3. IFFs in the global and South African context

IFFs, and policies to combat them, represent a complex policy area. Due to their inherently secret nature, IFFs are difficult to analyse, quantify, and address. This chapter first provides a short synopsis of recent analysis of IFFs. It discusses how IFFs and their relevant subcomponents should be categorised. This is followed by a general overview of current global IFF estimates, a description of IFFs’ harmfulness for developing countries, and measurement challenges. The chapter then focuses on the South African context. It discusses some of the country’s still prevailing risks for IFFs, presents related estimates from the literature as well as important policy steps which have recently been implemented by South African authorities to rein in IFF leakages.

Defining IFFs is a challenging task. In the past two decades several attempts by international institutions and global policy actors to establish a widely acceptable definition have been made. Despite the emergence of certain commonalities in the various definitions, however, there remains no universally accepted definition. As a result, different terms in the area of IFFs have often been used interchangeably, obscuring the nuances relevant to designing appropriate policy responses or even successfully analysing and measuring IFFs. Only relatively recent definitions, for instance, departed from combining capital flight, the movement of funds abroad for better returns, with IFFs as revenue or proceeds derived from corruption, commercial activities or crime (Baker, 2005[1]). Based on this distinction, UNECA (2013[2]) referred to IFFs in its broadest sense as “money illegally earned, transferred or used”.

Since then consensus about the cross-border nature of IFFs has emerged. The flows are generated by transactions, thereby contravening national or international laws. While, for instance, the World Bank has adopted the relatively broad notion of IFFs as money illegally earned, transferred or used, the cross-border element has strongly been emphasised. IFFs have been defined as cross-border movement of capital that is associated with illegal activity in the local jurisdiction (World Bank, 2017[3]). However, despite the term ‘earned, transferred or used’ being commonly used in the context of IFFs, the term “illicit financial flows” is still the subject of disagreement for statistical purposes. Without a consistent and commonly agreed upon definition, sound and comprehensive measurement as well as international comparison of IFFs will remain a challenge.

The 2030 Agenda for Sustainable Development of the United Nations identifies the reduction of IFFs as a key target for the Sustainable Development Goals (SDGs). In an effort to create an indicator to measure progress towards achieving this SDG target1, UNCTAD and UNODC, in collaboration with national statistical offices, most recently established a Conceptual Framework for the Statistical Definition and the Measurement of IFFs (the framework) (UNODC/UNCTAD, 2020[4]). While the final SDG indicator envisions the measurement of the total value of IFFs, the framework seeks to establish a finer typology of IFFs and decomposes the aggregate category into a set of subcomponents to increase its applicability for statistical measurement and policy guidance. The resulting statistical definition considers IFFs as “financial flows that are illicit in origin, transfer or use, that reflect an exchange of value and that cross country borders”. However, while three potentially overlapping forms of IFFs are specified in the above definition (origin, transfer and use), they are mapped on to two alternative types of IFFs. Specifically, UNODC/UNCTAD suggest that IFFs can be generated at two fundamental stages, namely during income generation and income management (though presumably IFFs could be generated if income is spent illegally as well). In the framework, IFFs can be linked to cross-border transactions in the context of producing illicit goods or services or directly generating illicit flows, but also to the use of illicit income for investment or consumption or illicit management of income generated from legal economic activity.

The recently proposed framework by UNODC and UNCTAD offers a further IFF typology. The framework therefore suggest that there are four main categories of activities that may generate IFFs: i) illicit tax and commercial activities, ii) illegal markets, iii) corruption, and iv) exploitation or theft-type activities and financing of terrorism and crime. These four main categories are shown in Figure 3.1.

Most of these IFF categories can be related to clearly-defined crimes with associated financial flows crossing borders. IFFs from illegal markets involve criminal activities where income is generated through the exchange or trade of illegal goods or services. They include any type of trafficking in goods, such as drugs and firearms, or services, such as smuggling of migrants. IFFs from corruption include bribery, embezzlement, abuse of functions, trading in influence or illicit enrichment. Again, where the associated flows cross borders these would be IFFs. Similarly, financial flows related to exploitation-type activities such as slavery or trafficking in persons as well as financing of crime and terrorism that cross borders fall under the IFF definition as well. Proceeds from all of these activities are considered IFFs when they lead, directly or indirectly, to cross-border flows or when financial assets are transferred to commit crimes or have their origin in the above crimes (UNODC/UNCTAD, 2020[4]).

In the UNODC/UNCTAD framework illicit tax and commercial IFFs emerge as a category subdivided into illegal and legal activities. The illegal component of this category refers to illicit practices by legal entities or individuals with the aim to conceal revenues or reduce tax liabilities. Resulting IFFs include practices such as tariff, duty and revenue offences, tax evasion, contravention of exchange controls, competition offences and market manipulation amongst others, most of them unobserved.

The key form of IFF generated legally in the UNODC/UNCTAD framework is tax avoidance. Tax avoidance, while legal by definition, can also drain domestic resources and reduce government revenues. Given that these effects are negative and in some ways similar to the effects resulting from illegal IFFs, the UNODC/UNCTAD framework suggests that financial flows associated with tax avoidance should also be considered illicit where they cross borders. The report also notes that, while tax avoidance generally involves tax planning undertaken within the letter of the law, there is often substantial legal ambiguity over when aggressive tax planning becomes evasion. The line between evasion and avoidance can also be difficult to draw for statistical purposes where the legal environment around tax planning, particularly for MNEs, is changing rapidly. In such a context, the European Commission (2017[5]) suggests that the range of tax avoidance thus should be considered as a continuum of activities with fluid boundaries between legal tax planning and illegal tax evasion as the extremes.

The debate over whether tax avoidance should be included in the definition of IFFs remains highly contested. Fundamentally, suggesting that legal activities can be included in a definition of illicit activity introduces room for complexity, subjectivity and error into the IFF area which is not helpful for policy analysis or measurement. In the context of the SDG indicator and the underlying framework, tax avoidance, though generally considered legal, has been included as an IFF by UNODC/UNCTAD. However, strictly categorising tax avoidance or base erosion and profit shifting (BEPS) as IFFs may result in suggesting that legitimate economic activities associated with MNE investments are in fact illicit. This could include, for instance, MNEs funding investments in developing countries through debt financing, which would fall within the terms of current law. Such categorisation has the potential to conflate quite different policy challenges, which is unlikely to assist policy makers in properly diagnosing the problem and developing the appropriate policy response. It could also have the effect of diverting policy attention from other previously mentioned IFF categories, such as corruption, flows related to smuggling, trafficking, terrorist financing, exchange control contravention, and tax evasion. None of this is to understate the egregious nature of, or the extent to which tax avoidance is a problem, but it is a recognition of the difficulties that arise when tax avoidance is incorporated into the definition of IFFs. Automatically subsuming tax avoidance into IFFs may also lead to distorted IFF estimates as will be discussed further below. This is particularly the case when MNE activities have been the result of tax base choices made by countries. While the line between legal and illegal in the area of tax avoidance can often be contested by taxpayers and tax authorities, the line between licit and illicit forms of tax avoidance seems impossible to draw and is potentially highly subjective. This renders the challenge of accurately measuring IFFs even more difficult. For these reasons, this study excludes tax avoidance from its definition of IFFs.

IFFs cause fiscal and economic damage to countries in a variety of ways. For developing countries in particular, tax- and customs-related IFFs often represent substantial amounts of lost or forgone tax revenue that could have otherwise been spent on reducing inequality and fighting poverty or fostering the structural transformation of their economies amid global challenges such as climate change. IFFs may also drain foreign exchange reserves, undercut legitimate trade and jeopardise the business environment. IFFs also undermine the investment base of a country. When capital leaves countries it undercuts their ability to finance development goals and invest in their economies. This may result in reduced scope and quality of public services, reduced public transfers or job creation, and ultimately, reduced confidence in public institutions.

Apart from the financial aspect, IFFs also undermine the trust of citizens. By evading taxes and channelling financial flows illicitly across borders, IFFs may undermine tax morale which may result in further erosion of the tax base, resulting in lower domestic tax compliance along with reduced cross-border tax compliance. IFFs often allow the proceeds of corruption and crime to flow overseas making it more challenging for authorities to tackle crime and reduce corruption. Assessing IFFs and successfully combatting financial leakages thus can not only improve domestic resource mobilisation but also support a government’s social contract with its citizens and help achieve a more favourable business climate for private sector investment.

Despite definitional challenges, because IFFs pose such significant risks to countries, there have been several attempts to estimate illicit flows on an international scale. The estimates are as heterogeneous as the methodologies applied and the underlying activities assumed to result in IFF-generating proceeds. For instance, in their most recent study based on trade data, GFI (2021[6]) estimate that total illicit flows from developing countries reached USD 1.6 trillion in 2018. While developing Asia accounted for around USD 390 billion or 25% of all flows, the African continent lost around USD 84 billion, accounting for 5% of global IFFs. For 2000 to 2008, AU/ECA (2015[7]) estimate cumulative IFFs from Africa due to trade mispricing of USD 162 billion, exhibiting an increasing trend over time. Other global estimates on criminal proceeds involving money laundering in 2009 amount to USD 2.1 trillion, an equivalent of about 3.5% of global GDP (UNODC, 2011[8]).

Other recent studies focus more on the global scale of individual tax evasion and offshore wealth. O’Reilly et al. (2019[9]) conduct an analysis into tax evasion and global offshore wealth in international financial centres (IFCs). They find that, based on bank account data from the BIS, global offshore wealth peaked at USD 2.5 trillion in 2008 and fell by around 42% to around USD 1.4 trillion until 2019. This drop in offshore deposits has largely been associated with the global expansion of tax transparency. Another study by Alstadsaeter et al. (2018[10]) reports that the stock of hidden offshore wealth has remained equal to about 10% of world GDP between 2001 and 2007. This order of magnitude corresponds to non-compliant foreign wealth of about USD 5.6 trillion in 2007.

Results for individual countries such as South Africa have very often been derived from the global or regional estimates. For instance, GFI (2021[6]) estimate average illicit outflows of around USD 20 billion for South Africa every year during 2009 - 2018. AU/ECA (2015[7]) suggests that South Africa’s annual IFFs are around USD 14 billion, measured in terms of GDP and relative to financial market size. According to Signé et al. (2020[11]), the country emitted a total of USD 441.5 billion during the period from 1980 to 2018. This would result in average annual outflows of about USD 12 billion. Related to individual tax evasion, Alstadsaeter et al. (2018[10]) estimate South African total individual wealth in offshore destinations to amount to about USD 43.5 billion – an equivalent of 11.8% of South Africa’s GDP in 2007.

However, the lack of a commonly agreed definition, inherent measurement challenges and the reliance on aggregated data render most IFF estimates imprecise and result in relatively high numbers. Moreover, studies following different approaches may estimate IFFs in ways that may or may not overlap. For example, studies based on trade-mispricing may also measure some IFFs associated with money laundering or tax evasion. Being aware of these challenges is therefore vital when trying to estimate and to interpret any results on IFFs.

Measurement challenges in relation to IFFs arise in particular through their hidden nature. Due to being inherently secret, most conventional strategies to quantify IFFs by using existing data as, for instance, through surveys of households, businesses or financial institutions are unlikely to reveal IFFs. While some IFFs are remitted illicitly through the financial system, other IFFs (e.g. cross-border smuggling of cash) are not and need to be considered separately. Associating payments that are not illegal in their method of transfer with either domestic crime or corruption is challenging, as is associating payments that may have been earned and transferred legally with illicit use. As a result, IFF estimates often rely on proxies for illicit activity to estimate the size of IFFs. These approaches seldom allow for disaggregation of different kinds of IFFs into any of the subcategories in any of the frameworks discussed above.

The most commonly-cited measures of IFFs are generated using top-down estimation methods. This involves measuring IFFs using errors and asymmetries in aggregated macroeconomic data derived from global trade statistics, portfolio investment statistics or statistics on foreign direct investment. By exploiting mismatches in bilateral data series, however, these approaches often conflate statistical measurement problems with IFFs, potentially resulting in inflated numbers. For example, a common approach to measuring IFFs from trade mis-invoicing is to compare import data as recorded by Country A from Country B with export data as recorded by Country B to Country A. Where differences between these data exist, it is these asymmetries that are assumed to derive largely from IFFs. However, such discrepancies can also result from differences in the timing of imports and exports in statistics, differences in the categorisation of imports and exports by different countries, and simple measurement error (Collin, 2020[12]).

Some researchers have suggested that estimates based on mirror trade statistics may be unreliable due to these issues. In general, these challenges hail from the availability of suitable data, the overly general or mixed methods that are applied, the arbitrary and/or simplifying nature of the assumptions, the aggregation of categories, and a lack of the country-specific context when analysing and measuring IFFs (Nitsch, 2016[13]). Moreover, IFFs for individual countries are often simply derived from the global aggregate by accounting for a country’s share in global trade or the regional economy as shown above.

In contrast, bottom-up methods to estimate IFFs rely on more granular micro-data, for instance from tax audits or individual tax returns. These methods try to identify the different individual IFF components and produce estimates for each, which may then be aggregated horizontally or from a lower to a higher level to obtain a national estimate. While they also differ in their degree of sophistication, the resulting accuracy of any estimate may be considered higher compared to the top-down estimation due to the reliance on microdata and activities that are directly related to the generation of IFFs.2

The second major challenge in measuring IFFs emerges from a lack of capacity or weak institutions to identify them in the first place. Developing countries, which very often face very significant IFF challenges, may lack reporting mechanisms or enforcement capacities to successfully detect and combat illicit outflows. For instance, the increasing reliance on tax data for the analysis of compliance demands data and analysis approaches that may strain the capacity of developing country tax administrations (Kennedy, 2019[14]). Developing countries have also been slower to benefit from progress in tax transparency, particularly due to resource constraints. This is particularly the case with respect to the CRS for AEOI, where many developing countries have not been able to put in place the required institutional processes to qualify to receive information. Out of the group of 108 jurisdictions that have committed to implement the standard by 2021, 32 are developing countries and 22 of them have started to reciprocally exchange to date (OECD, 2021[15]).

IFFs thus need to be analysed and measured in a country-specific context. They need to be considered in light of an economy’s economic structure as well as its institutional environment and capacities. Differences therein not only determine IFF categories relevant in the specific context and their respective size but also the capacity of national authorities to effectively assess and combat any illicit outflows. Moreover, a country-specific perspective allows for a more tailored approach to estimating IFFs, likely using the kind of data best-suited to exploring a relevant IFF-generating activity. For these reasons while the estimates derived in Chapter 6 are estimates of all IFFs, much of the policy discussion in the present study concentrates on individual tax-related illicit flows as one major subcomponent of South Africa’s IFF landscape.

IFF risks in South Africa largely stem from its economic structure and geographic position. According to the recently concluded evaluation by FATF as well as consultations with national authorities, IFFs in South Africa are still considered a major threat to tax revenues and the overall economy – despite significant efforts undertaken and progress achieved.3 IFFs can generally be divided into two large categories, namely flows pertaining to individuals and flows more closely related to corporates. While individual IFFs, for instance, consist to a large extent of direct transfers of funds, for example as the proceeds from illegal activities or funds wired offshore for the purpose of tax evasion, corporate IFFs are largely related to illegal trade activities. Inter-relationships and mutually reinforcing effects between the different categories exist and make them even harder to detect, demanding a more systemic view from authorities to combat IFFs successfully.

Tax offences with regard to income taxes have been characterised as being one of the most predominant forms of tax evasion (FATF, 2021[16]). Against this background, this report assesses one of the main elements of South Africa’s IFF environment. The main domestic proceeds-generating crimes in South Africa have been identified as tax crimes, corruption and bribery, fraud, those linked to credit card abuse, and environmental and resource type crimes. Tax crimes encompass evasion of a broad range of taxes and fees, including corporate and personal income and customs and excise taxes as well as tax fraud (e.g. VAT fraud). Fraud includes Ponzi schemes, other investment, cyber-fraud, and digital banking frauds as well as those involving virtual assets. Criminal proceeds from these activities, income disguised as donations or gifts, or legal income earned are often illicitly channelled outside the country and into IFCs thereby contravening exchange control regulations. Given the criminality associated with the manner in which they are generated, such activities are unlikely to be tax compliant. They may also involve foreign investment allowances, crypto assets, financial emigration or funds that are transferred through Common Monetary Area (CMA) countries.

South Africa’s geographic position as a key financial hub in the Southern African region and a relatively developed economy when compared to its neighbours potentially exposes it to the threat of foreign proceeds of crime, resulting in illicit in- and outflows. While funds can be laundered in or through South Africa from the region due to its relatively large financial sector, the country has also been described as a potential market and a transit point for trafficking in illicit goods, such as illicit drugs, or people smuggling. Corporate IFFs on the other hand are mostly related to trade mis-invoicing, commodities smuggling, transfer mispricing or advance payments (FATF, 2021[16]).4

Since the global financial crisis, the South African authorities have joined a number of multilateral tax transparency initiatives. While these initiatives have been part of the global effort to enhance tax compliance with respect to foreign source income, most of them have also been undertaken with the goal of mobilising government revenues. Most initiatives are thus related to countering tax-related IFFs such as tax evasion on a multilateral basis by facilitating the exchange of information (EOI) for tax purposes.

South Africa has been at the forefront on the African continent to join the global fight against tax-related IFFs. In 2009, the country became an early member of the Global Forum on Transparency and Exchange of Information for Tax Matters (the Global Forum). Signing the Convention on Mutual Administrative Assistance in Tax Matters (MAAC) in November 2011 expanded South African’s information exchange network substantially. Established in 1988 and amended by Protocol in 2010 to allow for broader country participation, the MAAC not only provides for bilateral and multilateral EOI (including spontaneous exchange, EOIR, and AEOI5), it also includes assistance in recovery, the service of documents and can facilitate joint audits among its signatories. Moreover, the MAAC is not only a valuable tool for fighting tax evasion; it also has the potential to support other law enforcement purposes such as fighting corruption and money laundering (OECD/Council of Europe, 2011[21]). Over 140 jurisdictions have already joined the convention, broadening South Africa’s network of exchange of information partners. Following the entry into force of the MAAC in early March 2014, South Africa also publicly committed to the early adoption of AEOI under the Common Reporting Standard (CRS) in May 2014. Joined by 44 other jurisdictions, the country announced the activation of automatic information exchange by 2017 with the first batch of taxpayer information having been exchanged in January the same year. As at March 2022, South Africa has 105 activated CRS agreements for partner jurisdictions to send CRS information.

Apart from global initiatives, increasing mutual assistance in tax matters in a developing country and regional context has become an important element in fighting tax evasion. Even prior to the introduction of the MAAC, the African Tax Administration Forum (ATAF) developed an EOI agreement open to signature by its members in 2012. The Agreement on Mutual Assistance in Tax Matters (AMATM) was set up to mutually provide ATAF member countries with tax information on a spontaneous or automatic basis or upon request.6 Although twelve African countries have already signed the AMATM (Botswana, Eswatini, Gambia, Ghana, Lesotho, Liberia, Malawi, Mozambique, Nigeria, South Africa, Uganda and Zambia), and it has entered into force following ratification by four countries as required by the Agreement, only seven countries have so far ratified the AMATM (Gambia, Lesotho, Liberia, Mozambique, Nigeria, South Africa and Uganda). Moreover, on 17 August 2013, South Africa also signed a similar agreement between member countries of the Southern African Development Community (SADC), the Southern African Development Community’s Agreement on Assistance in Tax Matters (SADCA). This agreement is not in force yet (AU/ATAF/OECD, 2021[22]).

South Africa has also been a founding member of the Africa Initiative, a programme created by the Global Forum to enhance EOI capacities among African countries. The Africa Initiative was launched in 2015 as a partnership between the Global Forum, its African members and a number of regional and international organisations and development partners: African Tax Administration Forum, Cercle de Réflexion et d’Échange des Dirigeants des Administrations Fiscales, World Bank Group, France (Ministry of Europe and Foreign Affairs) and the United Kingdom (Foreign, Commonwealth & Development Office). Since then, additional partners (African Development Bank, African Union Commission, European Union, Norway, Switzerland and the West African Tax Administration Forum) joined the initiative. The Africa Initiative is open to all African countries and currently has 33 African member jurisdictions. The Africa Initiative’s work fits into broader agendas, as tax transparency is an opportunity to stem illicit financial flows and increase domestic resource mobilisation, which are central to the African Union Agenda 2063 and the Sustainable Development Goals (AU/ATAF/OECD, 2021[22]). Since 2021, South Africa has been the Vice-Chair of the Initiative.

The Africa Initiative has resulted in significant engagement at the ministerial level to ensure political buy-in and sustained momentum in the area of tax transparency. Initially set up for a period of three years (2015-2017), the Initiative was renewed for a second phase (2018-2020) in 2017 and for a third phase (2021-2023) in 2020. During the plenary a call for action was made through a landmark document, the "Yaoundé Declaration" which urges the African Union to begin a high-level discussion on tax cooperation and illicit financial flows and their link to domestic resource mobilisation. Supported by the OECD and the governments of France and the United Kingdom since its inception, this call for action to increase tax compliance, fight IFFs and foster domestic resource mobilisation has been signed by 33 signatories, including the African Union Commission. South Africa signed the declaration in September 2018 (OECD, 2021[23]).

The active exchange of information (EOI) between tax authorities can be considered the most effective way to enquire about offshore wealth and thus address tax evasion. The success of international tax co-operation, however, relies on the effective implementation of international standards. Successful implementation of EOI standards requires governments to ensure the availability of legal ownership information, beneficial ownership, accounting and banking information, the access to that information and its effective exchange with foreign partners, based on international agreements in force. The use of available exchange agreements in turn provides the government with a more comprehensive picture of its taxpayers’ international financial affairs, their degree of compliance with domestic tax law and supports the fight against tax-related IFFs.

Adherents to the two internationally agreed standards on the exchange of information on request (EOIR) and of individual financial account information automatically (AEOI), are regularly monitored by the Global Forum to ensure effective implementation. As an early adherent to both standards, the country’s institutional setting for EOI has been reviewed several times in the past. South Africa has implemented the necessary institutional requirements for successful implementation and operation of EOI (AU/ATAF/OECD, 2021[22]). The country’s tax administration has an EOI strategy implemented. The EOI infrastructure in the form of a Competent Authority delegation, a dedicated EOI unit and staff as well as documentation procedures with database tracking tools have been in place. Moreover, revenue gains through EOI are monitored and additional taxes collected show that EOI has had a positive impact on revenue mobilisation in South Africa.

Relative to most other African countries, South Africa’s EOI network is wide with the possibility of exchanging information with more than 150 jurisdictions (OECD, 2021[24]). Over time, the country has entered into a large number of bilateral agreements, including through regional initiatives in Africa. Of the currently more than 100 bilateral agreements, the earliest entered into force in 1956 with Zambia (Figure 3.3). Since the mid-1990s the number of overall EOIR agreements has increased strongly, experiencing another surge after the global financial crisis due to the MAAC. The network of EOIR agreements with IFCs expanded later and has only recently experienced a substantial rise in bilateral relationships.

South Africa has been reviewed on its implementation of the EOIR standard twice by the Global Forum, in 2012 and 2021. In 2012, South Africa was found compliant overall with regard to the 2010 Terms of Reference for both the legal implementation of the EOIR standard as well as its operation in practice (OECD, 2012[25]). Two recommendations were issued with regards to further developing its EOIR network and to better monitoring the availability of identity information related to partnerships. Since then, South Africa has implemented the first recommendation and now has a broad EOIR network, especially with the entry into force of the MAAC in 2014. Despite its sizeable network, however, requests through bilateral agreements have only been submitted sporadically to other jurisdictions.

In 2021, the new EOIR review has stressed the need for continued progress to ensure availability of beneficial ownership information as described in the second round of EOIR reviews. Due to a strengthening of the standard in 2016, the second review concludes that overall South Africa has a legal and regulatory framework “in place but needs improvement” regarding the availability of ownership, accounting and banking information as well as the quality and completeness of outgoing EOI requests (OECD, 2021[24]). In particular, improvements have been recommended with regards to the availability of regularly updated beneficial ownership information on all bank accounts and all relevant entities and arrangements. It also recommends South Africa to ensure that accounting information is available for a period of five years for all relevant companies, including those that redomicile out of the country.7

Following South Africa’s public commitment to an early adoption of AEOI in 2014, exchanges have been taking place on an annual basis since 2017. Through the wide-spread adoption of the MAC, the global network of AEOI relationships has significantly increased within a relatively short period of time. South Africa signed the MAAC in 2014 and activated the associated CRS Multilateral Competent Authority Agreement (MCAA) in time for the first exchanges in 2017. To allow domestic financial institutions to collect and report the information to be exchanged, South Africa also amended the Tax Administration Act of 2011 and introduced CRS regulations (OECD, 2020[26]). As a result, South Africa currently has automatic information exchanges activated with 105 jurisdictions, 15 among them are in Africa, and more jurisdictions are expected to join the global network in the near future.

South Africa’s legal framework for AEOI implementation was recently assessed during the AEOI peer review in November 2021 (OECD, 2021[27]). The quality of its confidentiality and data safeguards framework to ensure confidential exchange of data without the direct or indirect revelation of any taxpayer information under the CRS was assessed as fully compliant. The country’s legal framework implementing the AEOI Standard was found to be in place and fully consistent with the requirements of the AEOI Terms of Reference. This includes South Africa’s domestic legislative framework requiring Reporting Financial Institutions to conduct the due diligence and reporting procedures (core requirement 1) and its international legal framework to exchange the information with all of South Africa’s Interested Appropriate Partners (core requirement 2). The Global Forum is currently reviewing South Africa’s effectiveness in relation to AEOI in practice together with other jurisdictions with results due to be published by the end of 2022. Capacity development is progressing with 48 South African tax officials having participated in Global Forum training on EOI in 2020 (AU/ATAF/OECD, 2021[22]).

During the last decade, voluntary disclosure programmes (VDPs) have become an internationally accepted practice to broaden the tax base and collect additional tax revenues across the world. Since 2009, over 90 VDPs have been implemented in more than 25 countries (O’Reilly, Parra Ramirez and Stemmer, 2019[9]). The programmes often differ significantly in terms of length and legal consequences of application. Moreover, the domestic implementation of voluntary disclosure has in the past often coincided with the signature of bilateral EOIR treaties or the commencement of AEOI.8 As a result, many VDPs have encouraged taxpayers to declare non-compliant offshore deposits or repatriate hidden assets. Others have primarily targeted non-compliance with regards to other tax laws. South Africa has followed this path as well and introduced a total of 3 different VDPs since 2010. Depending on their targeted taxpayers and tax offences, the programmes have either been administered by SARS alone for tax defaults or in collaboration with SARB when dealing with contraventions of exchange control regulation due to undeclared offshore assets or foreign capital income.9

South Africa’s first VDP was introduced on 27 October 2010 and lasted until 31 October 2011. The programme had a tax component, administered by SARS, and an exchange control component, administered by SARB’s financial surveillance department. Taxpayers could thus either regulate their tax affairs, declare exchange control contraventions, or both by applying simultaneously to all programme components. Any tax default or exchange control contravention had to have occurred before the current tax year. While the main prerequisite for successful applications for tax defaults was that the information was not previously disclosed to SARS, SARB required detailed information of foreign assets and structures, their market value and whereabouts.10 Successful applicants to SARS benefited from interest, penalty and additional tax relief; SARB did not impose administrative penalties. Criminal investigations were not initiated after full tax payment or full disclosure of contraventions.

One year later, SARS introduced a permanent VDP administered under the Tax Administration Act 28 of 2011 with effect from 1 October 2012. The general purpose of the programme is to enhance voluntary compliance in the interest of good management of the tax system and the best use of SARS resources (Republic of South Africa, 2012[28]). The VDP aims to encourage taxpayers to come forward on a voluntary basis to regularise their tax affairs and avoid the imposition of understatement penalties or other administrative fines. Penalties for late submission of returns are charged. Criminal prosecution will also not be initiated. The VDP is applicable to all taxes administered by SARS except for Customs and Excise. Both individuals and companies may apply to the programme provided they fulfil certain qualifying conditions.

As with the previous VDP, voluntary disclosure is required for a successful application. In the case of outstanding tax returns of which SARS has already been aware or where an investigation was already underway at the date of application, an application would be rendered void. Additional conditions include that the application must be formally completed and the application cannot result in a refund to the taxpayer. Upon approval of the application, an agreement is concluded between SARS and the applicant that reflects the outcome of the application process. A successful VDP assessment gives effect to the VDP agreement, and typically includes the disclosed additional taxable income and, depending on the length of time since the default occurred, interest and late payment penalties. In addition, the VDP allows for prospective applicants to make an initial “no-name” disclosure in order to obtain a non-binding opinion from SARS as to whether the applicant would qualify for relief and the extent of any relief.

Alongside the VPD, National Treasury proposed a Special Voluntary Disclosure Programme (SVDP) in October 2016. Taxpayers with undisclosed offshore assets and foreign income only had a limited window for submissions from 1 October 2016 to 31 August 2017 to declare their contravention of exchange control regulation and tax default and apply for relief. With AEOI commencing in September 2017, the SVDP was implemented as a last-minute opportunity for non-compliant taxpayers to disclose their assets while at the same time using the threat of global detection to encourage applications. Taxpayers who missed this deadline could still make use of the normal VDP process to disclose offshore income.

The programme was administered by SARS and SARB and relief was offered relative to their competencies. The undeclared income that originally gave rise to the foreign asset was to be exempt from income tax, donations tax and estate duty liabilities that had arisen in the past. Only 40% of the highest value of the aggregate of all assets situated outside South Africa between (or deemed to be between) 1 March 2010 and 28 February 2015 that were derived from undeclared income were be included in taxable income and subject to tax in South Africa in the 2015 tax period. Investment earnings and other taxable events prior to 1 March 2015 were exempt from tax. Interest on tax debts arising from the disclosure only commenced from the 2015 year of assessment. Moreover, SARB charged different levies in contravention of exchange control regulations. Depending on a repatriation of funds, a 5% levy on the value of non-declared assets was imposed if assets were repatriated, with a 10% levy applying in case of non-repatriation. Levies had to be paid from foreign sourced funds. A levy of 12% arose if the 10% could not be paid from foreign sourced funds. Compared to previous VDPs, similar provisions applied with regards to criminal prosecution and the imposition of fines or administrative penalties (SARS, 2016[29]).

Apart from increasing overall taxpayer compliance, high-net worth individuals (HNWIs) have been identified by SARS as important contributors to the South African economy but also as a potential source for IFFs. Their relatively high wealth and complexity of tax affairs may provide opportunities for aggressive tax planning and cross-border tax evasion. For these reasons, SARS established a stand-alone unit to monitor HNWI’s compliance in 2008/09. Over time, the number of very wealthy individuals in South Africa’s society has increased, together with their reported income. For instance, based on SARS statistics, Hundenborn, Woolard and Jellema (2019[30]) report an increase from 482 individuals with taxable income of more than ZAR 10 million in 2011 to 1048 individuals in 2014. During the same period, the maximum declared taxable income rose from ZAR 110 million to over ZAR 150 million.

Given their importance and potential revenue contribution, SARS attempted to provide a differentiated approach (treatment strategy) to ensure voluntary compliance. Individuals were classified as a HNWI based on a formal definition setting out a qualifying criteria. Since then, SARS’ monitoring of this income tax segment has changed over time and various organisational formats have been tested. In 2015/16, the HNWI Unit was entirely disbanded with very few audits taking place in that year and the following year (Zondo Commission, 2022[31]).

After the National Treasury’s announcement in February 2021, the High Wealth Individual (HWI) Unit re-started its operations with the aim to improve compliance of wealthy individuals. The new HWI Unit will have a limited focus on individuals with assets worth at least ZAR 75 million. At present its tax base consists of 1408 individuals (AU/ATAF/OECD, 2021[22]). Subsequently the SARS Commissioner announced the establishment of the HWI Taxpayer Segment to improve compliance of the identified individuals through a one-stop, superior service. The focus of the restructured HWI segment is to increase voluntary compliance of HWI taxpayers by providing clarity and certainty and making it easier to comply with their obligations, while at the same time improving SARS’ ability to detect and respond to risks posed by the taxpayer segment. Having been receiving AEOI data on foreign financial accounts since 2017, EOI is expected to play a key role in the detection of possible non-compliance in relation to cross-border transactions and assets held offshore.


[10] Alstadsaeter, A., N. Johannesen and G. Zucman (2018), “Who owns the wealth in tax havens? Macro evidence and implications for global inequality”, Journal of Public Economics, Vol. 162, pp. 89-100, https://doi.org/10.1016/j.jpubeco.2018.01.008.

[32] Andersson, J., F. Schroyen and G. Tosvik (2019), The impact of international tax information exchange agreements on the use of tax amnesty: evidence from Norway, https://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID3463645_code1391513.pdf?abstractid=3463645&mirid=1.

[22] AU/ATAF/OECD (2021), Tax Transparency in Africa 2021 - Africa Initiative Progress Report, OECD Publishing, https://www.oecd.org/tax/transparency/documents/Tax-Transparency-in-Africa-2021.pdf.

[7] AU/ECA (2015), Track it! Stop it! Get it! Report of the High Level Panel of Illicit Financial Flows from Africa, https://repository.uneca.org/bitstream/handle/10855/22695/b11524868.pdf?sequence=3&isAllowed=y.

[1] Baker, R. (2005), Capitalism’s Achilles Heel: Dirty Money and How to Renew the Free-Market System, Wiley.

[12] Collin, M. (2020), “Illicit Financial Flows: Concepts, Measurement, and Evidence”, The World Bank Research Observer, Vol. 35/1, pp. 44-86, https://doi.org/10.1093/wbro/lkz007.

[5] European Commission (2017), Aggressive tax planning indicators – Final Report, Publications Office of the European Union, https://doi.org/10.2778/847061.

[16] FATF (2021), Anti-money laundering and counter-terrorist financing measures – South Africa, Fourth Round Mutual Evaluation Report, FATF, http://www.fatf-gafi.org/publications/mutualevaluations/documents/mer-south-africa-2021.html.

[6] Global Financial Integrity (2021), Trade-Related Illicit Financial Flows in 134 Developing Countries: 2009-2018, https://secureservercdn.net/

[30] Hundenborn, J., I. Woolard and J. Jellema (2019), “The effect of top incomes on inequality in South Africa”, International Tax and Public Finance, Vol. 26, pp. 1018-1047, https://doi.org/10.1007/s10797-018-9529-9.

[18] Johannesen, N., T. Torslov and L. Wier (2020), “Are Less Developed Countries More Exposed to Multinational Tax Avoidance? Method and Evidence from Micro-Data”, The World Bank Economic Review, Vol. 34/3, pp. 790-809, https://doi.org/10.1093/wber/lhz002.

[14] Kennedy, S. (2019), “The potential of tax microdata for tax policy”, OECD Taxation Working Papers, No. 45, OECD Publishing, Paris, https://dx.doi.org/10.1787/d2283b8e-en.

[13] Nitsch, V. (2016), Trillion dollar estimate: Illicit financial flows from, http://nbn-resolving.de/urn:nbn:de:tuda-tuprints-54379.

[9] O’Reilly, P., K. Parra Ramirez and M. Stemmer (2019), “Exchange of information and bank deposits in international financial centres”, OECD Taxation Working Papers, No. 46, OECD Publishing, Paris, https://dx.doi.org/10.1787/025bfebe-en.

[15] OECD (2021), Global Forum on Transparency and Exchange of Information for Tax Purposes - Reinforcing Multilateral Co-operation in Tax Matters for a Fair and Inclusive Recovery (2021 Global Forum Annual Report), OECD Publishing, https://www.oecd.org/tax/transparency/documents/global-forum-annual-report-2021.pdf.

[24] OECD (2021), Global Forum on Transparency and Exchange of Information for Tax Purposes: South Africa 2021 (Second Round, Phase 1): Peer Review Report on the Exchange of Information on Request, Global Forum on Transparency and Exchange of Information for Tax Purposes, OECD Publishing, Paris, https://dx.doi.org/10.1787/fed716dd-en.

[27] OECD (2021), Peer Review of the Automatic Exchange of Financial Account Information 2021, OECD Publishing, Paris, https://dx.doi.org/10.1787/90bac5f5-en.

[23] OECD (2021), The Yaoundé Declaration, OECD Publishing, https://www.oecd.org/tax/transparency/what-we-do/technical-assistance/Yaounde-Declaration-with-Signatories.pdf.

[26] OECD (2020), Peer Review of the Automatic Exchange of Financial Account Information 2020, OECD Publishing, Paris, https://dx.doi.org/10.1787/175eeff4-en.

[17] OECD (2013), Action Plan on Base Erosion and Profit Shifting, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264202719-en.

[25] OECD (2012), Global Forum on Transparency and Exchange of Information for Tax Purposes -  Peer Reviews: South Africa, OECD Publishing, Paris, https://dx.doi.org/10.1787/2219469x.

[21] OECD/Council of Europe (2011), The Multilateral Convention on Mutual Administrative Assistance in Tax Matters: Amended by the 2010 Protocol, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264115606-en.

[28] Republic of South Africa (2012), Government Gazette, 4 July 2012, https://www.gov.za/sites/default/files/gcis_document/201409/a282011.pdf.

[29] SARS (2016), External Guide - Special Voluntary Disclosure Programme, https://www.sars.gov.za/wp-content/uploads/Ops/Guides/GEN-VDP-02-G02-Special-Voluntary-Disclosure-Programme-External-Guide.pdf.

[11] Signé, L., M. Sow and P. Madden (2020), Illicit financial flows in Africa: Drivers, destinations, and policy options, https://www.brookings.edu/wp-content/uploads/2020/02/Illicit-financial-flows-in-Africa.pdf.

[2] UNECA (2013), The State of Governance in Africa: The Dimension of Illicit Financial Flows as a Governance Challenge. Third ECA Meeting of the Committee on Governance and Popular Participation (CGPP), United Nations Publishing, https://repository.uneca.org/handle/10855/22142.

[8] UNODC (2011), Estimating illicit financial flows resulting from drug trafficking and other transnational organized crimes, https://www.unodc.org/documents/data-and-analysis/Studies/Illicit_financial_flows_2011_web.pdf.

[4] UNODC/UNCTAD (2020), Conceptual Framework for the Statistical Measurement of Illicit Financial Flows, https://www.unodc.org/documents/data-and-analysis/statistics/IFF/IFF_Conceptual_Framework_for_publication_15Oct.pdf.

[20] Wier, L. (2020), “Tax-motivated transfer mispricing in South Africa: Direct evidence using transaction data”, Journal of Public Economics, Vol. 184, pp. 1-16, https://doi.org/10.1016/j.jpubeco.2020.104153.

[19] Wier, L. and H. Reynolds (2018), Big and “unprofitable”, https://www.wider.unu.edu/sites/default/files/Publications/Working-paper/PDF/wp2018-111.pdf.

[3] World Bank (2017), Domestic Resource Mobilization and Illicit Financial Flows - Board Update, https://documents1.worldbank.org/curated/en/877291492623853466/pdf/IFFs-DRM-Board-Note-Master-vFINAL-04102017.pdf.

[31] Zondo Commission (2022), Judicial Commission of Inquiry into State Capture Report: Part 1 - Volume 1: South African Airways and its Associated Companies, https://www.statecapture.org.za/site/files/announcements/638/Judicial_Commission_of_Inquiry_into_State_Capture_Report:_Part_1_Vol._1:_SAA_(18_MB).pdf.


← 1. This means a significant reduction of IFFs as stated in SDG target 16.4.

← 2. Consult, for instance, Collin (2020[12]) for a more detailed view on IFF measurement challenges and Kennedy (2019[14]) on the importance of microdata.

← 3. Comments by participants expressed during consultations on South Africa’s IFF risks with National Treasury, SARS, the Financial Intelligence Unit, the Reserve Bank, as well as FATF and academics are gratefully acknowledged.

← 4. Transfer mispricing has been analysed in detail from the South African perspective by Wier (2020[20]).

← 5. This standard in AEOI requires the automatic exchange each year of information collected and reported by Financial Institutions on the financial accounts and assets they maintain for non-resident taxpayers.

← 6. The text to the Agreement is available at: https://events.ataftax.org/includes/preview.php?file_id=46&language=en_US.

← 7. As the COVID-19 pandemic prevented the holding of any onsite visits to South Africa, the 2021 review is limited to the adequacy of the legal and regulatory framework of South Africa to the standard of transparency and exchange of information on request. The review of the implementation of that framework in practice is scheduled to take place in 2022.

← 8. Norway, for instance, implemented a VDP prior to tax exchange agreements with Luxembourg and Switzerland (Andersson, Schroyen and Tosvik, 2019[32]).

← 9. Details of the impact of these programmes are discussed in Chapter 5.

← 10. Information has been retrieved from: https://www.sars.gov.za/wp-content/uploads/Docs/MediaReleases/2010/SARS-MR-2010-053-Media-release-on-SARS-and-SARB-for-the-Voluntary-Disclosure-Programme-5-November-2010.pdf.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2022

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at https://www.oecd.org/termsandconditions.