TRANSFORMATION 15(1991) ARTICLE CAPITAL FLIGHT UNDER APARTHEID* Zavareh Rustomjee Introduction The content of the 'mixed economy' is a central issue in the debate on post-apart- heid economic policy. The thrust of capital's input into this debate is to argue for an increased role for the market through deregulation and privatisation. In par- ticular, they threaten that if business 'confidence' is not maintained through accept- able state policies, foreign and domestic investment will not be forthcoming and indeed a net outflow of capital might result (eg Strauss, 1989). Some have argued that this has led to an increasing concern over business 'confidence' by sections of the Democratic Movement as reflected in an accommodative strategy in ANC discussion guidelines on future economic policy, particularly in regard to the question of ownership (financial Times, 01.10.90). While ANC economic policy is still at discussion stage, a cautious approach appears logical given the fact that capital flight has been a common experience of movements and parties elsewhere in the world that have followed progressive policies. Even developed market economies have not been immune to the devastat- ing adverse effect of flight on planned growth, and on the creation and distribution of national wealth (Harris, 1988). This paper contributes to the current debate, providing evidence that between 1970 and 1988, some $55 billion of capital has been systematically syphoned out of the South African economy through manipulation of the trading system. The magnitude of capital flight, in the face of what is generally regarded as a highly interventionist National Party government, reflects not only the power of private capital, but it also questions the bona fides of sections of the business and financial communities who were simultaneously arguing against sanctions and disinvestment in the 1970s and 1980s. More recently, while professing a commitment to South Africa and while attempt- ing to influence the debate on future economic policy, several of the conglomerates that dominate the economy have simultaneously restructured their operations, transforming themselves from South African multinationals into transnational corporations by placing portions of their assets beyond the reach of the future democratic state. This paper will show that the magnitude of flight provides an explanation as to how it is possible that domestic corporations can suddenly become subsidiaries of their own overseas branches as the recent example of De Beers has demonstrated. In section two of the paper various known mechanisms for capital flight are 90 TRANSFORMATION discussed. Section three focuses on flight through trade misinvoicing. Section four compares the magnitude of flight from South Africa with experiences elsewhere. An attempt to identify specific commodities that are misinvoiced is made in a detailed study of SA-USA trade in section five. Section six attempts to outline a research agenda which will more accurately identify conduits and culprits involved in capital flight and some policy implications are suggested in section seven. Capital Flight - Some Known Mechanisms Capital 'flight', according to Walter, appears to consist of a subset of international asset redeployments or portfolio adjustments - undertaken in response to a significant per- ceived deterioration in risk return profiles associated with assets located in a particular country - that occur in the presence of conflict between the objectives of asset holders and governments. It may or may not violate the law. It is always considered by the authorities to violate an implied social contract (Lessard and Williamson, 1987:105). Both individuals and corporations might engage in flight, which is often en- couraged by the existence of capital and exchange controls, a characteristic of the South African economy since the Sharpeville crisis of 1961 (Gidlow, 1976). Perhaps the crudest and best known mechanism is smuggling, whereby articles of high value are transferred outside national boundaries. More sophisticated methods exist via manipulation of the financial system, by using loopholes in existing legislation or by transgressing regulatory mechanisms. An example of this was the African Bank scandal which involved 'round tripping' of capital via the dual exchange rate system. Here rands were exchanged at a lower 'commercial' rate for foreign currency which was then exchanged again into rand at a higher 'financial' rate. While it was not purely an exercise in capital flight, it demonstrated the ease with which financial institutions can abuse the system (Ovendon and Cole, 1989:108-126). Unrecorded or inaccurately reported financial transactions which often reveal themselves in the Errors and Omissions item of the balance of payments are widely regarded as a measure of flight (see below). Capital flight can also take place through the trading system, and this mechanism is the subject of this paper. Suppose a corporation in South Africa were trading with a corporation in, say, the USA. A price would be agreed and the goods loaded onto a ship or aircraft, having passed through the necessary customs and excise for- malities. Documents pertaining to the transaction are channelled through the financial system via the respective corporations' bankers who, in South Africa's case are governed by Reserve Bank exchange control regulation. Payment for goods is usually carried out when the relevant financial authorities receive a Bill of Lading which confirms that goods of the specified value are on board the vessel. This segmented system therefore relies quite heavily on trust, on the integrity of all parties concerned and on the effectiveness of customs, excise and financial RUSTOMJEE 91 regulatory authority policing. It can be abused in a number of ways but, in all cases, abuse requires the collusion of transacting parties in respective countries. In cross-border trade, capital flight can take place if the agreed price for the commodities is higher or lower than the prevailing market price. An exporter might under-value goods, receiving less remuneration in the exporting country and retaining the balance in the country of destination. This can be done through agreement with an arms-length trading partner but is easier to effect by trading with a subsidiary. Effectively the exporter is syphoning capital out of the national economy. This is known as 'transfer pricing' and is a common practice of transna- tional corporations (TNC), who use this procedure to maximise profits of global operations by manipulating individual subsidiary accounts in, for example, high tax countries to reduce profits through over/under-pricing goods and services supplied by one or other of the TNC offices. Where this is done using single invoices with false values, it is very difficult to detect but there is no reason to suppose that this does not occur in South Africa. However, more than one invoice is often issued for a single transaction, one presented to the financial authorities for exchange control purposes, another to the customs authorities for purposes of establishing excise duty and perhaps a third to the relevant authorities of the trading partner. Such activities are made easier if the cross-border transaction is between subsidiaries of the same firm. This is often referred to as double invoicing and can be detected in a number of ways. Firstly, the authorities themselves can cross-reference foreign exchange transactions with trade transactions. Secondly, cross-referencing between customs and financial authorities of trading partners can uncover misinvoicing and transfer pricing. Finally, independent researchers can obtain a broad picture of the extent to which this mechanism is being used as a conduit for capital flight by comparing the official statistics of trading partners. The method used in this study to estimate flight does not take account of single invoice transfer pricing and estimates of flight are therefore conservative. Capital Flight through Misinvoicing TVade Trade statistics are normally compiled by the relevant authorities using the invoices declared by importers and exporters. Duties and tariffs are usually levied on invoiced values of imports. To estimate capital flight through misinvoicing, world trade as reported by South Africa is compared with trade as reported by her trading partners in IMF Direction of Trade Statistics Yearbooks. The differences between the two can be ascribed to disguised capital movement through misinvoic- ing; time lags; differences between stated destination of exports and the actual importing country; and inadequate specification of trade partners, eg 'Africa, not specified' and 'Special categories'. One would expect time lags to be ironed out over the 18 year period under consideration. Similarly, differences between stated and actual destinations would 92 TRANSFORMATION not affect the sum total of world imports from South Africa but merely its distribu- tion. Differences in specification of trade category and destination are the major obstacles to overcome and are dealt with below, leaving any resulting differences to fall under the category of capital flight. Given the consistency and direction of reported differences in trade statistics and the likelihood of flight from South Africa rattier man from its major trading partners, it is not unreasonable to assume that the statistics of South Africa's trading partners, mainly developed market economies, are a more accurate reflection of the value of trade. South Africa reports trade under three categories; 'Specified' and •Unspecified destinations and 'Special Categories'. Specified destinations refer to trade invoiced under the SITC or other appropriate standard to, or from, specified trading partners. Trade with countries that are 'Unspecified' (mainly African and Asian) is an^attempt to overcome sanctions by hiding the destination or source of trade. 'Special Categories' (IMF Statistics) or 'Unclassified Goods' (RSA Statistics) include 'military stores, petroleum products and specie (gold)... and balance of payment adjustments'.5 This category accounts for around 35% of South African trade, varying from year to year. Trade data reported by South Africa's trading partners are summarised in the world tables of IMF Direction of Trade Statistics (DOTS) and exclude trade falling under 'special categories'and'unspecified destinations1.6 While this study only compares specified trade, there is equal if not greater likelihood of capital flight in the latter categories, as has been evidenced by numerous scandals reported in the popular and financial press about secret armament deals, oil procurement fraud, etc (see for example Brzoska, nd). Until further study penetrates this shroud of secrecy, re- searchers have to work with what is statistically visible and estimates of capital flight that follow should therefore be considered to be conservative. The fact that South Africa's reported figures for exports and imports includes those of the South African Customs Union (Botswana, Swaziland, Lesotho and Namibia) have been accounted for by including in world aggregates any trade shown separately with these countries. The extent of misinvoicing imports and exports of trade between 1970 and 1988 is illustrated in Figure 1. For example, in 1988 while specified goods to a declared value of $13 677 million were exported, their actual value realised on sale in the destination country was $15166 million. Since only $13 677 million was exchanged into Rand in that year, the unrecorded capital outflow or flight amounted to the difference of $1489 million. Between 1970 and 1988, exports were underinvoiced by an average of 20% of South African exports after allowing for FOB/CIF differences. The average of 20% for South Africa is consistent with other work in this area. Gulati, quoted in Cuddington (1986a), provides estimated average (1977-1983) export underinvoic- ing as percentages of trade with industrial country partners of between 6.9% for Venezuela to 33.6% for Mexico. RUSTOMJEE 93 FIGURE 1 MISINVOICINQ IMPORTS AND EXPORTS 2000 1000 -J 3 -1000 -2000 •II |T| D rtft 1 -3000 -4000 r i • i • i • I • r ^ i i • r 1970 197S 1982 1986 1972 1976 1980 1984 1988 Source : Direction of Trade Statistics Yearbook, 1977, 1982, 1988, IMF, Washington. Simultaneously, imports were underinvoiced by an average of 4% of South African imports resulting in relatively smaller unrecorded capital inflows to South Africa. While this appears surprising, it conforms with observed patterns of other major studies of misinvoicing. Bhagwati, Krueger and Wibuswadi in Bhagwati (ed, 1974) find the underinvoicing of exports to be the major conduit of flight in most cases, often accompanied by lower magnitudes of underinvoicing imports, together resulting in net flight. The work of Gulati (1985) has yielded similar results. Lessard and Williamson (1987) have argued that likely motives for underinvoicing imports include the reduction of customs charges or the 'circumvention of import controls designed to limit the permissible value of imports'. Given the existence of exchange controls, South African importers are only likely to underinvoice imports to reduce tariffs or circumvent quantitative restrictions if they have other means of access to foreign exchange, perhaps through underinvoicing exports or through dividend repatriation. The extent to which the South African economy and foreign trade is dominated by a handful of conglomerates is a key to explaining this phenomena. It is likely that the competitive edge of corporations in a limited domestic market is enhanced by underinvoicing imports, the foreign exchange being obtained through other activities which might still result in net flight of capital. 94 TRANSFORMATION The Magnitude of Flight Capital flight is an unrecorded capital flow and is, therefore, not reflected in the statistics of official South African National Accounts.7 Recorded flows include Foreign Direct Investment to, and disinvestment from, the economy; Portfolio Investment and disinvestment which includes stocks, shares and bonds issued by private companies and public corporations like ESCOM and other parastatals; Long and Short-term foreign loans.8'9 Recorded statistics include components that might be regarded as capital 'flight', such as 'hot money' or short-term speculative outflow by the local non-bank private sector and certain long-term capital flows associated with the post-1985 refusal by international banks to 'roll over' existing loans. This study has adopted a conservative approach by associating flight only with un- recorded flows. Within the national accounting system, provision is made for Errors and Omis- sions in balancing the final accounts. However, the Errors and Omissions item is widely regarded as an indicator of capital flight via the financial system and has been incorporated in most empirical studies of flight - Cuddington (1986) and Lessard and Williamson (1987), for example. In this study capital flight that takes place via the trade system, namely the sum of misinvoiciiig imports and exports, has been added to Errors and Omissions to obtain a composite unrecorded capital flow which has been superimposed against official recorded flow on Figure 2. FIGURE 2 NET CAPITAL FLOWS : RECORDED AND UNRECORDED 3000 2 -6000 , . , . , . , . , . , 1970 1974 1978 1982 1986 Source : International Financial Statistics Yearbook, Various Years 1970-1988 & Direction of Trade Statistics Yearbook, 1977, 1982, 1988, IMF, Washington. (1) Positive * Capital inflow to S.A. Negative = Capital outflow from S.A. RUSTOMJEE 95 Major recorded outflows have taken place in periods of clear political crisis, namely 1970-1980 and 1985-1988 with net inflows before 1977 and between 1981 and 1984. The magnitude and consistency of unrecorded capital flow are startling and raise questions on the extent to which official statistics reflect reality. How do such magnitudes compare to other estimates of flight from South Africa and flight from other countries? Comparative Studies of Flight The magnitude of unrecorded outflows since 1970 at current prices is $37.1 billion ($55.3 billion in 1988 prices). Note that South Africa's foreign debt stood at $21.2 billion in 1988. These sums are considerably larger than other independent estimates of flight from South Africa largely because the definition of 'flight* varies widely and makes comparisons difficult. The Morgan Guarantee Trust Company (1986) method deduces flight by examining annual stock changes in foreign assets and liabilities. While they assume that current account figures for imports and exports are accurate, the methodology of this paper does not. Hence they estimate flight from South Africa at $13 billion (our estimate $18.8 billion) between 1976 and 1982 and $4 billion (our figure $9.8 billion) between 1983 and 1985. Their study is one of the few published estimates of flight from South Africa. Extensive work has been carried out elsewhere on capital flight, particularly in Latin America, although there are conflicting estimates on the magnitude and definition of flight from individual countries. Table la compares the 'errors and omissions' component of South African flight while lb compares South African flight through trade misinvoicing. Under the Errors and Omissions item on the balance of payments account, the flight to debt ratio and flight as a percentage of GDPratio for South Africa are within the range of Latin American results although slightly lower than the norm. Comparisons of flight through trade misinvoicing indicate a much higher flight to debt ratio for South Africa than the norm. This is due to the relatively low level of South Africa's debt exposure. Capital flight via the conduit of trade appears to be favoured by asset holders in South Africa, Peru and Chile as evidenced by the comparable levels of flight in Table lb. In the case of Chile, capital flight through trade contrasts with net inflows of capital under the 'errors and omissions' item for the same period 1974-1984. It suggests that asset holders not only chose to deploy their assets geographically, but also face choices in the mechanisms that they use. In this case, flight is favoured through the trading system rather than via the financial/banking system - of course the two are connected, depending on particular procedures and formalities followed by respective countries. Thus it is possible to understand how, for example, between 1981 and 1984 there could be positive net recorded inflows into South Africa (Figure 2) with simultaneous larger net un- recorded outflows. It confirms the observation of Khan and Ul Haque (1985) of the rationality 'for domestic residents (who)... borrow abroad, use the proceeds of such 96 TRANSFORMATION borrowings to finance domestic investment, and at the same time invest their domestic savings abroad'. Resident capital outflows coinciding with non-resident or resident inflows is a wholly rational action, for in times of crisis, the state as bearer of the monetary constraint bears the cost of foreign debt. Capital flight can, therefore, be regarded as an entirely rational decision for South African capital, particularly those sections of capital operating in an international environment. In summary therefore, the magnitude of capital flight from South Africa is consistent with flight from other countries given that there are differences in regulations governing trade and exchange controls and other differences in state, capital and class relations that might have determining effects on flight. SA-USATVade 1985 In 1985, the USA was the largest importer of South Africa's specified exports, importing some $1 512 million or 17.5% of total specified exports of $8 628 million. A detailed (yet preliminary) breakdown of SA-USA trade for 1985 by commodity type according to CCCN (formerly Brussels Tariff Nomenclature) classification indicates that underinvoicing exports to the USA amounts to $724 million, about 32% of reported US imports. Underinvoicing appears significant for trade in precious and semi-precious stones, including industrial and gem diamonds, pearls, silver and platinum group metals; chemicals and allied products; works of art; and mineral products. This emphasis on flight via primary commodity export is rein- forced by the dramatic rise in capital flight in the 1980s, coinciding with the South African economy reverting to dependence on primary commodity exports as observed by Black (1989). The extent of underinvoicing may be tempered slightly by the positive value of $435 million under 'unclassified' goods, that is goods falling under Sections 1-21 which have not been invoiced as such. With the possible exception of Works of Art, the goods traded fall squarely within the sectors of the South African economy dominated by the conglomerates. Diamonds are the monopoly preserve of De Beers; about 90% of platinum is produced by the duopoly of Anglo American controlled Rustenburg Platinum and SANLAM/Gencor control- led Impala Platinum; chemicals production is dominated by Anglo American subsidiary AECI, SANLAM controlled Sentrachem and the state controlled SASOL corporations; and most mineral production is controlled in varying proportions by the six major mining, houses of Anglo American, Rand Mines, Gencor, JCI, Anglovaal and GFSA. Research Implications Since South African trade is highly monopolised, and since export trade has been the main conduit for capital flight, it is reasonable to assume that the bulk of flight has taken place via the conglomerates that dominate the economy, particularly in the light of evidence from USA-SA trade. Much of the national wealth that has flown abroad can, therefore, be linked to the growing trend in off-shore operations RUSTOMJEE 97 of South African corporations (Kaplan, 1983a; 1983b). The recent move by De Beers and Minorco to insulate their international operations from South African state control is therefore a logical accompaniment of the process of capital flight.12 Urgent research tasks, therefore, include the identification of commodities and trading partners involved in misinvoicing trade and the mechanisms used. Detailed studies are required on individual trading country partner trade patterns using a similar methodology to that in the USA-SA trade case study. Within this, it is necessary to penetrate the complexity and secrecy surrounding South African capital flows and trade (particularly 'Unspecified' and 'Special' categories) which, while serving to circumvent sanctions, have contributed to concealment of flight. It may be possible by working backward from commodities identified in such studies to identify the mechanisms and actors involved in misinvoicing trade. This might also be facilitated by examining the structure and extent of South African off-shore operations in order to establish whether flight capital is concealed in balance sheets of off-shore subsidiaries and whether that national wealth is beyond the control of a post-apartheid state. The apparent anomaly of underinvoiced imports, implying a net inflow of capital particularly during the 1980s requires further research into South African import controls. In a general scenario of capital outflow, the only conceivable rationale for effective capital import through underinvoicing imports would be if, firstly, un- recorded (with respect to South African authorities) capital was available off-shore to finance the inflow and, secondly, that capital export was possible by some other means. If, as suggested above, there is linkage to conglomerate activity, research will be necessary on the ways in which conglomerates are controlled and on how such power might be more democratically exercised. This includes the areas of regulation through anti-monopoly legislation, promotion of competition strategies and the transfer of ownership. Capital flight has been financed by falling national reserves, growing public/state indebtedness, rising foreign liabilities and the measured fall in profitability of the South African economy. The perception of falling profitability has been vigorously embraced from various positions on the political spectrum, increasingly so from corporate capital. A recent survey of pension fund managers in the UK revealed a reluctance to invest in South Africa on grounds of profitability rather than on the issue of sanctions or political risk. However, the magnitude and consistency of flight suggests falling profitability to be as much a result of massive capital flight, averaging 7.0% of GDP over a period of 18 years, as its cause and therefore any arguments statistically based on measured profitability require careful treatment. Questions of risk and profitability are part of a broader question of the deter- minants of capital flight, which is not addressed by this paper and which requires research. This would include, for example, macro-economic policies on exchange gg TRANSFORMATION rates, capital controls and the extent of state regulation. It would also include the form that capital has assumed in the South African economy and the role of capital flight in corporate restructuring. Policy Implications There are several components of a policy framework to which the results of this study point. These include the direct control of trade, increasing resources to police trade and financial exchange activities, restructuring national accounting statistical methods, examining the structure of corporate capital with a view to weakening its power in the economy and increasing mass participation in the monitoring and preventing of capital flight. The USA-SA trade case suggests that underinvoicing of mineral exports is taking place. South Africa's dominant position as a world supplier of strategic minerals provides an opportunity to shift the terms of trade in favour of primary commodity producers through co-operative marketing agreements. Robins (1990) has outlined the benefits to a post-apartheid state, firstly, to assume control of all minerals marketing and, secondly, to conduct sales in conjunction with other mineral producers. The successful operation of De Beers Central Selling Organisation provides a clear example of possibilities in this regard and might further enhance regional co-operation strategies. The negative effects of massive capital flight may more than offset the possible negative aspects of such a centralised operation. Despite the existence of exchange and excise controls, capital flight continues. The routine monitoring and cross-checking of South Africa's international trade and financial transactions is clearly necessary. Ovendon and Cole (1989:124) report that a financial investigation of this nature was instituted at the highest levels of the South African Reserve Bank following the 1985 debt moratorium. The creation of a specialised body to strengthen customs and excise controls is appropriate. This could be an in-house operation or might involve the use of international professional organisations such as SGS. 17 Capital flight via manipulation of the trade system is enhanced by conditions of monopoly and oligopoly. After all, how can an agent that overinvoices imports compete with others that do not? How does an exporter pay for production costs if exports are consistently underinvoiced? The study indicates that capital controls alone have been insufficient to deter conglomerates from engaging in flight and a post-apartheid state may need to retain and even strengthen capital controls in the short term. Beyond this, the extent of conglomerate power will have to be addressed. Statistical monitoring at the national accounting level clearly requires restructur- ing to demystify statistics and to make trade more visible. One way of doing this might be to publish trade figures by category of goods as well as by corporation. The availability of such statistics might expose the extent of conglomerate control of trade and will facilitate popular participation in the monitoring process. The evidence of flight raises questions regarding the role of export orientation, RUSTOMJEE 99 which has been raised in the debate on the future growth path. It is clear from this study that without mechanisms to prevent flight via trade, such a growth path may only exacerbate the drain of capital from the South African economy. Finally, since flight has severe adverse effects on society, the participation of trade union and civic organisations in the monitoring process outlined above is essential. There is no apparent precedent for this in experiences elsewhere in the world and the respective organisations might have to make innovative inputs into this process. This should not be too difficult since trade union members in South Africa already have some experience in monitoring trade through their exposure of evasion of sanctions by, for example, stuffing local newspapers into the packing of relabelled goods.19 Conclusion There are, then, two pictures of net capital flow to and from the South African economy. The first is from official statistics which has been referred to by much of the current literature on the economy, debt and crisis of the South African state (see Gelb, 1988; Harris, 1987; Morris & Padayachee, 1988; Padayachee, 1988,1989; Hirsch, 1989; Pillay, 1979). The second, as revealed here, is the unofficial reality of consistent and large net capital outflows from the economy apparently taking place before the appearance of any clear crisis. The implications of this for the accuracy of official balance of payment statistics have prompted some to argue, on the same grounds, that the debt of some Latin American countries is largely fictitious. Further attention needs to be given to the way in which national account- ing statistics are reported. The evidence of the study paints a picture of an economy that has been systemati- cally 'milked' of capital since 1970. Cumulative capital flight between 1970 and 1988 has amounted to some $37.1 billion ($55.3 billion in 1988 prices),20 an average of 7.0% of GDP per annum and more than twice the size of the country's 1988 foreign debt. This study underlines the undoubted failure of the apartheid state to prevent capital flight, despite stringent exchange controls and poses a crucial challenge to any post-apartheid state. While Harris (1989) has pointed to the need for the post-apartheid state to 'confront (private capital) head on by... taking the initiative in defining confidence building measures', the evidence of flight since 1970 casts serious doubts on the intentions of sections of capital. Confidence building measures might not be enough. While there is evidence that the bulk of capital flight is linked to trade dominated by conglomerates and that corporate restructuring has been facilitated by capital flight, further detailed research is required. Factors that determine or influence capital flight have not been dealt with in this paper and research will be necessary to inform future policy. No doubt the argu- ments of free marketeers that attribute flight as a response to the highly interven- tionist policies of the apartheid state will be raised even though there was no 100 TRANSFORMATION abatement of flight during the early 1980s when National Party policies shifted towards increasing financial liberalisation and deregulation. Whether future policy be regulatory or strongly interventionist or otherwise, the economic and social damage caused by capital flight will have to be balanced by the costs and possible inefficiencies that might result from such policies. Notes * This analysis is based on calculations of capital flight, the details of which are available from the author (address enquiries to the editors). 1. De Beers are to transfer their head office from South Africa to Switzerland, from where international operations will be controlled {Financial Times (UK), 10.03.90). 2. See Kaplinsky (1979) for an account of how a TNC systematically underinvoiced Kenyan pineapple exports, fine-tuning the capital flight procedure to obtain the maximum tax and operating subsidies offered by the state. 3. There are exceptions, particularly where tariffs are used in order to protect local industries. In such cases local prices might form die basis for valuing the imported commodities. This issue is not explored in this paper. 4. In the IMF Direction of Trade Statistics (DOTS), these refer to trade that falls within DOTS code 199. 5. See RSA Foreign Trade Statistics, 1985. Petroleurnproducts only fell into this category after 1979 and accounts for the steep rise in imports of 'Special Categories' after 1980. 6. All comparisons are made on an inclusive CIF (carriage, insurance, freight) basis. Where FOB values (free on board - excluding CIF) are given in IMF^DOTS data, these are increased by the relevant FOB/CIF ratios as published in IMF International Financial Statistics. Exports and imports shown on the country pages of IMF statistics are as reported by South Africa, either FOB, CIF or a mixture. In the summary tables, imports are adjusted to CIF equivalents and exports adjusted to FOB equivalents. This explains why adding up the subcategories in the country table yields a different (£l0%) lue to DOTS total who publish an Annual Yearbook as well as monthly h denominated mainly in US dollars and local currencies. In addition, the IMF publish an annual Yearbook and a monthly entitled Balance Of Payment Statistics, denominated mainly in Special Drawing Rights (SDR). Differences, if any, between the IMF-IFS and IMF-BOP statistics are due largely to time lags in the reporting of data. This section of the study is based on IMF, International Financial Statistics because: a) Relevant SARB statistics have not been published since 1986. These relate mainly to capital flows and to foreign assets and liabilities: b) Most major studies of capital flight have used IFS data, facilitating comparative national studies that might follow from this paper; c) Although data with similar titles are reported differently by the SARB (For example, the SARB might use different criterion to define short-term capital outflow), the aggregate outflows of capital are consistent There is close correlation between IFS capital flows and figures deduced from SA Reserve Bank data. 8. Because they are relatively small, the categories 'Private Unrequited Transfers' (mainly wage transfers of migrant and expatriate labour) and 'Recorded Unrequited Transfers' (mainly non-repayable grants and aid) have been excluded. 9. Published statistics for financial flows cover the territories of South Africa and Namibia. For a useful detailed account of what exactly each component constitutes, see Clarke (1978). 10. Morgan Guarantee (1986) define flight as: the counterpart of recorded net direct investment flows plus increases in gross external debt less recorded outflows through current account deficits less Duild-up of foreign assets by the banking system and official monetary authorities. 11. For an example of the history and detailed workings of one of the largest conglomerates, Anglo American, see tones (1984). For a broader view of the conglomerate structure of capital in South Africa see Rustomjee (1990). Uttembourgnead office (The Times (UK), 09.05.90). ' ^ ^ 13. Here there is a parallel with the experience of some Latin American countries in the 1970s RUSTOMJEE 101 whereby private capital outflow was matched by growing international indebtedness of the national state. 14. Mohr (1989) quotes Andre Hamersma in the Financial Mail (02.12.88): We have become an unsuccessful economy. Capital leaves not s o much for political reasons but because people don't make money in SouUi Africa anymore. We musn't think that after 1991 or when we have reorganised debt, money will flow in. 15. See UK pressreports on the results of a survey by Pensions Investment Research Consultants in 16.1 have in mind for example, Natrass (1989), who has used recorded national statistics to illustrate cyclical but falling profitability in the South African manufacturing industry since 1948. Natrass uses this to criticise the empirical basis of the regulation school perspective. 17. Societe General de Surveillance, a private Swiss based consortium, provides a trade valuation (relative to world prices) and checking service for a percentage of value of goods checked. In the case of Kenya, Kaplinsky (1978) found that the costs of utilising SOS outweighed savings in capital flight 18. Murray (1980:141) has argued that the monopolisation of economies in the production sphere together with the monopolisation of trading in a global context of trends towards concentration and transnationalisation must not only be recognised but also quantified. He points out the inadequacy of categorising monopolised trade only in terms of commodities (SITC categories, etc). To understand what is really going on, categorisation by trader is required as well. As at1980, Brazil and Mexico appeared to be the only two countries in the world that were systematically monitoring this. 19. This example is quoted in an SA press interview with a South African company specialising in falsely relabelling goods to conceal their SA origin. 20. Using the consumer price index from IMF, IFS statistics. References Bhagwati, JN (ed) (1974) - Illegal Transactions In International Trade (New York: North-Holland Oxford/American Elsevier). Black, A (1989) - 'The Crisis in SA Manufacturing Growth: issues and alternatives' (Paper presented to the Lausanne Colloquium, Lausanne, Switzerland, 8-13 July). Brzoska, M (undated) - 'Shades of Grey - ten years of South African arms procurement in the shadow of the mandatory arms embargo' (Working paper N o l 3 , Institut fur Politische Wissenschaft, Centre for study of Wars, Armaments and Development, Universitat Hamburg). Clarke, S (1978) - 'Changing Patterns of International Investment in South Africa and the Disinvestment Campaign' (Foreign Investment in South Africa : Discussion Series 3, Anti-Apartheid Movement, London, September). Commonwealth Committee of Foreign Ministers on Southern Africa (1989) - The Sanctions Report (London :Penguin). Cuddington, J (1986) - 'Capital Flight: issues, estimates and explanations' {Princeton Studies in International Finance 58, December). Direction of Trade Statistics Yearbook 1988 (Washington: IMF). Franzsen. DG (1983) - 'Monetary Policy in South Africa, 1932-1982', in South African Journal of Economics, 51(1). Gelb, S (1988) - 'Making Sense of the Crisis', in Transformation, 5. Gidlow, RM (1976) - 'Exchange Control and The Blocked Rand Mechanism', in South African Journal of Economics, 44(1). Gulati, S (1985) - 'Capital Flight Through Faked Trade Invoices: 1977-1983, A Statistical Note' (Columbia University, New York). Harris, L (1987) - "The South African Economy at an Impasse' (Paper presented at meetings of the African Studies Association, University of Denver, Colorado, USA, 19-23 November). Harris, L (ed} (1988) - New Perspectives on the Financial System (Croom Helm. UK). Harris, L (1989) - "The Mixed Economy of a Democratic South Africa' (Paper presented to the Lausanne Colloquium, Lausanne, Switzerland, 8-13 July). Hirsch, A (1989) - "The Origins and Implications of South Africa's Continuing Financial Crisis', in Transformation, 9. Innes, D (1984) - Anglo American and the Rise of Modern South Africa (London: Heineman). International Financial Statistics, Various monthly issues (Washington: IMF). International Financial Statistics Yearbook 1988, (Washington: IMF). Kaplan, D (1983a) - 'South Africa's Changing Place in the World Economy', in South African Review Two (Johannesburg: Ravan). Kaplan, D (1983b) - 'The Intemationalisation of South African Capital : South African direct f°^*»t investment in the contemporary period' (Paper presented at the Southern A r n o a 102 TRANSFORMATION Studies Conference, University of Edinburgh, 30 May -1st June). Kaplan, D (1988) - 'Telecommunications in South Africa -an industry at the crossroads1 (Unpublished paper. University of Cape Town). Kaplinsky. R (1978) - 'Report on Foreign Exchange Leakages with Particular Reference to Transfer Pricing7 (Unpublished mimeo). Kaplinsky, R (1979) - •Export-Oriented Growth: a large international firm in a small developing country', in World DevelopmentH• Khan, MS and Nadeem Ul Hague (1985) - 'Foreign Borrowing and Capital Flight: a formal analysis . in IMF Staff Papers, 32(4). Lessard, D and John Williamson (eds) (1987) - Capital Flight and Third World Debt (Washington DC- Institute for International Economics). Mohr P (1989) - •External Constraints on Economic Activity and Economic Policy on Soufli MOnr> U ' y^^^p^pTe^tedtotrreLausanMClkiiIS^itl^ Morgan Guarantee Trust Company of New York (1986) - 'LDC Capital Flight1, in World Financial MiTtes/March. Morris, M and V Padayachee (1988) - 'State Reform Policy in South Africa , in Transformation, 6. Murray R (1980) - Transfer Price. Multinationals and the State" (unpublished manuscript). Nattrass, N (1989) - -Post-War Profitability in South Africa: a critique of regulation analysis in South Africa', in Transformation, 9. Ovendon, K and Tony Cole (1989) - Apartheid and International Finance : a program for change, (Harmondsworth: Penguin). Padayachee, V (1988) - 'Private International Banks, The Debt Crisis and the Apartheid State 1982-1985', in African Affairs, 87(348). Padayachee, V (1989) - South Africa's International Financial Relations..1970-1987: history, crisis and transformation' (Unpublished DPhil thesis. University of Natal. Durban). Pillay, V (1979) - 'Transnational Capital and the Growth of the South African Corporate State' Corporations in South Africa, London, 2-4 November). Quarterly Bulletin (various) Pretoria: South African Reserve Bank). Robins, (1990) Rustomiee. Z (1990) - 'Regulating South Africa's Corporate Oligopoly' (Unpublished MPhil dissertation. Institute of Development Studies, University of Sussex). Strauss. C (1989) - "The Role of the Private Financial Sector in a Future South Africa' (Paper presented to the Lausanne Colloquium. Lausanne. Switzerland, 8-13 July). TABLE 1a : CAPITAL FUGHT(1) • ERRORS AND OMISSIONS COMPARATIVE INTERNATIONAL ESTIMATES (US* BILLION) CUMULATIVE CUMULATIVE FOREIGN FLIGHT CUM. FLIGHT GDP DEBT TO FLIGHT 1974-1984 1974-1984 AT DEBT ASA AT 1984 AT 1984 END RATIO %OF PRICES PRICES 1984 % CUM.GDP (4) (5) ARGENTINA -20.8 850 45.8 -45 -2.4 CHILE 3.7 193 20.0 19 1.9 MEXICO -56.0 1657 97.3 -58 -3.4 PERU -2.6 190 13.2 -20 -1.4 SOUTH AFRICAO) -7.7 711 24.3 -32 -1.1 URUGUAY -1.7 60 3.3 -52 -2.8 VENEZUELA -15.5 654 34.5 -45 -2.4 Source of LA statistics: Cuddington, John T, Macroeconomic Determinants of Capital Right an econometric investigation, in Lessard and Williamson (1987). RUSTOMJEE 103 TABLE 1b : CAPITAL FLIGHT(2) - TRADE MISINVOICINQ COMPARATIVE INTERNATIONAL ESTIMATES (US* BILLION) CUMULATIVE CUMULATIVE FOREIGN FLIGHT CUM. FUGHT GDP DEBT TO FUGHT 1974-1984 1974-1984 AT DEBT AS A AT 1984 AT 1984 END RATIO % OF PRICES PRICES 1984 % CUM.GDP (4) (5) ARGENTINA -6.8 850 45.8 -14.8 -0.8 CHILE -11.2 193 20.0 -56.0 -5.8 MEXICO -18.9 1657 97.3 -19.4 -1.1 PERU -5.7 190 13.2 -43.2 -3.0 SOUTH AFRICAG) -28.4 711 24.3 -101.9 -4.0 URUGUAY -0.4 60 3.3 -12.1 -0.7 VENEZUELA 0.5 654 34.5 1.4 0.1 Source of LA statistics: Gulati, SK, A Note On Trade Misinvoicing, in Lessard and William- son (1987). (1) 'Right' is equivalent to the sum of errors & omissions, 1974-1984 in 1984 prices, to conform with Cuddington's definition. (2) 'Right' is equivalent to the sum of import and export under/over invoicing. (3) Source of S.A. debt: South African Reserve Bank, Quarterly Bulletin, June 1989. (4) Source : International Financial Statistics Yearbook 1989, IMF, Washington. (5)Positive = Capital inflow Negative = Capital outflow