Dutch investments in South Africa – The Netherlands as
an investor or a tax haven?
Annemijn M. van Seventer
*ABSTRACT
Efforts to reduce tax avoidance and to increase the transparency of international financial flows are key features of the deliberations at the OECD and the G-20, which requires a multilateral approach. But before changes on a global level can be made, it is necessary to first take a bilateral approach. Therefore, I selected the Netherlands and South Africa to examine the bilateral-level impact of investment flows of special purpose entities (SPE) and the related foreign direct investment (FDI) patterns. This paper describes investment patterns of 554 Dutch companies investing in business entities in South Africa. I compared characteristics of ‘fully’ Dutch companies and multinationals making use of a Dutch financial holding (Dutch SPEs). The research results suggest that there are significant differences in characteristics of FDI patterns between ‘fully’ Dutch companies and Dutch SPEs. These differences pertain especially to regional and sectorial patterns of Dutch companies and the ownership structures of the South African companies. In addition, 92% of total FDI investments from the Netherlands into South Africa were investments from Dutch SPEs with an estimated value of US$ 22.3 billion in 2010. When the use of SPEs will be prohibited, FDI from the Netherlands into South Africa may decline and the Netherlands will cease to be one of the largest investors in South Africa.
KEYWORDS: Foreign Direct Investment, Bilateral Investment Treaties, Special Purpose Entities
Amsterdam, 10th January 2014
“Dutch sandwich grows as
Google shifts €8.8 billion to
Bermuda”
(Financial Times, 10 October 2013)
“How Apple Sidesteps Billions in Taxes”
(The New York Times, 28 April 2012)
“Starbucks pays corporation tax in UK
for first time in five years”
(The Guardian, 23 June 2013)
TABLE OF CONTENT
I. INTRODUCTION………….……… 5
II. LITERATURE REVIEW….………. 8
III. HISTORICAL DESCRIPTION OF FDI ………...11
3.1 FDI in South Africa….………...11
3.1.1 FDI in South Africa between 1980 and 2012……….. 13
3.2 Sectorial patterns of FDI……….………... 16
3.3 M&As versus Greenfield……….……….……. 17
3.4 Investment patterns of Dutch FDI activities in South Africa.………....19
3.5 Summary of FDI patterns in South Africa.………... 21
IV. DATA COLLECTION & METHODOLOGY……….……… 22
V. ANALYSIS……….………...25
5.1 Regional patterns…….………...25
5.1.1 Regional patterns of the investors….……….……. 26
5.1.2 Regional patterns of local business entities in South Africa... 26
5.2 Sectorial patterns……….……... 27
5.2.1 Type of Entity……….……….……...27
5.2.2 NACE Rev. 2……….……….……...28
5.2.3 Primary, secondary and services sector….……….29
5.3 Ownership structures of parent companies investing in South Africa... 29
VI. RESULTS & DISCUSSION……….……... 30
6.1 Regional patterns of Dutch SPEs….………. 31
6.2 BITs and FDI.……….………32
6.3 SPEs and FDI.……….………... 33
VII. CONCLUSION……….……….33
VIII. LIMITATIONS & FUTURE RESEARCH………... 34
LIST OF ABBREVIATIONS BIT Bilateral Investment Treaty
BRICS Brazil, Russia, India, China, and South Africa CBS Statistics Netherlands
DNB Dutch Central Bank FDI Foreign Direct Investment GASME Global Alliance of SMEs GDP Gross Domestic Product
IIA International Investment Agreement
IISD International Institute for Sustainable Development IMF International Monetary Fund
ISIC International Standard Industrial Classification M&A Merger and Acquisition
NACE Statistical Classification of Economic Activities in the European Communities
NL The Netherlands
OECD Organisation for Economic Co-operation and Development OFC Offshore Financial Centre
SA South Africa
SARB South African Reserve Bank
SEO Scientific Institute for Economic Research
SOMO Centre for Research on Multinational Corporations SPE Special Purpose Entity
TNC Transnational Corporation
TRALAC Trade Law Centre for East and Southern Africa
I. INTRODUCTION
An increasing number of transnational corporations (TNCs) divert foreign direct investment (FDI) through conduit countries with favourable tax treaty networks.1 One of the main reasons why TNCs locate foreign affiliates in conduit countries is to avoid host country taxes. In such constructions, the profits are migrated to countries with more favourable tax regulations.2 This is referred to as tax treaty shopping. Tax treaties prevent double taxation by allocating taxing rights between the host country, where the income arises, and the home country, where the beneficiary of the income resides (Weyzig, 2012:911).3 One of the most rapidly growing uses of offshore financial centres (OFCs) is engagement in financial activities in a more favorable tax environment through special purpose entities (SPEs), also referred to as special financial institutions (SFIs). The International Monetary Fund (IMF) defines OFCs as follows: “Countries or jurisdictions with financial centres that contain financial institutions that deal primarily with non-residents and/or in foreign currency on a scale out of proportion to the size of the host economy. Non-resident owned or non-resident controlled institutions play a significant role within the centre. The institutions in the centre may well gain from tax benefits not available to those outside the centre.” The financial institutions mentioned in this definition are considered SPEs. According to the United Nations Conference on Trade and Development (UNCTAD), SPEs, which are frequently established in OFCs, can be described as follows: “Foreign affiliates that are established for a specific purpose or that have a specific legal structure; they tend to be established in countries that provide specific tax benefits for SPEs. They may not conduct any economic activity of their own and have few employees and few non-financial assets.” Recently, the expansion of FDI through OFCs and SPEs is brought to the attention of the international community. They attempt to limit the facilitation of tax avoidance schemes and improve the transparency of international financial flows of these OFCs and SPEs (UNCTAD, 2013:15). To examine the bilateral-level impact of investment flows of SPEs and the related FDI patterns, I selected the Netherlands (NL) and South Africa (SA).
1 Conduit country is a term used by the Centre for Research on Multinational Corporations (SOMO)
also referred to as tax haven or OFC.
2 Although tax considerations are the main driver for the use of OFCs and SPEs, there are other
motivations for TNCs to locate their foreign affiliate in a particular country. OFCs and SPEs are used for: tax neutral solutions for partner countries with different tax regimes; legal neutrality for shareholders dispersed across different jurisdictions; and to help firms from countries with weak institutions to set up an international business more easily and to gain access to international capital markets and legal systems (UNCTAD, 2013:17).
3 Tax treaties enable TNCs to reduce withholding taxes on dividend, interest and royalty payments
As a former coloniser as well as a staunch supporter of the fight against apartheid, the Netherlands has a long history with South Africa. Consequently, the Netherlands is an important investor in business entities in South Africa. In 2010, the Netherlands was the second largest foreign investor in South Africa (UNCTAD, FDI/TNC database 2013).4 There are several explanations for this, some relating specifically to South Africa, others more generic. An example of an explanation of the latter kind is the general importance of capital flows for the Dutch economy. The Netherlands is the country with the largest inward and outward FDI in the world in 2013, which is mainly due to the inclusion of SPEs in FDI stock data (DNB, 2013:6). In other words, firms located in the Netherlands invest extensively in other countries (not only in South Africa), and foreign firms invest extensively in the Netherlands. Many of these firms are SPEs. For example, 91 out of the 100 largest TNCs have at least one foreign affiliate in the Netherlands (Eikelenboom and De Groot, 2013; UNCTAD, 2005:18). One of the reasons why the Netherlands is regarded as an attractive country for SPEs is the large number of bilateral investment treaties (BITs) with other countries, especially with developing countries.5 According the Trade Law Centre of East and Southern Africa (TRALAC), BITs are “binding international agreements between two countries under which each country undertakes certain obligations with respect to investments made by nationals of the other country within its territory.” The past year witnessed numerous expressions of opposition to on-going BITs (UNCTAD, 2013:105). For example, the government of South Africa decided to terminate its BITs with the Belgo-Luxembourg Economic Union6, Germany, Spain, and the Netherlands (UNCTAD, 2013:108). Our understanding of the advantages and disadvantages of SPEs and BITs is restricted to the macro-economic level, as there is little information available about the impact of SPEs and BITs on FDI on the bilateral level. In this thesis I will try to fill this gap by answering the following main research question: What are the characteristics of ‘fully’ Dutch companies compared to multinationals making use of a Dutch financial holding when investing in business entities in South Africa? The main objective of this study is to describe FDI patterns in South Africa and the role of Dutch investors. Additionally, this study gauges how
4 The UK was the largest investor in South Africa with a share of 50% of total inward FDI stock in
2010.
5 The Netherlands has 96 BITs in 2013 (UNCTAD, 2013:232).
6 The Belgo-Luxembourg Economic Union is an economic and monetary union between Belgium and
investment patterns in South Africa would change when the use of (Dutch) SPEs would be prohibited, as recently attempted by the Organisation for Economic Co-operation and Development (OECD) and the G-20, and BITs would be terminated. The research results suggest that there are significant differences in characteristics of FDI patterns between fully Dutch companies and Dutch SPEs. These differences pertain especially to regional and sectorial patterns of Dutch companies and the ownership structures of the South African companies.
This research project has implications for academics, managers and investors, and regulatory authorities. It contributes to the literature by presenting a detailed description of FDI patterns between the Netherlands and South Africa. Furthermore, treaty shopping and international investment agreements are under increasing pressure from the international community (UNCTAD, 2013:15). Once efforts to reduce tax avoidance and to increase the transparency of international financial flows lead to the ‘actual’ plan of action, managers and investors will have to change their modus operandi in several respects. For example, management of TNCs need to adjust their tax structures, and investors need to reconsider the location of their foreign affiliates. Moreover, aforementioned efforts require a multilateral approach (UNCTAD, 2013:15). But before changes on a global level can be made, it is necessary to first take a bilateral approach. For example, when a government of a conduit country decides to change their tax treaty network, TNCs will shift their operations to other OFCs. Therefore, an adequate understanding of FDI patterns on the bilateral level must precede the formulation of a multilateral strategy. Further, the results of this study show that fully Dutch companies and Dutch SPEs have different FDI patterns. Consequently, this study helps regulatory authorities become aware of in which sectors FDI activities of SPEs take place, where these activities are located, and which ownership structures are mainly used. Multilateral organisations might take these patterns into account when formulating international tax policies. Moreover, the estimated value of SPEs in South Africa is calculated, which clarifies the actual impact of Dutch SPEs on total inward FDI stock in South Africa.
crosstabs to describe FDI patterns in South Africa by fully Dutch companies compared to multinationals making use of a Dutch financial holding. In section VI, the results are discussed in the light of the literature. The final section formulates the conclusion, mentions some limitations of this study and makes suggestions for future research.
II. LITERATURE REVIEW
After 2008, mainly in response to increasing pressure on public finances as a result of the global financial crisis, the international community renewed and strengthened its efforts to reduce tax avoidance and increase the transparency of international financial flows (UNCTAD, 2013:15). The improvement of tax transparency and the promotion of information exchange are key features of the deliberations at the OECD and the G-20. Although most international efforts to combat these issues have focused on OFCs, investment flows through SPEs are becoming more important (UNCTAD, 2013:17). Addressing the increasing challenges of tax avoidance and tax evasion requires refocusing international efforts. The main difference between tax avoidance and evasion is that tax avoidance, unlike tax evasion, is a legitimate way to avoid taxes. According to UNCTAD (2013:18), the first step in dealing with these concerns is to establish a closed list of “acceptable” or “benign” uses of OFCs and SPEs. According to the Internationalisation Monitor (CBS, 2012:95), further research is necessary to divide FDI into two parts: the amount that is actually invested in a particular country and the part that passes through the country without having any durable economic impact. This would enrich the currently available numbers, as it would uncover interdependency between countries. Consequently, a descriptive study on bilateral-level FDI could contribute to existing studies and investment reports of regulative authorities. Overall, there is a lack of research into treaty shopping that is based on data of a single home and host country (Weyzig, 2012:911). This being the case, I selected the Netherlands and South Africa to examine the bilateral-level impact of investment flows of SPEs and the related FDI patterns.
(UNCTAD, 2013:101).7 Opposition to on-going IIA negotiations around the world became particularly vocal in 2013 (UNCTAD, 2013:105). Especially BITs were being criticised, largely because more than 1,300 of the 2,857 BITs that were in place could be terminated by both parties at any time before the end of 2013 (UNCTAD, 2013:109). The government of South Africa, for example, decided to terminate or renegotiate several of its BITs, as they believed that BITs failed to stimulate inward FDI in South Africa (Wöcke et al., 2013:14).8 From the perspective of the South African government, the economy has sufficient attractiveness for inward FDI (A. van Oudheusden). In September 2012 South Africa informed the Belgo-Luxembourg Economic Union through a notice of termination that it would not continue with the existing BIT, which was set to expire in March 2013 (UNCTAD, 2013:108). In October 2013, South Africa also terminated its BIT with Germany, which is the fourth biggest source of FDI in South Africa (TRALAC, 2013). In November 2013, South Africa terminated its BIT with the Netherlands as well. Furthermore, the South African government expressed their intention to revoke its BITs with other European partners, too. And the doubts about the beneficial effects of BITs are by no means limited to the international policy community.
The global network of more than 2,800 BITs has been built to promote FDI. Nevertheless, after a decade of research it is far from clear that BITs do in fact lead to an increase of FDI. Hallward-Driemeier (2003:22) analyses twenty years of bilateral FDI flows from 20 OECD countries to 31 developing countries and finds little evidence that BITs have stimulated additional investments. This study also implies that BITs should be considered complements to rather than substitutes for high-quality institutions and local property rights (Hallward-Driemeier, 2003:23). Hallward-Driemeier’s findings were confirmed by a study by Gallagher and Birch (2006). They examined FDI inflows from the US into 24 host countries in Latin America. The empirical results indicated that signing a BIT with the US does not have an independent effect on FDI from the US (Gallagher et al., 2006:972). Gallagher et al. (2006),
7 Other IIAs refer to “economic agreements, other than BITs, that include investment-related
provisions (for example, framework agreements on economic cooperation), investment chapters in economic partnership agreements and free trade agreements” (UNCTAD, 2013:118).
8 The fact that the major sources of FDI (UK, Germany, the Netherlands, and Switzerland) all have
though, did discover a correlation between the total numbers of BITs signed and the amount of FDI flowing to Latin America. At the same time, the authors stated that BITs may help, but cannot stimulate FDI in the absence of a stable and growing economy (IISD, 2009). Tobin and Rose-Ackerman (2003) studied FDI inflows from the US as well, but instead of investigating Latin America they focused on a set of 48 developing countries. The results show a weak relationship between BITs and FDI. However, BITs are not always ineffectual, as the authors found evidence that BITs attract greater amounts of FDI when a country has a relatively high level of political risk (Tobin et al., 2003:31).
Several other studies did find a positive and statistically significant relationship between BITs and FDI. Like Hallward-Driemeier (2003), Neumayer and Spess (2005) studied the FDI flows from OECD countries to developing countries. Their results are mixed. Neumayer et al. (2005) state that developing countries that sign more BITs with developed countries receive more FDI. Also, there appears to be a trade-off in the sense that developing countries that give up some of their domestic policy autonomy by binding themselves to foreign investment protection receive more FDI. Neumayer et al. found limited evidence that BITs function as substitutes for institutional quality (Neumayer et al., 2005:1575). Salacuse and Sullivan (2005) study the interrelation between BITs and FDI and compare the effectiveness of US and OECD BITs. Their results suggest that US BITs have a positive and significant relationship with the host country’s overall FDI inflows, whereas the impact of other OECD BITs is weaker (Salacuse et al., 2005:104). Egger and Pfaffermayr (2004) investigated the relation between BITs and FDI by analysing FDI stocks from OECD countries into both OECD and non-OECD countries. They demonstrate a significant positive correlation between FDI and approved BITs as well as a positive but less salient interrelation between FDI and signed BITs. Busse, Koeniger and Nunnenkamp (2008) demonstrate that BITs promote FDI inflows to developing countries from developed countries. On the basis of these findings, they suggest that policy makers of developing countries should strive for BITs to promote FDI (Busse et al., 2008:24).
America, has never even entered into a BIT (IISD, 2012). Future statistics on FDI stock will show whether FDI flows increase or decrease after the termination of BITs with investor countries. In order to understand the line of reasoning behind the South African government’s stance on BITs, a brief review of the country’s historical record regarding FDI is essential. Section VI discusses the impact of BITs on FDI in South Africa in detail.
III. HISTORICAL DESCRIPTION OF FDI
The last two decades, FDI from TNCs played a significant role in connecting national economies and building integrated international production systems. The attractiveness of South Africa as a destination for inward FDI stock fluctuated. These fluctuations were partially due to the prevailing ‘dual economy’ in South Africa, where an economy comparable to that of an industrialized country co-exists with an economy of a developing nation (A. van Oudheusden; Wöcke et al., 2013:2). On the one hand, South Africa has a sound regulatory and legislative environment for investment, a sophisticated business sector and globally competitive financial markets. On the other hand, pervasive poverty, high-income inequality, challenges in health care and education, and inefficient labour markets exist side by side. I will start this section by providing a historical description of FDI trends in South Africa between 1980 and 2012. I have examined patterns of FDI in South Africa before and after 1994. In that year, South Africa ended the apartheid regime and held its first democratic elections. It would be interesting to see what happened with respect to international investment activities after the 1994 overhaul of the South African government. In the following sections, patterns of both inward and outward FDI stocks will be examined. Sectorial trends in FDI activities will also be addressed, as will the patterns in different types of FDI entry modes. Finally, I will analyse patterns in Dutch FDI activities in South Africa. This section concludes with a summary. The information is drawn from previously published studies, World Investment Reports of UNCTAD, Quarterly Bulletins of the South African Reserve Bank (SARB), and statistics of the OECD. These different sources allow for a comparison between FDI stock data including and excluding SPEs. Unlike the OECD, UNCTAD and the SARB both include SPEs in the FDI stock data.
3.1 FDI in South Africa
that will improve their standing as investment countries (Pigato, 2000:2). During the 1970s and 1980s, Sub-Saharan Africa was regarded as an unattractive location for foreign investment, due to the region’s political and economic instability, poorly skilled labour force, inadequate infrastructure, and geographical location, far away from export markets (Fedderke et al., 2006:740; Pigato, 2000:5; UNCTAD, 1999). In this period, the capability of African countries to attract FDI hinged on natural resources and local markets (Asiedu, 2006:63; Morisset, 2000:4; Pigato, 2000:3). However, the level of FDI in Sub-Saharan Africa has increased significantly since the 1980s.
There are two ways of analysing FDI patterns. One focuses on flows, the other on stocks. FDI flows refer to the amount of FDI over a period of time and consist of annual changes in share capital, reinvested profits and other investments, including loans (CBS, 2013). FDI stocks refer to the total accumulated value of foreign owned assets at a particular point in time, consisting of capital participations, loans, and other liabilities. An example will clarify this distinction (see table 1). At the end of 2011 the amount of South Africa’s inward FDI stock was US$ 134.4 billion. By accumulating the FDI flows of all countries that invested in South Africa in 2012, we arrive at the total amount of US$ 4.6 billion. This amount is added to the FDI stock at the end of 2011, resulting in US$ 139 billion inward FDI stock at the end of 2012. In other words, the difference in FDI stock between the beginning of 2011 and the end of 2011 is the cumulative value of FDI flows plus a correction for changed exchange rates or changed prices. FDI stocks include previously invested stock on FDI to provide a nation’s position with respect to total FDI. Therefore, the focus will be on FDI stocks.
Table 1. Example of FDI stocks and flows
Example:
The amount of inward FDI stock of SA 01/01/12 US$ 134.4 billion
PLUS: Net foreign investments in 2012 (flow) US$ 4.6 billion
The amount of inward FDI stock of SA 31/12/12 US$ 139 billion
assets of the reporting economy at the end of the year. Outward FDI stock refers to the investments of the reporting country held abroad, listed as liabilities of the reporting country at the end of the year. Official FDI data is regularly and directly collected from central banks, statistical offices or national authorities on an aggregated and disaggregated basis for the FDI/TNC database (UNCTAD, 2013).
3.1.1 FDI in South Africa between 1980 and 2012
As illustrated by figure 1, both inward and outward FDI increased significantly between 1980 and 2012. In this period the level of inward stock increased from US$ 16.5 billion to US$ 139 billion. Outward stock increased from US$ 5.5 billion in 1980 to US$ 82.4 billion in 2012. In the period between 1980 and 1985 the inward stock was larger than the outward stock. Until 1985 South Africa can be regarded as a typical developing country, as foreign investments easily exceeded South African investments abroad. In other words, the South African economy experienced a net inflow of capital until 1985. Between 1985 and 1995 a shift took place, as the amount of outward FDI became larger compared to the amount of inward FDI, eventually even resulting in a net outflow of FDI capital. However, since 2000 inward FDI stock exceeds outward stock, and South Africa is back to a net inflow of FDI capital. According to A. van Oudheusden, fluctuations in FDI patterns are mainly influenced by political actions. In addition, a certain relationship exists between the fluctuations in FDI patterns and the exchange rate of the ZAR/US$ (A. van Oudheusden). To provide a detailed overview of certain FDI trends, the periods before and after 1994 will be discussed separately.
Figure 1. FDI stock in SA 1980 – 2012 (source: UNCTAD FDI/TNC database 2013)
Between 1980 and 1994, the amount of outward FDI stock increased from US$ 5.5 billion to US$ 19.1 billion (figure 2). Unlike outward FDI, inward FDI declined from US$ 16.5 billion in 1980 to US$ 12.6 billion in 1994. Between 1980 and 1988, a significant decrease of inward FDI stock is noticeable. In this period, several trade and financial sanctions were imposed by countries, such as the US, against South Africa mainly due to the Apartheid regime (A. van Oudheusden). From 1980 to 1985, inward FDI stock is larger than outward FDI stock, which means there was a net inflow of capital. In contrast, from 1985 to 1994 the level of outward FDI stock surpassed the amount of inward FDI stock, meaning that there was a net outflow of capital. In 1985, political unrest contributed to significant capital outflows and panic selling of the Rand resulting in the declaration of a debt standstill by the South African government (Wöcke et al., 2013:3). Expressed as a percentage of gross domestic product (GDP), the inward and outward FDI stocks saw only a modest increase between 1990 and 1994.9 Inward stock as a percentage of GDP increased from 8.2% in 1990 to 9.3% in 1994. Between 1990 and 1994, outward stock as percentage of GDP increased from 13.4% to 14.1% respectively.
Figure 2. FDI stock in SA 1980 – 1994 (source: UNCTAD FDI/TNC database 2013)
Between 1995 and 2012, the amount of inward FDI stock increased from US$ 15 to US$ 139 billion. Since 1995, South Africa emerged from an isolated economy towards a strong pole of
9 Statistics of FDI stock as a percentage of GDP before 1990 are not included in the UNCTAD
attraction for foreign investors (A. van Oudheusden). In 1999, a peak of inward FDI is detected (see figure 3). Between 1999 and 2001, the final repayment of the debt standstill (imposed by the government in 1985) has been made, indicating a positive turnaround of South Africa’s international credit rating. The outward FDI stock grew from US$ 23.3 billion in 1995 to US$ 82.4 billion in 2012. In that same year inward FDI stock was almost twice as large as outward FDI stock. As figure 3 demonstrates, between 1995 and 1998 the outward FDI stock was slightly higher than the inward FDI stock. Since 1998, the inward FDI stock has been larger than the outward FDI stock, resulting in a net inflow of FDI capital. As illustrated by figure 3, there is an upward trend in both inward and outward FDI stock. Expressed as a percentage of GDP, both inward and outward FDI have increased dramatically between 1995 and 2012: from 9.9 to 35.6% and from 15.4 to 21.1%.
Figure 3. FDI stock in SA 1995 – 2012 (source: UNCTAD FDI/TNC database 2013)
Summing up, during the past two decades inward and outward FDI in South Africa has increased substantially. This is in line with global patterns, as more and more other developing countries, too, are emerging as outward FDI investors (UNCTAD, 2005:5). For example, the BRICS (Brazil, Russia, India, China, and South Africa) are becoming the leading sources of FDI among emerging investor countries, not only as major recipients of
FDI, but also as sources of outward FDI (Sandrey, 2013:8; UNCTAD, 2013:4). Furthermore, TNCs from emerging markets such as Malaysia, South Africa, China and India (in this order) are the largest developing-country investors in Africa in terms of FDI stock (UNCTAD, 2013:40). But while the outward FDI from developing countries increased, outward FDI from developed countries into developing countries declined by more than US$ 274 billion in 2012. Especially Belgium, the US, and the Netherlands experienced major declines in their outward FDI stocks (OECD, 2013; UNCTAD, 2013:6).
3.2 Sectorial patterns of FDI
FDI activities can be classified into three main sectors: the primary, secondary, and tertiary sector. The primary sector consists of agriculture and mining activities. The secondary sector encompasses manufacturing, construction, and electricity, gas and water. The tertiary sector or services sector comprises commerce, transport and communication, and financial and other services. From what we currently know about FDI stocks it appears that sectorial structures of international production have changed profoundly over the past two decades (UNCTAD, 2013).
Table 2. Sectorial distribution of inward FDI stock, 2002 and 2012
Sector/Industry 2002 as % 2012 as %
Primary 137.0 13% 341.1 12%
- Agriculture, forestry and fishing 44.2 4% 71.0 3%
- Mining and quarrying 92.7 9% 270.1 10%
Secondary 254.8 24% 531.1 19%
- Manufacturing 204.2 19% 341.0 12%
- Electricity, gas and water 26.5 2% 84.5 3%
- Construction 24.1 2% 105.6 4%
Services 673.9 63% 1 948.1 69%
- Wholesale and retail trade, catering and accommodation 143.2 13% 455.5 16%
- Transport, storage and communication 101.0 9% 257.4 9%
- Finance, insurance, real estate and business services 207.7 19% 597.9 21%
- Community, social and personal services 222.0 21% 637.3 23%
Total inward FDI stock 1 065.7 100% 2 820.3 100%
Source: SARB Quarterly Bulletin, various issues (amounts in billion South African Rand) Interesting to mention is the motor vehicle industry in South Africa. The country has a well-developed auto assembly industry since the 1970’s and the South African government supports investments in the automotive sector through incentive programs that encourage production for the global market (A. van Oudheusden). Companies such as BMW and Daimler influences the manufacturing sector significantly.
3.3 M&A versus Greenfield
Many foreign investors conduct cross-border mergers and acquisitions (M&As) transactions through SPEs located abroad, mainly for international tax planning purposes. In other words, SPEs are often established to serve as an acquisition vehicle in M&A transactions. This study gauges how investment patterns in South Africa would change when the use of SPEs would be prohibited. This means that notice should be taken of patterns in FDI entry modes. There are two main modes through which FDI activities can be initiated, cross-border mergers and acquisitions and Greenfield investments (UNCTAD, 2009:97). While cross-border M&As imply the purchase or sale of existing equity, Greenfield investments refer to investments that are altogether new (OECD, 2008:87). Data on FDI entry modes is derived from the World Investment Reports issued by UNCTAD. Statistics of cross-border M&As are available from Thomson Reuters and data on Greenfield investments are available from fDi Markets of the Financial Times.10
Patterns in M&A transactions can be analysed on the basis of the value of sales and purchases.11 Net cross-border M&A sales is calculated as follows: the amount paid for companies in the host economy to foreign TNCs minus the amount paid for foreign affiliates in the host economy (UNCTAD, cross-border M&A database 2013). Additionally, net M&A purchases are calculated as follows: purchases of companies abroad by home-based TNCs minus sales of foreign affiliates of home-based TNCs (UNCTAD, cross-border M&A database 2013). This study adopts the perspective of South Africa (as the host economy), which is why only M&A sales are examined. A complete overview of M&A transactions in South Africa is provided in appendix 1, which includes the number and value of both sales and purchases. The value of M&A sales fluctuated drastically during 1990-2012 period (see table 3). For example, in 1998 a net value of – US$ 5.7 billion was reached compared, 2001 saw a peak of US$ 11.8 billion. The negative value of – US$ 5.7 billion implies that the sales of foreign affiliates in South Africa are larger than the sales of companies in South Africa to foreign TNCs. The M&A balance differ from sector to sector. For example, in 2012 cross-border M&A sales took place primarily in the secondary (44%) and services sector (40%) (UNCTAD, 2013).12
Table 3. Values of M&A sales
M&As 1990 1995 2000 2005 2 010 2 011 2 012
Value of Sales - 15 540 308 5 092 3 934 6 632 - 879
Source: UNCTAD FDI/TNC database 2013 (values in US$ million)
Data of Greenfield investments can be analysed by source or by destination.13 Only the values of South Africa as a destination of Greenfield investments are analysed, since South Africa is perceived as the host economy (vis-à-vis FDI) during this study.14 The values of Greenfield investments, with South Africa as a destination, remain fairly constant between 2003 and
11 Data only covers those M&A deals that involved an acquisition of an equity stake of more than
10%. Furthermore, data on sales refer to the net sales by the region/economy of the immediate acquired company as data on purchases refer to the net purchases by the industry of the ultimate acquiring company (UNCTAD, cross-border M&A database 2013).
12 Percentages are based on the values of cross-border M&A sales and are calculated based on
statistics of the UNCTAD FDI/TNC database 2013.
13 Data refers to the estimated amounts of capital investment.
14 A complete overview of Greenfield investments in South Africa is provided in appendix 2 including
2012 (see table 4).15 Even though, high values of Greenfield investments are detected in 2008 and 2011 with volumes of US$ 13.5 billion and US$ 12.4 billion respectively. A shift in sectorial patterns is detected as well, as 18% of total Greenfield investments concerned the primary sector in 2003 compared to only 4% in 2012.16 The value of Greenfield investments in the secondary sector decreased between 2003 and 2012 from 52% to 43%. The services sector, on the other hand, saw a growth of the Greenfield investments from 31% in 2003 to 53% in 2012. Going by the data on Greenfield investments, it is fair to say that a sectorial shift has taken place in South Africa from the primary towards the services sector.
Table 4. Value of Greenfield investment by destination
Greenfield investment 2003 2005 2010 2011 2012
Value by Destination 4 452 3 658 6 819 12 413 4 571
Source: UNCTAD FDI/TNC database 2013 (values in US$ million)
From the data presented above, we can conclude that the value of M&A sales in South Africa fluctuated significantly between 1990 and 2012. This is a marked contrast to the value of Greenfield investments, which remained fairly constant in South Africa between 2003 and 2012. Furthermore, the share of Greenfield investments as a share of the total FDI is increasing. Especially developing countries receive more and more FDI through Greenfield projects (UNCTAD, 2005:10), although the largest Greenfield FDI projects in South Africa in the period 2008-2010 were generally smaller than the largest M&A deals of 2010.
3.4 Investment patterns of Dutch FDI activities in South Africa
Between 1990 and 2000, approximately 90% of total FDI in Sub-Saharan Africa came from a small number of countries, most importantly the UK, France, the US, Japan, Germany, and the Netherlands (Pigato, 2000:6). A similar pattern can be observed regarding recent inward FDI activities in South Africa (SARB, 2012). Nevertheless, FDI in South Africa is coming from an increasing group of countries, including Asian economies such as Japan, China and India. But despite this increase of Asian FDI, European economies are still the largest source of inward FDI stock in South Africa (SARB, 2012).
15 Before 2003 no data is recorded on Greenfield investments by UNCTAD.
16 Percentages are based on the values of Greenfield investments and are calculated based on statistics
To analyse patterns of Dutch FDI activities in South Africa, data have been obtained from two sources, namely the databases of the SARB and the OECD. The documentation of the two data sources is different, however, as the SARB, unlike the OECD, includes SPEs in their FDI stock data. By combining the two data sources, the value of Dutch SPEs in South Africa can be estimated. First, I will describe historical investment patterns of Dutch FDI activities in South Africa based on statistics of the SARB (including SPEs). Second, FDI patterns of the estimated value of Dutch SPEs in South Africa will be outlined. Third, I will present the share of total inward FDI stock in South Africa contributed by the Netherlands. A distinction is made between the contribution of total Dutch investments (including SPEs) and the contribution of Dutch SPEs to the total amount of inward FDI in South Africa.
The inward FDI stock from the Netherlands into South Africa increased considerably during the past two decades. As illustrated by table 5, the total value of Dutch investments in South Africa (including SPEs), increased from US$ 1 billion in 1995 to US$ 24.3 billion in 2010 (SARB, 2012).17 In that same year 17.5% of total inward FDI stock was contributed by Dutch companies. As table 4 shows, there is a significant difference between data of the SARB and the OECD that can be attributed to the inclusion of SPEs in the SARB data.18 As table 5 illustrates, the estimated value of Dutch SPEs increased from US$ 0.8 billion to US$ 22.3 billion between 1995 and 2010. Furthermore, it appears that a large share of the total investments from the Netherlands into South Africa is contributed by Dutch SPEs. For example, in 2010 SPEs accounted for 92% of the total Dutch investments (see table 5). Table 5. Dutch investments in SA
1995 as % 2000 as % 2005 as % 2006 as % 2007 as % 2008 as % 2009 as % 2010 as % 1. Total investments, incl. SPEs 1.0 100% 0.8 100% 2.2 100% 3.3 100% 4.1 100% 3.9 100% 10.8 100% 24.3 100% 2. Total investments, excl. SPEs 0.2 19% 0.5 57% 0.8 36% 1.0 31% 1.5 37% 1.5 38% 1.8 17% 2.0 8% Investments by SPEs (1-2) 0.8 81% 0.3 43% 1.4 64% 2.3 69% 2.6 63% 2.4 62% 9.0 83% 22.3 92%
Source: SARB, Quarterly Bulletins various issues (amounts in US$ billion)
17 However, Dutch inward FDI stock in South Africa declined from US$ 2.5 billion in 2004 to US$
2.2 billion in 2005. The period 2008-2010 shows a remarkable increase from US$ 3.9 billion to US$ 24.3 billion.
18 A detailed description of the calculations of inward FDI stock (including and excluding SPEs) from
As table 6 shows, the share of Dutch companies in South Africa’s inward FDI increased from 6.7% in 1995 to 17.5% in 2010.19 At 16.1%, the share of Dutch SPEs was almost as big, meaning that SPEs account for the lion’s share of Dutch FDI into South Africa. In 1995, SPEs accounted for only 5.4%, which suggests a marked growth in the period 1995-2010.
Table 6. Share of Dutch investments of total inward FDI stock in SA
1995 as % 2000 as % 2005 as % 2006 as % 2007 as % 2008 as % 2009 as % 2010 as % Total inward FDI stock in SA 15,1 100% 47,4 100% 78,6 100% 90,4 100% 106,7 100% 76,6 100% 102,3 100% 138,7 100% Investments by NL in SA (incl. SPEs) 1,0 7% 0,8 2% 2,2 3% 3,3 4% 4,1 4% 3,9 5% 10,8 11% 24,3 18% Estimated value of Dutch SPEs 0,8 5% 0,3 1% 1,4 2% 2,3 3% 2,6 2% 2,4 3% 9,0 9% 22,3 16%
Source: SARB, Quarterly Bulletins various issues (amounts US$ billion) 20
In sum, the FDI stock from the Netherlands in South Africa declined between 1995 and 2000, but has been increasing ever since. Especially between 2005 and 2010 there was a strong growth. Furthermore, it is clear that a large share of the total Dutch investments into South Africa is contributed by Dutch SPEs. Finally, the Netherlands is becoming increasingly important as an investor in South Africa, and the contribution of Dutch SPEs to total inward FDI in South Africa is on the rise.
3.5 Summary of FDI patterns in South Africa
The overall increase of inward and outward FDI activities in South Africa is substantial. In addition, it appears that during the last decade the inward FDI stock has seen a minor shift (in relative terms) from the primary and secondary sector towards the services industries. In 2002 as well as in 2012, the services sector accounted for a larger share of total inward FDI stock in South Africa than the primary and secondary sector. Furthermore, the share of Greenfield investments in South Africa’s total FDI is increasing. At the same time, though, the share of Greenfield FDI projects is smaller than the share of M&A deals. The estimated value of Dutch SPEs in South Africa increased between 1995 and 2010 from US$ 0.8 billion to US$
19 A detailed description of the calculations of total inward FDI stock in South Africa is provided in
appendix 4.
20 The values of total inward FDI stock in table 6 differ from the values provided in figure 3, since the
22.3 billion. In 2010 Dutch investments accounted for 17.5% of the inward FDI stock in South Africa. 16.1% of total the inward FDI stock was contributed by Dutch SPEs.
These statistics raise several questions. For example, what are the characteristics of these Dutch SPEs? Where are their parent companies located, and in which sectors are they mainly operating? And are the Dutch SPEs active in different regions, sectors and kinds of companies than Dutch companies that are active in South Africa?
IV. DATA COLLECTION AND METHODOLOGY
It is not possible to make assertions about the differences between the activities of Dutch companies and Dutch SPEs solely on the basis of the historical descriptions of FDI stock in South Africa. In keeping with the main objective of this study, the characteristics of Dutch companies and Dutch SPEs are compared. This comparison is based on data from several sources. First, data is retrieved from the Dutch daily (Het Financieele Dagblad). Two data journalists, Eikelenboom and De Groot investigated the role of Dutch organisations in Africa. In collaboration with Company.info, they composed a database of more than 2,000 Dutch organisations that are investing or participating in business entities in Africa in 2011. About 38% of these Dutch organisations are fully Dutch companies. The other 62% consists of multinationals investing in Africa through Dutch financial holdings (Eikelenboom and De Groot, 2013). It should be noted that multinationals with a Dutch financial holding include both SPEs and subsidiaries, a distinction that will be explained in more detail later on. Company.info manages an online database with information about more than 2.5 million Dutch organisations. This database contains recent information on companies, obtained from different sources such as the Dutch Chamber of Commerce, the insolvency register, and through the analysis of annual reports of organisations. All new businesses are required to register at the Dutch Chamber of Commerce, and every company that is registered, is also included in the database of Company.info. In collaboration with Company.info, the two data journalists established a database listing Dutch organisations that are responsible for outward Dutch FDI to Africa. This database does not include information of the year in which a certain FDI in Africa has been made.
Africa is one of the emerging BRICS countries that are rapidly becoming more important in the world economy. Finally, South Africa is a gateway to other African countries; the way Hong Kong is a gateway to China. A total of 554 Dutch companies, investing in business entities in South Africa in 2011, are incorporated in the database. Of these 554 companies, 220 (39.7%) are fully Dutch companies. The other 334 (60.3%) are multinationals investing in business entities in South Africa through a Dutch financial holding. The database records information on the following variables:
1. Parent company
2. Country in which the parent company is headquartered 3. Name of the Dutch business entity
4. Name of the business entity located in South Africa
5. Location in South Africa where the business entity is located
6. Percentage of the parent company participating in the local business entity
Regarding the 554 companies, it should be noted that this concerns the number of transactions by Dutch companies in South Africa. Consider the hypothetical parent company X, which is investing in four business entities in South Africa through a Dutch business entity. Even though company X is one company, I will consider these four transactions as separate instances of Dutch FDI, since investments are made in four different business entities in South Africa.
Second, business information of the parent companies is retrieved from ORBIS, an electronic statistical database containing comprehensive information on 120 million companies worldwide. It includes listed as well as unlisted companies. By making use of ORBIS, detailed business information was collected concerning the type of entity and the sectors in which Dutch companies are participating. ORBIS provides different sector classifications, such as the Bureau van Dijk (BvD) major sector, US SIC core codes, and the NACE Rev. 2 main sectors. The latter will be applied during this study since it is often used by organisations like UNCTAD, the OECD, and the World Bank to classify FDI activities.21 NACE Rev. 2 is the statistical classification of economic activities in the European
21 Detailed descriptions of the sector classification of BvD major sector and US SIC core codes are
Community and is based on the International Standard Industrial Classifications of all economic activities (ISIC core codes).
Third, interviews are conducted to provide background information on the statistics of FDI patterns in South Africa. First, an interview is accomplished with Francis Weyzig, who completed a PhD research on taxation and developments, analyzing the effects of Dutch tax policy on taxation of multinationals in developing countries. Second, I conducted an interview with Albert van Oudheusden, who is a Dutch consultant in South Africa and lives there for 44 years. The interviews had an open character in order to gather as much valuable background information as possible.
This study is written from the perspective of South Africa, meaning that Dutch companies investing in business entities in South Africa are regarded as inward FDI and South African companies investing in business entities outside South Africa is considered outward FDI. Since this study examines investment patterns of Dutch companies in South Africa, only patterns of inward FDI activities are analysed. According to UNCTAD (2009:92), data on FDI stocks should include information to the sector or economic activity of both the parent company and the foreign affiliate. However, as information of local businesses in South Africa is often not available in the ORBIS database, only the parent companies will be taken into consideration in the current research project. In addition, the location of the headquarters of the parent company and the country in which the parent company is located are checked to see if they match.
As previously stated, by Dutch companies we mean ‘fully’ Dutch companies and multinationals investing in business entities in South Africa through a Dutch financial holding. These Dutch financial holdings can be divided into SPEs and subsidiaries that establish physical operations in the Netherlands. According to the business information from ORBIS, the majority of the Dutch financial holdings are SPEs with only a handful of registered employees in the Netherlands.22 The data on English parent companies investing in South Africa through a Dutch financial holding are illustrative in this regard.23 16% of these Dutch
22 I took “only a few” to mean less than 7 registered employees.
23 United Kingdom is chosen since the UK is considered as the largest investor of total inward FDI
holding companies have on average 260 employees,24 33% does not include the number of employees registered in the Netherlands, and the other 51%, has an average of only 2.4 registered employees.25 Furthermore, 76% of English parent companies are investing in business units in South Africa within the NACE Rev. 2 main sector K (Financial and insurance activities).26 Therefore, I considered Dutch financial holdings SPEs or ‘mailbox’ companies. Henceforth, multinationals investing in South Africa by making use of a Dutch financial holding are considered as Dutch SPEs.
By combining data from the data journalists of Het Financieele Dagblad, Company.info, and ORBIS, I established my own data set. Using this dataset, several crosstabs were set up to provide an all-encompassing investment pattern of Dutch organisations investing in business entities in South Africa. Results of these crosstabs are described and analysed in the following section.
V. ANALYSIS
Several crosstabs are provided to answer the main research question that was formulated in the introduction. Regional patterns, sectorial patterns as well as ownership structures of parent companies investing in business entities in South Africa are examined. Due to the unequal size of Dutch companies and Dutch SPEs, it has been decided to focus on percentages rather than on absolute amounts. In addition, Dutch SPEs make larger investments in business entities in South Africa than Dutch companies. Approximately 40% of the Dutch companies are fully Dutch companies, which invested a total of US$ 2 billion in 2010. The other 60% are Dutch SPEs, which invested US$ 22.3 billion in 2010.
5.1 Regional patterns
This section describes regional patterns of the parent companies investing in business entities in South Africa and the location of these business entities in South Africa.
24(53+53+36+1140+69+244+214) / 7
25(1+3+1+1+1+4+1+1+5+5+7+3+1+3+1+3+1+1+2+4+4+1+1) / 23
26 “In section “K - Financial and insurance activities”, two classes have been introduced that go
5.1.1 Regional patterns of the investors
There is a certain regional pattern regarding multinationals with a Dutch financial holding. As explained above, the country in which the headquarters of the parent company is located is considered the country of origin. Almost 86% of the multinationals with a Dutch financial holding are originally located in other European countries (44.3%) and North America (41.3%). The remaining 14% is contributed by parent companies originally located in Africa (3.9%), Asia (10.2%), and South America (0.3%). Furthermore, only five countries – the US (39.5%), the UK (13.5%), Switzerland (8.4%), Germany (7.2%), and Japan (6.3%) – account for almost 75% of the multinationals. Several examples of multinationals of which the parent company is located in one of these five countries are: Hewlett-Packard (US), Experian (UK), ABB Group (Switzerland), BASF (Germany), and Fujitsu (Japan).
5.1.2 Regional patterns of local business entities in South Africa
Table 7. FDI investments in SA by Dutch companies and the location in SA
Cape Town Johannesburg Pretoria Rivonia Sandton Other Total
Dutch Company 36 24 9 0 1 62 132 as % 27.3% 18.2% 6.8% 0.0% 0.8% 47.0% 100% Dutch SPE 12 57 6 12 12 75 174 as % 6.9% 32.8% 3.4% 6.9% 6.9% 43.1% 100% Total 48 81 15 12 13 137 306 5.2 Sectorial patterns
The ‘Type of Entity’ classification and the ‘NACE Rev. 2’ are used as a framework to provide an overview of the relationship between FDI investments in South Africa by Dutch companies and the sector in which they are operating.
5.2.1 Type of Entity
First, the relation between FDI investments in South Africa by Dutch companies and the ‘Type of Entity’ will be analysed. ‘Type of Entity’ is broken down into the categories ‘Financial Company’, ‘Industrial Company’, ‘Bank’, ‘Foundation or Research Institute’, ‘Insurance Company’, and ‘Mutual and Pension Fund/Nominee/Trust/Trustee’. As no information is missing, the sample size is 554. The results are listed in table 8. It is clear that both Dutch companies (72.7%) and multinationals (90.1%) are primarily industrial companies. In this respect Shell, Harsco, and Techint can be considered as examples. However, the rate of industrial companies is higher among multinationals than among the Dutch companies. Another major category is the financial companies, which account for 21.4% of the Dutch companies and for 6.9% of the multinationals. The contribution of other types of entities, such as banks, foundation or research institutes, insurance companies, and mutual and pension funds, is negligible.
Table 8. FDI investments in SA by Dutch companies and the Type of Entity
Bank Financial Foundation Industrial Insurance Mutual Total
Dutch Company 6 47 3 160 0 4 220
as % 2.7% 21.4% 1.4% 72.7% 0.0% 1.8% 100%
Dutch SPE 3 23 1 301 6 0 334
as % 0.9% 6.9% 0.3% 90.1% 1.8% 0.0% 100%
5.2.2 NACE Rev. 2
This section examines the relation between FDI investments in South Africa by Dutch companies and the ‘NACE Rev. 2 main sectors’. The NACE Rev. 2 lists 18 sectors (see table 9). As the relevant information in ORBIS was missing for two companies, the sample size is 552. As table 9 shows, about 60% of the fully Dutch companies are operating in manufacturing (28.4%), wholesale and retail trade (17%), and financial and insurance activities (14.7%). An example of a Dutch manufacturing company is Akzo Nobel. Approximately 60% of FDI investments of multinationals with a Dutch financial holding, concerns manufacturing (a striking 51.5%) and professional, scientific and technical activities (10.5%). Examples of manufacturing companies investing in South Africa through a Dutch SPE are BMW and General Electric. The differences between Dutch companies and multinationals are clear: while both groups are active in manufacturing, Dutch companies focus more on wholesale and retail trade and financial and insurance activities than multinationals. Both Dutch companies and multinationals invest in professional, scientific and technical activities, although the latter (10.5%) more emphatically than the former (7.3%). Finally, Dutch companies spend 4.6% of their investments on construction, compared to only 1.8% for multinationals. For the other NACE Rev. 2 main sectors, the differences are minor.
Table 9. FDI investments in SA by Dutch companies and the NACE Rev. 2
Dutch Dutch
Company as % SPE as % Total
A - Agriculture, forestry and fishing 8 4% 1 0% 9
B - Mining and quarrying 13 6% 19 6% 32
C - Manufacturing 62 28% 172 51% 234
D - Electricity, gas, steam and air conditioning supply 0 0% 7 2% 7
E - Water supply; sewerage, waste management and remediation activities 2 1% 0 0% 2
F - Construction 10 5% 6 2% 16
G - Wholesale and retail trade; repair of motor vehicles and motorcycles 37 17% 16 5% 53
H - Transportation and storage 7 3% 14 4% 21
I - Accommodation and food service activities 3 1% 4 1% 7
J - Information and communication 8 4% 22 7% 30
K - Financial and insurance activities 32 15% 18 5% 50
L - Real estate activities 7 3% 3 1% 10
M - Professional, scientific and technical activities 16 7% 35 10% 51
N - Administrative and support service activities 5 2% 14 4% 19
P - Education 3 1% 3 1% 6
Q - Human health and social work activities 1 0% 0 0% 1
R - Arts, entertainment and recreation 1 0% 0 0% 1
S - Other service activities 3 1% 0 0% 3
5.2.3 Primary, secondary and services sector
The sectors of the NACE Rev. 2 can be categorised into the primary (A – B), secondary (C – F) and services sector (G – S).27 The percentages for these aggregated categories are shown in table 10, which shows that the secondary (47%) and the services sector (46%) absorb the vast majority of Dutch FDI in South Africa. Dutch companies are mainly active within the services sector (56%), whereas Dutch SPEs are mostly operating in the secondary sector (55%). Statistics of the SARB (2012) bear out that in 2012 the secondary sector accounted for 19% of total inward FDI stock, and the services sector for 69%. This means that countries generally prefer to invest directly in the services sector. Dutch companies, on the other hand, are investing in both the secondary and the services sector, with the qualification that fully Dutch companies operate mainly in the services sector and Dutch SPEs in the secondary sector. Thus, it appears that it is interesting for TNCs to invest in the secondary sector through a Dutch holding rather than directly. This is not surprising, since the Netherlands has a relative large services sector, which is also borne out by the Dutch FDI activities in South Africa.
Table 10. Sector classification of Dutch companies and Dutch SPEs in SA Dutch Companies as % Dutch SPEs as % Total as % Primary (A - B) 21 10% 20 6% 41 7% Secondary (C - F) 74 34% 185 55% 259 47% Services (G - S) 123 56% 129 39% 252 46% Total amount 218 100% 334 100% 552 100%
5.3 Ownership structures of parent companies in South Africa
This paragraph addresses the relation between FDI investments in South Africa by Dutch companies and the ownership structures of business entities in South Africa. As the data were missing for 44 companies, the sample size is 510. The data are laid out in table 11. Since a relatively small number of Dutch companies have a minority share (a share of less than 49%) in business entities in South Africa, these categories are taken together. Furthermore, the classification of ‘100% ownership’ is considered as a separate category. As table 11 illustrates, the majority of both Dutch companies and multinationals possesses 100% ownership of the business entities in South Africa. For example, Philips, Tetra Laval, and
27 Appendix 7 provides a detailed description of the NACE Rev. 2 and the primary, secondary, and
SAB Miller. Although, multinationals are classified to a larger extent into the category ‘100% ownership’ compared to Dutch companies with 70.6% and 56.9% respectively. A relatively large percentage of Dutch companies (15.6%) has a minority share in a South African company, compared 8% of the multinationals. A similar pattern can be observed regarding the 80-89% category, as Dutch companies (5.2%) are more strongly represented in this category than multinationals (1.3%). In the remaining categories the differences are negligible. These results suggests that TNCs that invest in South Africa through a Dutch SPE are more likely to have 100% ownership of a business entity than Dutch companies. However, it is unknown whether investments of Dutch companies involved an existing foreign affiliate (M&A) or a newly established or acquired business entity (Greenfield investment).
Table 11. Percentages of parent companies investing in business entities in SA
0-49% 50-59% 60-69% 70-79% 80-89% 90-99% 100% Total Dutch Company 33 22 3 17 11 5 120 211 as % 15.6% 10.4% 1.4% 8.1% 5.2% 2.4% 56.9% 100% Dutch SPE 24 23 8 25 4 4 211 299 as % 8.0% 7.7% 2.7% 8.4% 1.3% 1.3% 70.6% 100% Total 57 45 11 42 15 9 331 510
VI. RESULTS AND DISCUSSION
value of these Dutch SPEs was US$ 22.3 billion in 2010. This means that SPEs account for 92% of the total value of Dutch investments. In other words, 60% of all Dutch companies investing in South Africa represent 92% of the total Dutch investments (see table 12). During the past two decades the contribution of fully Dutch companies investing in business entities in South Africa is negligible (see table 6 and 12). Dutch investments in South Africa mainly consist of investments from Dutch SPEs instead of investments of fully Dutch companies. Therefore, we can state the Netherlands is more or less a tax haven or OFC instead of an investor when examining inward FDI from the Netherlands into South Africa.
Table 12. Value of investments and number of Dutch companies investing in SA
Value of Number of
investments as % companies as %
"Actual" Dutch companies 2.0 8% 220 40%
Dutch SPEs 22.3 92% 334 60%
Total value / amount of Dutch companies 24.3 100% 554 100%
Source: SARB, Quarterly Bulletin and database of Eikelenboom and De Groot (values in US$ billion)
6.1 Third country involvement in Dutch SPEs in South Africa
