EPA) between the two countries. Based on similar bilateral agreements, Filipino candidates for nurses and certified care workers came to Japan in the following year. Partly due to the fact that Japan is a receiving country for foreign nurses and care workers, discussions on the international migration of healthcare workers in Japan have focused on the issues facing a receiving society, such as how to establish good relationships between patients, elderly people, and their families on the one hand, and foreign nurses and care workers on the other, and what are the required approaches in multi-cultural workplaces (Carlos et al. 2008). As the declining birthrate and aging population are expected to further advance in the coming years, there will be an increasingly urgent need to tackle and resolve the problem of the expected shortage in nurses and healthcare workers. However, the international migration of medical and healthcare workers poses an equally, or even more, serious problem for Asian * Research Fellow, African Studies Group, Area Studies Center, Institute of Developing Economies (IDE-JETRO). This research was subsidized by the Grants-in Aid for Scientific Research of the Japan Society for the Promotion of Science 2008-2010 (Project No. 20710195), International Migration of Nurses and its Impact on the Health System of the Sending Countries Case Studies of the Philippines, South Africa, Ghana and Zimbabwe (Leader: Chizuko Sato). 1

and African countries, as this means the loss of their valuable human resources in the health sector. Taking South Africa as an example, this paper discusses the extent and some characteristics of the international migration of nurses in the late 1990s and early 2000s, and examines how it affected the provision of nurses in South Africa. According to research by the OECD (2007: 175), the number of South African nurses working in OECD member countries in around the year 2000 was the second largest after that of Nigerian nurses among African countries. Although South Africa has by far the largest economy of all African countries, the black majority government which was born following the abolition of apartheid in 1994 has adopted a policy of not issuing working permits to medical and healthcare workers from neighboring southern African countries in principle, so as to prevent brain drains from these countries to South Africa (SA-DOH 2006b; 2010). Therefore, South Africa basically does not accept nurses from other African countries and remains a sending country of nurses to the U.K. and other English-speaking countries. In the meantime, South Africa has a great demand for healthcare services, with an estimated 5.6 million people living with HIV in 2009 which is the largest in the world (UNAIDS 2011: 24). Although the scale of emigration of South African nurses has declined since the late 2000s, this paper argues that the experience of rapid increase in nurse emigration from the late 1990s to the mid-2000s left an imprint in the development of human resources for health policy in the country. Human resources in the healthcare sector primarily consist of physicians and nurses. In terms of manpower shortage, the scarcity of physicians is more serious in South Africa. Nevertheless, this paper focuses on nurses for the following two reasons. First, nurses are numerically the largest category among the healthcare workers, who are expected to be the backbone of the primary healthcare approach that the post-apartheid government has adopted as a strategy to expand public health services to rural and poor people (Chabikuli et al. 2005). Second, while the migration of physicians has been prevalent for several decades, the international migration of nurses is a relatively new phenomenon that has become conspicuous since the late 1990s (Kingma 2007). The recent rise in interest in the international migration of healthcare workers is due to a surge in the movement of nurses and other healthcare workers with medium-level technical expertise, who make up the majority of the workforce in the healthcare sector. This paper consists of four sections. In the first section, I will discuss theoretical issues concerning the international migration of nurses from the perspective of the sending countries. In the second section, the trends and characteristics of the international migration of South African nurses will be described, followed by discussion of the reasons for the rapid increase in their international migration in the late 1990s. In the third section, the conditions of the domestic supply of nurses in South Africa from the late 1990s to date will be examined in 2

order to determine whether the increase in the international migration of nurses has had any negative influence on the domestic supply of nurses. In the final section, I will discuss the contents of the human resources for health policy of the post-apartheid government and analyze the measures taken by the Department of Health against the increasing migration of nurses. Three Theoretical Approaches to the International Migration of Nurses from the Perspective of Sending Countries In analyzing the international migration of nurses from the viewpoint of sending countries, there are at least three theoretical approaches. The first approach emphasizes the fact that nurses are professional workers with specialized skills and expertise who should be distinguished from unskilled workers, and therefore views the international migration of nurses as a serious problem of brain drain from sending countries (SAMP 2001). Some researchers have even used the term brain hemorrhage instead of merely brain drain, due to the sheer scale of the expansion of the international migration of physicians, nurses and other healthcare workers in recent years (Interview, Galvez-Tan). One of the most critical aspects of the ongoing brain drain is that most training schools for physicians and nurses are operated publicly in many African countries. This means that the cost of the human resource development is covered by the developing countries, while the fruit of their education is enjoyed by developed countries (Dolvo 2007). This problem pertains to the economic cost of the international migration of nurses. The second approach sees an increase in the international migration of nurses from the viewpoint of the feminization of migration. This is because the nursing and healthcare professions have traditionally been taken up by women in most countries. Related to this approach is the theory of the global care chain, which has originally been applied to the international migration of domestic workers (Hochschild 2000). The global care chain refers to the chain of care work that is generated when women emigrate from developing countries as domestic workers and look after the children of their employers in receiving countries, while the children of the emigrating workers are taken care of by their grandmothers and other extended family members who remain in their home countries. While this concept was originally applied to the chain of care work for children, it can also refer to care work for elderly people, which is increasingly in demand as the declining birthrate and aging population proceed. Nurses and care workers take care of elderly people in foreign countries, while their own parents and grandparents have to be looked after by their extended families at home. Other problems have also been pointed out, such as the collapse of marital relationships and juvenile delinquency among children due to the long-term absence of wives and mothers. These problems are considered the social and psychological costs of the international migration of nurses. 3

The third approach perceives the international migration of nurses from the viewpoint of a dilemma in human security, by highlighting the fact that the nursing profession is directly linked to a society s quality of life, which is grounded in the medical and health care services. Broadly speaking, human security refers to the state of people who are free from fear and free from want, as defined by the U.N. Commission on Human Security (Commission on Human Security 2003). However in reality, definitions, concepts and interpretations of human security vary according to the user, partly due to there being two differing fields of discussion the security theory and the human development theory. Nonetheless, human security is a more advanced idea than the traditional concept of national security, because it places people, rather than nations, at the center of the discussion and prioritizes the protection of people from not only physical violence, but also from structural violence. While the international migration of nurses contributes to meeting healthcare and nursing needs in the receiving countries, it may also exert a negative influence on the same needs in the sending countries. This represents a dilemma in human security, which can be identified using the concept formulated in the human development theory. This is essentially the same problem as the brain drain described above; however, discussion here is focused more on the direct relationship between those brains and human welfare through the nursing and healthcare professions. There is another aspect of the dilemma in human security caused by the international migration of nurses. While the migration may improve the economic situation of the nurses (workers) themselves and their families, it may possibly have a negative influence on the healthcare needs of the general public of the sending country. As globalization advances, migration means expanded opportunities for migrating people. In a sense, movement across borders has become a basic human right that can be enjoyed by any citizen of any country. However, it may also mean deteriorated healthcare services for members of the sending country as a whole due to the loss of nurses. The above three approaches, which respectively focus on the migration of professional workers, female workers, and medical and healthcare workers, do not contradict or supersede each other. Nonetheless, this paper places emphasis on the third approach, i.e., dilemmas in human security. This is because the international migration of nurses has to be understood from a broader perspective covering the needs of both the sending and the receiving societies, rather than just looking at the needs of the migrating nurses themselves and their families. In this way, we can discuss the optimal policies and measures which can satisfy the needs of both the sending and receiving countries, as well as of both the migrating nurses and the general public of their home countries. 4

Trends and Characteristics in the International Migration of South African Nurses There are no statistical data on the overseas outflow of South African nurses. 1 The only available data that we can refer to are the number of requests for the issuance of nurse certificates. Emigrating nurses are commonly required to submit these certificates to the overseas medical institutions employing them, or to the employment agencies who refer them to the overseas employers. These nurse certificates are issued by the South African Nursing Council (SANC). 2 It must be noted that the number of requests for the issuance of nurse certificates only represents the number of nurses who hope or intend to work overseas, and thus does not necessarily reflect the number of South African nurses who have actually emigrated to other countries. Bearing this in mind, examination of the available data indicates that the number of requests for the issuance of nurse certificates, filed by South African nurses, increased rapidly since 1994. In other words, the international migration of South African nurses is a phenomenon that has occurred since the abolition of the apartheid regime. The annual number of requests remained at around 500 from 1991 to 1995, but it tripled to about 1,800 in 1998. It then increased dramatically in the late 1990s and the early 2000s (OECD 2004: 125). From 2001 to 2003, about 4,000 requests were filed annually. However, the number declined sharply to below 2,500 in 2004 and the declining trend continued afterwards. The breakdown of requests by destination country for the issuance of nurse certificates from 2001 to 2008 is shown in Table 1. Table 1: Verifications Issued by the South African Nursing Council 2001 2002 2003 2004 2005 2006 2008 2001-08 U.K. 2,567 2,336 2,790 1,746 910 218 109 10,676 Australia 430 461 467 347 330 470 664 3,169 U.S.A 267 420 360 163 157 83 50 1,500 Ireland 253 528 109 32 41 20 34 1,017 New Zealand 237 161 156 57 84 84 214 993 Canada 87 38 108 27 20 16 61 357 Middle East 41 10 70 10 4 1 0 136 Others 56 48 51 29 46 100 48 378 Total 3,938 4,002 4,111 2,411 1,592 992 1,180 Source: SANC website (last accessed: 26 September 2012). Note: For 2005, the data is available only for the period from January to November. The data for 2007 is not available. 1 The emigration and immigration statistics, published by the Statistics South Africa, are not useful for this purpose, because such statistics only identify people who leave the country with the intention of permanent emigration, and not all South African nurses working in other countries have such intention. 2 The SANC publishes statistics on the number of requests for the issuance of nurse certificates up to January 2009 on its website, but subsequent figures have not been disclosed. 5

The popularity of the U.K. as the principal destination for emigrating South African nurses is also confirmed by Clemens and Pettersson (2008: 23), which analyzed the census data of the nine major receiving countries (the U.K., the United States, France, Canada, Australia, Portugal, Spain, Belgium and South Africa) in around the year 2000, in order to identify the numbers of African-born physicians and nurses in these countries. Their research indicated that the total number of nurses who had been born in South Africa and worked in the eight receiving countries (i.e., excluding South Africa) stood at 5,105, of whom 56% (2,884 nurses) lived in the U.K., followed by those living in Australia, the U.S. and Canada. Apart from the nine countries examined by Clemens and Pettersson (2008), other destinations for South African nurses included Ireland, New Zealand, some Middle Eastern countries, in particular the United Arab Emirates, and some African countries such as Namibia and Kenya. The reasons for the emigration of South African nurses are usually explained using the push/pull concept. The push factors include the expectation for better wages in the receiving countries in both relative and absolute terms, the hope of escaping from their poor working conditions and environment, the lack of respect for the nursing profession in South Africa, and the hope of escaping from the poor security situation in South Africa. Because one out of every three or four people are estimated to be HIV positive in South Africa, the hope of working in a safe workplace free from the risk of HIV infection is also a strong push factor for the outflow of nurses. On the other hand, the pull factors include better remuneration and vigorous recruitment activities by developed countries that are facing shortages of human resources in the healthcare sector (Oosthuizen 2005). While it is important to shed light on the subjective views of migrating nurses and clarify both the push and pull factors that are driving their international migration, these factors alone cannot explain why the international migration of South African nurses has dramatically increased from the late 1990s to the early 2000s. Moreover they cannot explain why more than half of the emigrating South African nurses choose to go to the U.K. In this context, an interesting possibility is suggested through examination of the situation in the U.K., where more than half of the emigrating South African nurses worked in around the year 2000. Figure 1 shows the number of nurses from selected African countries, including South Africa, who newly registered with the Nursing and Midwifery Council of the U.K. (NMC-UK) from FY1998 to FY2007. This figure shows that the total number of South African nurses working in the U.K. is far larger than that of nurses from other African countries. It must be noted, however, that apart from South Africa and Nigeria, the overall number of nurses trained in African countries is limited. Therefore, in fact, the problem of outflow of nurses is more serious in other African countries than in South Africa, given the large percentages of emigrating nurses from their limited overall numbers. 6

Figure 1: Number of Newly Registered Nurses from Selected African Countries in the U.K. 2500 2000 1500 South Africa Nigeria 1000 Zimbabwe Ghana 500 Zambia 0 Source: NMC-UK (last accessed: 18 October 2011). Figure 1 also shows that the number of newly registered South African nurses in the U.K. zigzagged during this period, peaking in FY2001 and decreasing sharply after FY2003. The repeated increase and decrease in new registrations of South African nurses in the NMC-UK were caused by various regulations introduced from 1999 targeted at recruitment activities for foreign nurses by the National Health Service (NHS), as discussed later. The introduction of a series of regulations on the recruitment of foreign nurses was directly related to large-scale recruitment activities carried out by the NHS from the late 1990s to the early 2000s. This is clear from Figure 2, which shows the number of newly registered nurses in the U.K. and the number of foreign nurses among them from FY1998 to FY2007. Figure 2 indicates that increase in the registration of foreign nurses correlates with the increase in the overall number of newly registered nurses in the U.K. In FY1998, the proportion of foreign nurses among newly registered nurses in the U.K. stood at 20% (one out of five nurses), while a half of the newly registered nurses were foreigners in FY2001. Buchan (2007: 1323) points out that the increase in foreign nurses in the U.K. in the early 2000s was caused by the vigorous recruitment of nurses from overseas by the NHS, in accordance with the NHS expansion policy under the Labour Party government that came into power in 1997. 7

Figure 2: Number of Newly Registered Nurses and Foreign Nurses in the U.K. 40000 35000 30000 25000 20000 15000 10000 5000 Registered Nurses (Total) Foreign Nurses 0 Source: NMC-UK (last accessed: 18 October 2011). Note: EU/EEA nurses are excluded from foreign nurses. Foreign nurse means a nurse who was trained outside the U.K. Since the mid-2000s, the influx of foreign nurses to the U.K. decreased due to the increased production of British nurses and the introduction of international agreements to control the aggressive recruitment of foreign nurses by the NHS. In FY2006, the proportion of foreigners to the overall number of newly registered nurses declined to 17%. This indicates that the increased migration of South African nurses from the late 1990s to the early 2000s was essentially the result of large-scale recruitment activities by British public hospitals. The number of newly registered South African nurses in the U.K. has declined since FY2003, while during the same period the demand for general nurses in the U.K. has somewhat been satisfied, except in the case of nurses with special expertise. It may be said that the surge of incoming foreign nurses into the U.K. from the late 1990s to the early 2000s was a temporary phenomenon attributable to the medical and healthcare policies of the British government. 3 The repeated increase and decrease in the number of newly registered South African nurses in the U.K. from FY1998 to FY2005 was also a result of political intervention by the British government. The vigorous recruitment for foreign nurses by the NHS faced many criticisms 3 The U.K. is also losing its own nurses to other English-speaking countries such as the U.S., Australia, and New Zealand (OECD 2007). Even though the demand for foreign nurses is declining, the trend in the U.K. toward an aging society is set to continue. Consequently, it is highly likely that the demand for care workers, rather than nurses, will increase in the U.K. There are already cases of registered nurses from developing countries working as care workers at nursing homes for elderly people in receiving countries. Therefore, a decrease in the demand for nurses in the U.K. does not necessarily result in a reduced outflow of nurses from sending countries. 8

from African political leaders including Nelson Mandela, the first president in democratic South Africa, who condemned these activities as the poaching of valuable human resources from Africa. Consequently, the British Department of Health issued a guideline in 1999 whereby the NHS should not recruit nurses from South Africa or the Caribbean. In 2001, it further introduced a code of practice for the ethical recruitment of international healthcare workers by the NHS. This code of practice was expanded and strengthened in 2004 so as to cover not only the NHS but also recruitment and placement agencies and private companies and institutions that provide services to the NHS. The code of practice prohibited the NHS from recruiting healthcare workers from developing countries listed in the code, unless the government of the relevant country officially agreed to such activities. It also required the NHS to only work with recruitment and placement agencies that agreed to comply with the code (Buchan 2007: 1331). A similar code of practice concerning the ethics of the international recruitment of healthcare workers was adopted by the Commonwealth countries in 2003 (Commonwealth Secretariat 2003), and by the World Health Organization (WHO) in 2010 (WHO 2010). Although the code of practice was introduced with good intention, it was not legally binding and lacked penalty provisions. There were also loopholes, such the fact that the scope of the code did not include job hunting activities by individual workers from developing countries, and that the code did not cover non-nhs private hospitals and nursing institutions for elderly people, unless such institutions provided services to the NHS. Therefore, the outflow of nurses from African countries to the U.K. was not terminated completely. In fact, Buchan (2007: 1333) reveals that 3,000 nurses from the listed developing countries were newly registered in the NMC-UK in FY2004. However, it seems that the code had a short-term reduction effect, at least on South Africa, considering the fact that newly registered South African nurses decreased in FY2000 following the issuance of the guideline in 1999, and that the number declined again in FY2002 following the introduction of the code of practice in 2001 (see Figure 1). Moreover, the British government and the South African government signed a bilateral agreement in 2003, which focused on human resources exchange in the healthcare sector from the educational viewpoint. This bilateral agreement enables South African healthcare workers to work for the NHS for a specified period and subsequently return to public health institutions in South Africa with the acquired expertise. This agreement also encourages British healthcare workers to work at public hospitals in South Africa for a specified period (UK-DOH 2003). The South African Department of Health proudly states that this bilateral agreement succeeded in reducing the number of newly registered South African nurses in the NMC-UK by more than 55% from 2,114 in FY2001 to 933 in FY2004 (SA-DOH 2006c). The number of South African nurses registered in the NMC-UK continued to decline from FY2003, down to only 32 in FY2007. 9

Domestic Supply and Shortage of Nurses in South Africa African politicians have bitterly criticized the aggressive recruitment and placement of foreign nurses by British medical institutions. There have been repeated media reports linking the shortage of nurses in South Africa to the emigration of nurses (Shevel 2003; Munshi 2005). These criticisms and reports led to the introduction of bilateral and multilateral agreements concerning the recruitment and employment of medical and healthcare workers, as described in the previous section. However, the research by Clemens and Pettersson (2008) indicated that the proportion of South African nurses working overseas to all employed South African nurses was in fact merely around 5% in 2000. This brings us to question whether international migration is truly a root cause of the shortage of nurses in South Africa. This section attempts to examine this question by analyzing the domestic supply of nurses over the period during which the emigration of South African nurses increased. In South Africa, there are three categories of nursing qualifications: registered nurse (RN), enrolled nurse (EN), and enrolled nursing auxiliary (ENA). To become a registered nurse, one has to graduate from a four-year undergraduate course at a university or a nursing college. One has to take a two-year training course to become an enrolled nurse and one-year course to become an enrolled nursing auxiliary. An enrolled nurse can become registered nurse after taking a two-year bridging course. Figure 3 shows the number of nurses registered in the South African Nursing Council (SANC) from 1996 to 2011. The total number of nurses, combining all the three categories, remained at around 173,000 from 1996 to 2002, and then increased steadily from 2003, reaching 203,948 in 2007 and 238,196 in 2011. This means that the number of nurses in South Africa increased by approximately 138% from 1996 to 2011. Figure 3: Number of Nurses Registered in the South African Nursing Council 300000 250000 200000 150000 100000 50000 ENA EN RN 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: SANC (last accessed: 27 September 2012). Note: RN: Registered Nurse, EN: Enrolled Nurse, ENA: Enrolled Nursing Auxiliary. 10

The increase in the number of nurses registered in the SANC is also reflected in the increasing numbers of students joining nursing colleges and nurse training institutions. Figure 4 indicates the number of enrollments in the three categories of nursing courses from 1997 to 2011. It shows that students joining four-year courses to become a registered nurse have increased since 2001, following a slight decrease in the late 1990s. Enrollments in two-year courses to become an enrolled nurse have also increased since 2000, except in 2005 and 2006. The number of students entering one-year courses to become an enrolled nursing auxiliary increased from 1999 to 2004, but that figure subsequently took a downwards turn until 2010 when it jumped up again. The total number of students joining nursing schools and training institutions in all categories stood at 17,342 in 1997, which increased to 29,598 ten years later in 2007 and to 42,753 in 2011. In other words, the total number of enrollments in nursing courses increased 2.47-fold. On the other hand, data on graduates from nursing schools and training institutions in the same period indicate that there is a large gap between the numbers participating in four-year courses and the other two courses. Figure 4: Enrollments in Nursing Schools and Training Institutions in South Africa 25000 20000 15000 RN Course 10000 EN Course 5000 ENA Course 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: SANC (last accessed: 27 September 2012). Note: RN course (4 years), EN course (2 years), and ENA course (1 year). 11

Figure 5 shows that the total number of graduates from all three courses increased from 5,768 in 1997 to 13,236 in 2007, and to 15,589 in 2011. In particular, graduates from the enrolled nurse and enrolled nursing auxiliary courses have increased rapidly since 2002, as these courses can be completed in one or two years. In contrast, graduates from four-year undergraduate courses at universities and nursing colleges have not increased in proportion to the increase in enrollments. This indicates that the dropout rate stands high for four-year nursing courses. It is natural that the number of graduates from four-year courses declined between 2001 and 2005, since the number of enrollments also decreased in the corresponding five years (from 1997 to 2001). Still, the gap between enrollments and the number of graduates is large for four-year courses. While the average annual enrollments in four-year courses stood at 10,645 from 1997 to 2003, the annual average of graduates for the corresponding seven years (from 2001 to 2007) is as low as 1,837. This means that the completion rate of four-year courses during this period is only 17% on average. The average completion rate of four-year courses slightly increased to 20% for the period afterwards (2008-2011), when the annual average of graduates increased to 2,735. Failure to produce registered nurses through four-year courses at universities and nursing schools are partly compensated by bridging programmes taken up by enrolled nurses. More than 40% of registered nurses have been produced on average through bridging programmes each year since 1999. Figure 5: Graduates from Nursing Schools and Training Institutions in South Africa 8000 7000 6000 5000 4000 3000 2000 1000 RN Course EN Course ENA Course 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Source: SANC (last accessed: 27 September 2012). Note: RN course (4 years), EN course (2 years), and ENA course (1 year). 12

In addition to the high dropout rate of four-year courses, which inhibits improvement of the supply of nurses in South Africa, the fact that a large proportion of nurses who are registered in the SANC are not working is also a cause of the domestic shortage of nurses. There are no available data indicating a long-term trend concerning the actual number of nurses working for domestic medical institutions. However, Hall and Erasmus (2003: 537) indicated that 18.4% (nearly one out of five) of nurses registered in the SANC were not working and thus dormant in 2001. 4 A recent policy document published by the South African Department of Health in early 2012 also estimates that 18% of nurses who are registered with the SANC are not actively working in South Africa (SA-DOH 2012: 29). There is also the problem of the skewed distribution of nurses, caused by the dual structure of the South African healthcare sector. The healthcare sector in South Africa is separated into the public sector and the private sector. Figures 6 and 7 respectively indicate the number of registered nurses and the number of enrolled nurses who are working for public health institutions. The proportion of registered nurses working for public health institutions to all registered nurses registered in the SANC remained about 42% to 45% between 2000 and 2010. It slightly increased to 47% in 2011. The proportion of enrolled nurses working for public health institutions was relatively high at 63% on average in the early 2000s (2000-2003). However, as the number of enrolled nurses registered in the SANC increased in the late 2000s, the average proportion of enrolled nurses working for public health institutions declined to around 53% for 2005-2008 and to 48% for 2009-2011. Even though the supply of enrolled nurses increased in South Africa in the 2000s, there was no increase in enrolled nurses working for public health institutions. As mentioned earlier, not all nurses registered in the SANC are actually working in medical and health institutions. Just like in Japan, dormant nurses exist in South Africa in a certain proportion. Therefore, the actual proportion of registered nurses and enrolled nurses who work for public medical institutions, among all nurses working in South Africa, is estimated to be higher than is suggested from the above figures. The problem, however, is that South Africans who are covered by medical insurance and can thus receive treatment at private health institutions, only made up 16% of the total population in 2001 (McIntyre and Doherty 2004: 386). This means that only 40 to 60% of all the registered nurses and enrolled nurses employed in South Africa work for the public sector, even though 84% of all South Africans depend on public health institutions. 5 4 Note that the number of nurses registered in the SANC in 2001, quoted in Hall and Erasmus (2003), does not correspond to the data shown on the SANC website. 5 South African Department of Health has announced a plan to introduce the National Health Insurance scheme in order to provide health insurance for everybody, but its feasibility is still under discussion and at the moment 13

Figure 6: Proportion of Registered Nurses Employed by the Public Sector 140000 120000 100000 80000 SANC RN 60000 40000 20000 Public Sector 0 2000 2001 2002 2003 2005 2006 2007 2008 2009 2010 2011 Source: SANC and the Health Systems Trust (HST) (last accessed: 27 September 2012). Note: Data for 2004 is not available. Figure 7: Proportion of Enrolled Nurses Employed by the Public Sector 60000 50000 40000 30000 20000 SANC EN Public Sector 10000 0 2000 2001 2002 2003 2005 2006 2007 2008 2009 2010 2011 Source: SANC and HST (last accessed: 27 September 2012). Note: Data for 2004 is not available. it is not clear when and how it will be implemented. 14

A good number of nurses in the private sector actually work for NGOs, mining hospitals and pharmacy clinics which also take care of people without insurance. Therefore, the South African Department of Health cautions us when we discuss the population ratio for public and private sectors (SA-DOH 2012: 34). Nonetheless it is striking to see the shortage of nurses in the public sector which is reflected in the increased vacancy rate of nursing posts. In the early 1990s, nearly 80% of all nursing posts at public health institutions were filled. The vacancy rate of nursing posts increased to 25% in 2001 (Hall and Erasmus 2003: 537). This means that the shortage of nurses at public health institutions in South Africa became serious at the same time as the emigration of South African nurses increased. However, although the emigration of South African nurses took a downturn in the mid-2000s, the vacancy rate of registered nurse positions at public health institutions in South Africa continued to rise, reaching 31.5% in 2006, 36.3% in 2007, 40.3% in 2008 and 46.3% in 2010. 6 This means that close to half of all registered nurse positions are vacant at public health institutions in South Africa. In recent years, the shortage of nurses has not been limited to the public sector. Even private health institutions are facing a shortage of nurses, and the executive managers of some private hospitals and clinics in South Africa have started to import Indian nurses (Interview, Worral-Clare; Comins 2008). Moreover, the shortage of nurses in South Africa is expected to proceed further in the future. Figure 8 shows that the majority of South African registered nurses in 2011 were in their 40s (30%) or 50s (30%). Only 23% of all registered nurses were aged below 40 and only 4% were younger than 30. As for enrolled nurses, while 48% were younger than 40, the proportion of those aged below 30 was only 14%. In 10 to 20 years, when the generation who are currently in their 40s or 50s enter retirement age, the shortage of nurses, particularly registered nurses, will become even more severe. The profession of nursing used to be one of the few skilled jobs available to black women. Since apartheid was abolished and racial restrictions on professions were eliminated, there have been many job opportunities for young women. In this context, some South African researchers point out that the popularity of nursing jobs has declined among young women (Interview, Oosthuizen). On the other hand, the demand for nurses in South Africa is expected to increase continuously. Firstly, the population of South Africa is estimated to have increased to 50.59 million in 2011, from 44.81 million in 2001 (Statistics South Africa 2011: 3). Secondly, as the HIV infection rate in the nation remains high, nurses are required to provide treatment to patients as well as to conduct preventive and educational measures against further infection. Thirdly, as the country s primary healthcare policy advances, more and more clinics and hospitals are expected to be built in rural areas where hitherto only mobile clinics were available for healthcare services. This also contributes to an increase in the overall demand for nurses. 6 HST website, Vacancy Rate of Registered Nurse Posts in Public Sector (last accessed: 27 September 2012). 15

Figure 8: Age Distribution of Nurses (2011) a. Registered Nurses b. Enrolled Nurses 1% <29 30% 13% 3% 1% <29 4% 19% 30-39 40-49 50-59 60-69 30% >69 unknown 18% 27% 5% 1% 14% 34% 30-39 40-49 50-59 60-69 >69 unkno wn Source: SANC (last accessed: 27 September 2012). The skewed distribution of nurses in South Africa exists not only between the public sector and the private sector, but also between rural areas and urban areas. Table 2 shows that the population per nurse differs substantially between provinces. Compared to Gauteng and Western Cape provinces, where large cities such as Johannesburg, Pretoria, and Cape Town are located, the population per nurse was 2.8 times greater in Limpopo province and 2.4 times greater in Mpumalanga province in 1998. The gap has been reduced considerably in Limpopo province by 2011, but the gap still remained large in Mpumalanga and Eastern Cape provinces. Limpopo, Mpumalanga and Eastern Cape are provinces with many poor rural villages including former homeland areas. The shortage of human resources for health in these provinces remains much severer than in provinces with large cities. Table 2: Comparison of Population per Qualified Nurse between Provinces Provinces 1998 2003 2008 2011 Limpopo 458 414 274 244 Mpumalanga 387 417 344 294 Eastern Cape 291 323 304 280 Gauteng 165 195 189 186 Western Cape 162 190 195 184 Source: SANC (last accessed: 27 September 2012). 16

Human Resources for Health Policy in South Africa The above discussions have highlighted the fact that the challenges of the human resources for health policy for South Africa do not only concern the brain drain caused by international migration, but also involve many other aspects. The final section of this paper will discuss the details of South Africa s human resources for health policy and examine policy measures to increase the domestic supply and retention of human resources in the healthcare sector. The first comprehensive human resources for health policy since the end of apartheid was published by the Department of Health in 2006. It was titled the Strategic Framework for the National Health Human Resources Plan (hereafter Framework document ). In the background to the publication of this policy document, it is apparent that there were concerns about the outflow of human resources for health due to international migration. In the foreword to this Framework document, the South African Department of Health lists the problem of the outflow of human resources for health to richer countries as one of the major policy agendas that the government has to tackle, along with the movement of workers to private health institutions and the retirement of experienced staff (SA-DOH 2006a: iv). Before the Framework document was published, earlier policies to alleviate the shortage of human resources in the healthcare sector mainly focused on the retention of physicians. For example, even though the scarce skills allowance and the rural allowance introduced in 2003 were targeted at both physicians and nurses, the nurses who could actually receive the scarce skills allowance were limited to those specialized in cancer treatment and those working in operating rooms and ICUs. While physicians received a sum equivalent to 18 to 22% of their salaries as a rural allowance, the allowance paid to nurses was set at only 8 to 12% of their salaries, provoking many complaints from nurses. Complaints about the rural allowance also came from nurses who worked for large hospitals in urban areas, because they felt that their work at large hospitals, which were often the last resort for accepting and treating patients, was not duly appreciated (Interview, MaMkhize). Moreover, the South African government has received foreign physicians through bilateral agreements on the condition that they should work at hospitals in rural areas where the shortages are most severe. An agreement with the Cuban government in 1995 was aimed at accepting Cuban physicians in public hospitals in rural areas (Hemmett 2007). 7 Besides Cuba, South Africa has received physicians from Iran, Tunisia, Germany and U.K. through 7 In addition to the expatriation of Cuban physicians to South Africa, the bilateral agreement between South Africa and Cuba also includes a programme to train South African medical students in Cuba with a scholarship from the South African government. In this programme, South African medical students study for five years in a school of medicine in Cuba, before returning to South Africa to participate in internship and take the national qualification exam for medical practitioners. Once they pass the exam, they are obliged to undertake community service for one year and then work for a public health institution for at least five years (SA-DOH 2006d). 17

similar bilateral agreements (Interview, Groenewald; SA-DOH 2012: 41). Physicians from non-developing countries, particularly from Europe, Australia and New Zealand, are recruited and dispatched to remote hospitals through the Rural Health Initiative, a nonprofit organization established in 2007, which works with the South African Department of Health (Interview, Hudson). In mid-2011, about 10% of physicians registered with the Medical Council in South Africa are foreigners (3,004 people), including Nigerian (551), Congolese (Democratic Republic of Congo and Zaire: 463), British (265), Cuban (194) and Indian (124) (SA-DOH 2012: 41, Annex A Table 3). Since mid-1995 the South African government has as a matter of principle refrained from issuing working visas to healthcare workers from southern African countries that suffer from shortages of healthcare workers. This decision was taken in order to halt the flood of such applications as well as to avoid being criticized for depriving the Southern African Development Community (SADC) countries of their scarce resources. In late 1990s an agreement was reached among the health ministers of the SADC countries, whereby the member states should not deprive each other of human resources in the healthcare sector (SA-DOH 2006b). 8 In October 2001, the South African government announced that it would not recruit physicians and nurses from any of the G77 countries or Commonwealth countries, except in the case of acceptance based on special agreements such as government-to-government agreement (OECD 2004). The South African Department of Health states that this policy is aimed at discouraging the brain drain of healthcare workers from African countries to South Africa (DOH 2006a: 28). The fact that Nigerian and Congolese physicians occupy the largest and second largest number of foreign physicians respectively in South Africa in 2011 seems to betray this policy statement. However, certain foreign nationals such as those who are married to South Africans and those who are given refugee status are allowed to work in the healthcare sector, as long as they have suitable qualifications in order to register with their respective professional councils. In 2010, the South African Department of Health reiterated its principal policy stance of not recruiting foreign healthcare workers outside a government-to-government agreement (SA-DOH 2010). Because the South African government restricts the immigration of healthcare workers in principle, except in special cases based on bilateral agreements or the Rural Health Initiative, the country s policy for securing the supply of human resources in the healthcare sector should focus on the retention of domestic physicians and nurses. As mentioned earlier, the scarce skills allowance and the rural allowance did not provide sufficient incentives to nurses. 8 According to Clemens and Pettersson (2008), 500 nurses born in other African countries were working in South Africa in 2001. Most of them were Zimbabweans and Namibians. This figure made up only 0.3% of all nurses working in South Africa in the same year. However, the apartheid government actively recruited skilled/specialized workers from sub-saharan African countries in the late 1980s so that they could work in educational and health institutions in the independent homelands (Waller 2006: 4). It is not known how many of the nurses who came from other African countries during the above period still remain in South Africa. 18

Another retention policy introduced after 1994 is mandatory community service. 9 Community service was introduced for physicians in 1999, then expanded to various professions in the healthcare sector including dentists, pharmacists, and rehabilitation technicians, and finally introduced to nurses in 2007. According to this policy, nurses are required to work for health institutions in rural areas or in areas where the shortage of nurses is severe for one year, after completing their training courses. Unless they complete the required community service, physicians and nurses are not permitted to register with the Medical Council or the Nursing Council. This community service was introduced to nurses relatively recently, and thus it is still too early to determine how this system has been accepted by the nurses themselves. In the case of physicians, to whom the community service has been applied for more than 10 years, reactions seem to vary. The person in charge of human resources planning at the Department of Health states that many South African physicians consider the community service to be a kind of punishment and that it is extremely unpopular among them (Interview, Groenewald). On the other hand, the Dean of the School of Medicine, the University of Pretoria, says that some of their medical students began to see the roles of medicine in a new light through their interaction with rural residents, with whom they had never communicated until they began the community service (Interview, Mariba). The Framework document for human resources for health planning, formulated by the South African Department of Health in 2006, integrated the existing measures for the retention of healthcare workers into an overall policy framework and set numerical targets to increase the production of the healthcare workforce. The targets included doubling the annual number of graduates from schools of medicine from 1,200 to 2,400 by 2014, and increasing the annual number of registered nurse graduates from nearly 1,900 to 3,000 by 2011 (SA-DOH 2006a). Although the Framework document did not clarify how these increases would be realized, at least the latter target for registered nurse graduates was almost achieved thanks to the increased intake at university and nursing schools (see Figures 4 and 5 above). As for the training of physicians, the Hospital Association of South Africa, which represents private hospitals in the country, has shown a positive attitude toward opening private training institutions for this purpose. But the government has yet to accept such a proposal from private hospitals (Interviews Worral-Clare; Groenewald). 9 In order to alleviate the problem of the shortage of physicians, a new qualification called clinical associate was introduced, which can be acquired through a three-year undergraduate course and training. The first intake of such students was realized in 2006. However, the South African Medical Association is against the idea of regarding clinical associates as assistant physicians, given that their training period is only three years and thus shorter than that for registered nurses (Interview, Mariba). Partly due to the resistance from physicians, the training of clinical associates has not proceeded as originally planned. According to the HST website, 64 clinical associates are working in the public sector in 2012. This is the first year when the staff number data on clinical associates are available. 19