by Robert Davies

PLEASE NOTE: Readers wanting to reproduce and reference
this article should contact SAMP


Developing a new policy on undocumented migration that is sensitive to causal factors, regional realities, and the recommendation of the 1994 UN Conference on Population and Development that governments of countries of origin and of countries of destination should work together to"make the option of staying in one's own country a viable one for all people" (UN, 1994), requires understanding some of the dynamics of trade and investment relations in Southern Africa. The inherited highly polarised Southern African regional economy, and the distortions resulting from apartheid aggression and destabilisation, are without doubt major causal factors underlying what must be recognised as an abnormal movement of people across the sub-continent. Subjecting current policies for transforming these relations to critical review, and identifying ways to make the restructuring of the regional economy more effective in terms of producing opportunities for employment and a livelihood to formerly displaced, or currently marginalised, communities must be an essential part of a new regionally focused approach to this issue.

This paper essentially seeks to provide background information to facilitate discussion and debate. It begins with an overview of key features of the Southern African regional economy in the past, looks at the way in which trade relations between South Africa and the rest of SADC have evolved in the period since April 1994, examines the emerging trade and investment policy framework, and concludes by trying to highlight key challenges that lie ahead.

1. Historical Context

Southern Africa is a region in much more than a mere geographic sense of that term. As Wallerstein and Vieira (1992) have argued, Southern Africa has been a "social construct" since the third quarter of the nineteenth century. During the period of colonial rule a distinct regional political economy was forged with identifiable structures, patterns of relations and institutions. South Africa and most of the remaining Southern African Development Community (SADC) member states were drawn in, albeit in very different ways into a regional economy characterised by quite high levels of integration. South Africa, Namibia, Botswana, Lesotho and Swaziland all became members both of a customs and monetary union. Zimbabwe, Zambia and Malawi were grouped in a federation, and significant trade, investment and migration patterns were established between most of the current SADC member states.

Participation in the regional economy came to be of major importance to the domestic economies of several countries. Access to sources of labour, markets and resources in neighbouring countries all contributed significantly to the accumulation process in South Africa. Lesotho, Mozambique, Malawi, Botswana and Swaziland all became dependent for a substantial part of their foreign exchange earnings on transfers from citizens working in South Africa. Traffic passing through Mozambique's ports provided that country with another major portion of its convertible currency earnings. Customs revenue allocated to Botswana, Lesotho, Namibia and Swaziland (the BLNS countries) under the Southern African Customs Union (SACU) agreement came to account for a sizable portion of these countries' total state revenue - over 50% in the most extreme case of Lesotho.

Patterns of participation in the Southern African regional economy were, however, characterised by great unevennesses and by acute imbalances and disparities. Essentially, the principal poles of accumulation came to be located in South Africa (and to a lesser extent in Zimbabwe) while the other territories became incorporated in subsidiary roles as labour reserves, markets for South African commodities, suppliers of certain services (such as transport) or providers of cheap and convenient resources (like water, electricity and some raw materials). South Africa's visible exports to the rest of the region exceeded imports by a factor of around 4:1. This was a product not only of the stronger productive base of the South African economy but also of protective tariffs and non-tariff barriers of various kinds, which kept goods produced in regional states out of the South African market. This visible trade gap was partly sustained by the provision of services - transport, migrant labour remittances and hydropower - all of which became important sources of foreign exchange earnings for several neighbouring countries.

This pattern of relations was significantly disrupted by the conflicts and destabilisation of the 1970s and 1980s. The apartheid regime's policies of destabilisation and aggression had a major impact on neighbouring countries in the Southern African region. The United Nations Economic Commission for Africa estimated that 1,5 million people died between 1980 and 1988 in Angola and Mozambique as a direct or indirect result of Pretoria-sponsored wars. While these two countries bore the brunt of this assault all of the then SADCC member states were affected to some degree and the UN ECA estimates that economic and military aggression cost the then nine SADCC member states the equivalent of US $ 62,45 billion (UN ECA, 1989).

In addition, established patterns of economic interaction in the regional economy were disrupted in ways that negatively impacted on member states of the then Southern African Development Coordination Conference (SADCC). In particular, South Africa reduced its involvement in two important relationships in which it had historically been involved as a buyer and thus as a provider of revenue to other countries - migrant labour and transport. In the case of the former, the number of nationals from SADCC member countries employed in the mines declined from over 220.000 in 1975 to around 165.000 by 1991 (Chamber of Mines, 1991). In transport, the period saw both a decline in South Africa's use of facilities in other regional states, and witnessed an attempt by South African Transport Services (SATS) to divert traffic from landlocked countries which had historically used the services of other regional states. South African traffic through Maputo was cut to around 15% of pre-independence levels; while regional traffic passing through South African ports increased to a maximum of one-and-a-half times the level of 1981/2 by 1984/5.

This combination of low intensity war, destabilisation and economic disruption, stagnation and dislocation must be recognised as fundamental in creating the human insecurity that underlies past and present patterns of migration in the region. Large numbers of people were displaced by wars in Angola and Mozambique and a refugee movement was spawned which rapidly spread across the sub-continent. There were over a million Mozambican refugees in other SADCC countries at the beginning of the 1990s, while a further 200 - 300.000 had secured some limited protection in the South African "homelands" of kaNgwane and Gazankulu. 300.000 Angolans had sought refuge in Zaire, 100.000 in Zambia and 40.000 in Namibia in the period before the signing of the 1991 Bicesse ceasefire accord (i.e. before the impact of the resumption in fighting after the September 1992 election) (Sogge, 1992, p 13). Many more have spread across the sub-continent since.

It is widely accepted that Southern Africa needs to be placed on a new development-orientated growth path. While military threats to the security of states in the region have receded, human insecurity continues to be a reality of life with effects felt across national borders. Even South Africa, which has the largest and most advanced economy of all the countries of Southern Africa, faces major problems of poverty, unemployment and deprivation - the heritage left to it by apartheid. Employment prospects and living standards in neighbouring countries devastated by apartheid policies of destabilisation are, of course, even worse and continuing socio-economic crises in such countries continue to be felt across the region, inter alia in the form of "new problems" like large scale clandestine migration mainly to South Africa and significant "unrecorded trade" in arms and drugs.

Development is a notoriously slippery concept. It can generally be recognised as referring to an improvement of the human condition in all its aspects - economic, social, educational, intellectual and security. The UNDP's Human Development Index represents an important attempt both to conceptualise and measure development, essentially by refining conventional GDP/capita measurements to take account of income distribution, trends in average levels of nutrition, literacy and education.

Development is then a broader concept than economic growth. Economic growth refers to an increase in the output of goods and services. While growth is a necessary condition for development in countries characterised by poverty and underdevelopment, it alone is not sufficient. It is quite possible for there to be growth without development, or for there to be great disparities between the rate of growth and its impact on development. As such, as UNDP puts it, "The issue is not only how much economic growth but what kind of growth" (UNDP, 1992, p 2).

Southern Africa clearly needs to be placed on a development orientated growth path if it is to address the socio-economic causes of future potential security problems and create the climate of peace and security essential for sustained economic growth.

Within a comparatively short space of time it needs to find programmes and policies that will significantly increase employment, raise living standards and address urgent problems of housing, education and health. Unless there are pronounced improvements in all these areas across the sub-continent, insecurity in the form of violence, large scale migration, drugs and arms smuggling must be expected to continue and possibly even to increase. This would, in turn, rebound negatively on the investment climate and hence the prospects for economic growth.

Southern Africa needs to achieve all of this, moreover, in a changed and changing global context. Developments at the level of the world economy have decisively changed the terms of trade against many of the raw materials on which the region has historically depended, and it is now abundantly clear that conventional import substitution industrialisation programmes will be unable to generate sufficient growth. All the countries of the region thus face the challenge of finding new ways to engage more effectively with the world economy - at a time when the "rules of the game" are being changed largely to meet the needs of countries of the industrialised North.

A number of studies conducted in recent years (ADB, 1993) have argued a convincing case that significant benefits could accrue to all the countries and peoples of the region from a programme aiming to restructure regional relations in a way which permits increasing, mutually beneficial, sectoral cooperation and a more integrated, and less unbalanced, regional economy.

2. Patterns of Trade since 1994

South Africa's transition to a democratic government in April 1994 did not, and indeed could not, in itself guarantee that the region would be placed on a new development orientated growth path. It has, however, had a number of immediate short term impacts. Firstly, it has led to the "normalisation" of its relations with the region. The country's former pariah status has been definitively ended and with it whatever impediments still remained to acceptance as a normal trading partner or member of regional organisations. South Africa's political transition also removed, secondly, the main political factor underlying the patterns of insecurity in the 1980s. The installation of a democratic government has created a new climate in which South Africa no longer confronts the region as an adversary and this, plus the global trend towards multi-party democracy, clearly helped facilitate "internal settlements" that ended the wars in Mozambique and Angola. Both of these factors created, thirdly, potentially favourable conditions for a mutually beneficial restructuring of regional economic relations.

As far as trade is concerned, the current situation can be identified as one in which "normalisation" can be identified as having had the most evident impact to date, but also in which emerging new policy frameworks look set to bring about changes in the medium term. The following table shows how trade between the Southern African Customs Union (mainly South Africa) and non-SACU SADC member countries has evolved since 1993.

  Jan - Dec
Jan - Dec
Jan - Dec
Exports to non-SACU SADC Countries (R bn) 5,4 6,1 9,7 5,5
% increase
12,9% 53,2% 13,4%
Imports from non-SACU SADC Countries (R bn) 1,0 1,4 1,5 0,8
% increase
40% 7,1 % 6,7%
Ratio Exports:Imports 5,4:1 4,4:1 6,5:1 6,9:1

Source: Republic of South Africa, 1993, 1994, 1996, Absa Bank, 1996.

The table shows that SACU exports to other SADC member countries increased significantly between 1994 and mid-1996 (the latest date for which statistics were available at the time of writing), and now stand at an annualised total of aroun R 11 billion (or around 10% of total exports, compared to around 6,8% in 1993).

Such figures, however, underestimate both the extent and significance of such exports for South Africa. Officially published South African trade statistics refer to the trade of the whole of the Southern African Customs Union thus obscuring the significance of trade between South Africa and its partners in the Customs Union and understating the full significance to the South African economy of trade with other African countries. A calculation by officials of the Department of Trade and Industry estimated that exports to other SACU countries amounted to more than R 15 billion in 1993, more than the country's exports to either Asia or America. Trade with all other African countries amounted 31,7% of total trade if SACU countries are included, and 12,8% if they were not (DTI, 1995). Nearly three quarters of exports to non-SACU African countries go to other SADC countries.

An examination of the composition of these exports shows, moreover, that they have a disproportionate significance for South African manufacturing industries. Figures for 1994, for example, show that the non SACU SADC group was the destination of over a third of total SACU exports of machinery and appliances, of more than a quarter of vehicle exports, and of 21% of exports of chemical products, 39,1% of plastics and rubber products, 16,9% of foodstuffs and beverages and 13,8% of textiles and clothing (Republic of South Africa, 1994). Trade with the region is thus of very great significance for precisely those sectors that current industrial policy seeks to prioritise in promoting an export led contribution to economic growth and employment.

What the table also shows, however, is that while exports of very great strategic significance for growth and jobs in South Africa have shot up, imports from neighbouring countries have grown much more modestly from a lower base. While exports have doubled, imports that were less than 20% of exports to start with increased by less than 60% between 1993 and 1996. The inevitable result is that the imbalance in visible trade has widened. The table thus shows that the ratio of exports to imports has increased from just over 5:1 in 1993 to nearly 7:1 in 1996. These averages, moreover, conceal significant variances between individual countries. In the most extreme case, that of Mozambique (which is of course the country of origin of the largest number of undocumented migrants deported each year), exports from SACU in January to June 1996 were R1,03 billion against imports of only R 48,6 million, giving a ratio of over 21:1 - compared to 16:1 in 1993 (Republic of South Africa, 1993, 1996). The inescapable conclusion from the data in the table shows, in short, is that "business as usual" has tended to reproduce and indeed exacerbate polarisation.

This outcome is clearly a product of a number of factors. The larger size and greater capacity of the South African economy compared to those of its neighbours; the weakened domestic productive and export capacity of neighbouring countries, partly as a result of conflict and destabilisation in the past, and the "opening up" of the markets of neighbouring countries to imports, partly as a result of liberalisation policies pursued under structural adjustment programmes, are all without doubt contributory factors. So too, it must be acknowledged, are tariff and non-tariff barriers restricting access to the South African market by products from neighbouring countries. Although the Uruguay Round of GATT, and a change in domestic policy priorities, has led to a significant reduction in South African tariff levels overall, several products in which neighbouring countries have export potential and are currently competitive would compete with those of sectors that are"sensitive" in South Africa. As an increasingly sharp public exchange during the course of 1996 made plain, products like clothing and textiles and several agricultural commodities are still subject to tariffs and other restrictions which neighbouring country producers find prohibitive.

3. The Policy Context

Government policy since 1994 has been premised on a view that polarisation of the sort reflected in the above quoted figures is neither desirable nor sustainable. Southern Africa is seen as a market of major strategic significance for South Africa, with growth potential in precisely those manufactured export commodoties which South Africa needs to prioritise. However, expanding such exports is acknowledged to be unsustainable unless the needs and demands of neighbouring countries in terms of access to the South African market are accommodated, and recognition is given to the need to work towards more equitable trading arrangements in the region.

Willingness to work towards a new regional trade arrangement was proclaimed at an early stage and underscored by two institutional decisions taken soon after the installation of the new government: the accession to SADC in August 1994, and agreement to begin to renegotiate the Southern African Customs Union agreement in November 1994.

A strategic choice that had to be made early on was whether to work towards a new trading relationship on a bilateral or multilateral basis. The new government found itself bombarded with proposals from several individual SADC countries to either renegotiate existing, or negotiate new, preferential bilateral trade agreements. At the same time the SADC secretariat, which had for some time been working on a regional trade protocol, tabled for discussion a draft protocol to establish a Free Trade Area in the SADC region shortly before South Africa's accession to the organisation.

South Africa's preference was to work towards a multi-lateral, regional agreement with some flexibility to accommodate bilateral specificities. Several reasons appeared to underlie this choice, including the desire to avoid duplication, an expressed commitment to reinforcing efforts to promote regional cooperation and integration, and a sense that the new rules created by the Uruguay Round of GATT could pose problems for a purely bilateral route. With regard to the last point, influential legal opinion in South Africa (Blumberg, 1994) suggested that provisions of the Marrakesh Agreement appeared to be much more receptive than previous GATT rules to claims by classes or groups of countries to have preferences granted to similar countries extended to them on the grounds that not to do so would constitute discrimination. Any substantially new bilateral agreements could, it was suggested, might thus find themselves either rejected or compulsorily extended to other countries in the region (i.e. multi-lateralised) through WTO intervention. Moreover, to secure a waiver for a bilateral agreement from the WTO under Article 25, South Africa would probably have to present itself as a developed country making concessions to developing countries. This completely contradicts the stance it is taking in negotiations with major trading blocs, where the advantages of it being seen as a developing country are apparent.

Having decided in favour of seriously working towards a viable multilateral framework, the next question that had to be confronted was what kind of multi-lateral agreement to work towards in the region. The draft protocol prepared by international consultants commissioned by the SADC secretariat envisaged what was seen as a rather mechanical timetable for tariff phase downs leading to the establishment of a Free Trade Area. According to the proposal, countries were to be divided into categories according to the existing levels of overall tariff protection. Those with higher incidences of overall protection were to phase down more rapidly than the others. This formula was seen as taking no account of the capacities of the individual countries, nor of concrete conditions in specific industries or sectors. The proposal was also vague as to whether the region should move towards literal free trade (i.e. remove duties on all intra-regional trade) or a WTO-legal Free Trade Area, which required only the removal of duties on "substantially all" trade - a formulation subject to broader interpretation, and certainly allowing for significant exclusions.

A "quick and dirty" study by staffers of the South African Industrial Development Corporation on the potential distribution of costs and benefits suggested that the introduction of literal free trade would exacerbate rather than reduce polarisation. The study calculated that the SACU countries (and mainly South Africa) would benefit to the extent of a 1,1% increase in GDP, a 4,6% increase in total exports and an 8% increase in manufactured exports from the removal of all tariffs within the SADC area. It concluded, however, that the resulting increased competition to domestic manufacturing industries from South African imports would have a net negative impact on GDP of 4 of the remaining 6 SADC countries (IDC, 1995).What appeared to follow from this study was a point in fact made in several policy studies undertaken before 1994, viz that further South African access to the markets of other SADC countries would need to be carefully structured and phased. On the other hand the IDC study identified a range of agricultural and industrial products (including textiles and clothing, footwear, furniture and ferrous products) produced in the non-SACU SADC member countries which would be competitive in the South African market at lower tariff rates (IDC, 1995). All of this suggested the need for an asymmetrical arrangement (which might nevertheless be couched as an FTA for WTO purposes) in which South Africa opened up its market to a greater extent than would be required by other countries, and which operated on a somewhat differentiated basis country by country.

South African negotiators suggested that instead of a specifying some mechanical formula for tariff phase downs, a SADC trade protocol should focus on defining a process of negotiating a new regional trade regime, taking account of some of the above points. They argued that a distinction needed to be made between the tariff regime prevailing within the region and that towards the rest of the world, and called on partners in the region to identify products for which they wanted better access into the South African market, rather than engaging in sterile debates over whose average tariff level was lower than whose. This was, however, seen by several other countries - especially those that had under the impact of structural adjustment programmes cut their tariffs the most - as mere prevarication. They saw themselves as having "progressed far in liberalising markets and opening up [their] economies", but felt that this has not been matched by the strongest economy in the region (Business Day, 12/10/1995).

After being bogged down in such debate for many months, agreement was finally reached on a SADC Trade Protocol during 1996. The protocol itself was signed at the SADC summit held in Maseru in August (SADC, 1996). It essentially provides for a process leading to the establishment of a Free Trade Area (FTA) in the SADC region within eight years. The precise definition of the FTA and of the tariff phase downs required to reach it is left to be determined in various "rounds" of negotiations - the first to begin within six months of the protocol. The protocol thus takes up the South African proposal that there should be various "rounds" of negotiations, similar in concept to the GATT rounds, but with a much shorter timetable - in which countries make offers and requests. By focusing on process, the protocol creates space for a significant element of asymmetry to be incorporated - both in the timing of the implementation of obligations and in the final outcome, provided that the SADC FTA broadly conforms with the WTO requirements that the process be completed within a maximum of ten years and that it results in the removal of customs duties on "substantially all" trade. The protocol also calls specifically for "the establishment of a linkage" between trade liberalisation measures and "other areas of sectoral cooperation" within the SADC programme.

Discussions preceding the adoption of the protocol, and subsequent statements by officials from various countries, have made it clear that a major feature of the negotiations envisaged by the protocol is likely to be demands by SADC partners for improved access to the South African market, and that a significant element of asymmetry is envisaged favouring the smaller and more vulnerable SADC economies. South African government officials have indicated that preparations for the first round of substantive negotiations are on track, and that they intend to approach the negotiations with a holistic perspective which focuses on the overall patterns and dynamics of regional trade rather than the demands of specific sectoral interests. Government has also said that it regards implementing the SADC trade protocol as top priority, and that any reciprocal concessions that may have to be made in extra-regional trade negotiations (such as with the EU) must be phased in in such a way that SADC countries enjoy greater preferential access to the South African market.

The SACU negotiations were also reported to have made significant progress in the second half of 1996, after appdaring to have been stalled for some time. Agreement on a simplified approach to the revenue sharing formula, which would essentially freeze South Africa's share of the pool at current levels, was reached in October leaving only the institutional arrangements for the new Customs Union outstanding.

4.The Challenges for Trade & Investment

After a fairly long delay, the essential frameworks for restructuring regional trade relations appear at last to be being put in place. The challenge now is to give real content to those frameworks.

South African government officials have repeatedly stated that they are committed to working to make the SADC trade protocol and renegotiated SACU agreement key mechanisms to bring about a more equitable, mutually beneficial and thus sustainable pattern of regional trade. A revised bilateral agreement with Zimbabwe, and possible new bilateral with Zambia, are also on the cards as short term measures to be incorporated into the SADC trade protocol within three years. All of these processes, it is recognised, will - if they are to be meaningful - have to have as their main content, in the short run at least, the accommodation of demands/requests by SADC and SACU partners for greater access to the South African market. Maintaining a holistic, and regional perspective, against inevitable special pleading by sectional interests in South Africa will, however, be a major challenge.

While promoting a more balanced pattern of visible formal trade will be an important part of creating a more integrated and more equitable regional economy the overall distribution of costs and benefits needs to be looked at in terms of the whole of the balance of payments account: including invisible earnings from service provision and the capital account reflecting investment.

As indicated earlier, the deficit many countries in the region had with South Africa in visible trade was, historically, at least partly financed from invisible earnings arising from service provision. The provision of water, hydro-power, transport, tourism and other non-traditional services to South Africa have all been identified as having potential to rapidly boost foreign exchange earnings for neighbouring states - and, in several cases, are precisely those sectors that were most seriously affected by destabilisation policies. SADC has programmes run by coordinating units or commissions in all relevant sectors. They have a mixed record. Boosting SADC sectoral cooperation programmes, and addressing the thorny question of institutional development and reform within SADC have been identified as priorities. Achieving these objectives will, however, pose many challenges.

As far as investment is concerned, while a number of projects have been initiated by South African companies in neighbouring countries, these have not broken the mould of limited involvement in traditional sectors - mining, hotels and tourism, forestry and building/engineering contracting. The major new initiative that seeks to move beyond this is, of course, the Spatial Development Initiative or "corridor" programme. This programme, which is one of the flagships of government's infrastructure development and industrialisation strategy, aims to promote decentralised industrial development through providing transport and other infrastructure in partnership arrangements between the state and the private sector. The first such project, significantly, is the cross-border Maputo corridor project. This involves directly road, rail, harbour and telecommunications projects with an estimated value of just over R 1 billion. However, the Department of Trade and Industry has announced that 135 "bankable projects" have been identified with an investment potential of US $ 5 billion. These, it is estimated, could create 100.000 new jobs (Business Report, 14/8/1996). While these figures refer to the whole programme, including that part being undertaken in South Africa, Transport Minister, Mac Maharaj, has said that his department believed the project could double the Gross Geographic Product of Southern Mozambique within five years (Mozambique News Agency, AIM Reports, 87, 22/2/1996). A feature of the projects being undertaken under this initiative is their financing. Most of the infrastructure projects are put out to tender on a Build, Operate, Transfer (BOT) basis under which private sector consortia tender to build and operate toll roads and other facilities, and public money accounts for less than 10% of the total. At the time of writing, a preferred consortium had been identified for one of the key projects - the upgrading of the N4 toll road.

5. Conclusions

The successful implementation of the above described measures could, in my view, begin to shape an environment conducive to the emergence of a more balanced pattern of trade and investment relations in the Southern African region.

However, from the point of view of an examination of their potential impact on factors underlying undocumented migration, it should be stressed that their main impact on employment in countries of origin of undocumented migrants to South Africa is likely to be felt in the medium and longer, rather than in the immediate short, term. There are, moreover, a number of countervailing trends at work in several countries in the region. SAP-driven downsizing of public sectors and formal sector industries are on the cards in several countries: in the case of Mozambique, the trade union federation estimated that such programmes had led to the loss of 90.000 jobs in nine years (Mozambique News Agency, AIM Reports, 84, 16/5/1996). Complex struggles for land between commercial interests and rural peasant communities are also reported to be complicating the resettlement of displaced people in Mozambique (The Guardian 11/6/1996, Mozambique News Agency, AIM Reports, 86, 14/6/1996).

This would suggest that, from the standpoint of a search for ways to assist countries of origin to make the option of remaining in one's own country a viable one, the macro-policy measures now being put in place need to be complemented by a number of other measures likely to have a more immediate impact.

One priority might be to reform the regulatory environment in such a way that it was more tolerant, and indeed encouraging, of informal cross border trade. A substantial unrecorded trade clearly exists in the Southern African region. While it certainly includes an undesirable element - drugs trafficking and arms smuggling - it also embraces many small scale producers of agricultural commodities, crafts and other goods. It is a trade which cannot be stopped, already provides a livelihood to significant numbers of people from neighbouring countries and potentially to many more. Indeed, a significant number of those deported each year as illegal immigrants probably came to South Africa to trade rather than to stay. Examining ways to make both immigration and customs regulations more facilitative of informal trade clearly needs urgent attention.

Beyond this, more and better focused bilateral and multi-lateral cooperation in sectors with high employment potential - such as agriculture, transport - also needs to be encouraged.