- 1 See Musselli and Bürgi Bonanomi (2020) for a discussion of different conceptualisations of IFFs. Th (...)
1Illicit financial flows (IFFs) are cross-border transfers of money or assets that have an illegal origin or are illegally transferred or used in at least one of the jurisdictions concerned.1 They can be related to income generation as well as income management. The former refers to illicit cross-border activities that produce income, including, for instance, the shipment of illegal drugs. The latter includes illicit transactions that transfer funds that have a licit origin, as well as flows stemming from licit activity that are used in an illicit way (see UNCTAD and UNODC, 2020, 7–8). IFFs are known to severely affect the fiscal space and drain public revenues of many low- and middle-income countries through a variety of mechanisms (Mbeki et al., 2015).
2In this chapter, we argue that a specific element of public finance systems—the use of tax expenditures (TEs)—can play an important role as an enabler or facilitator of IFFs. The term ‘tax expenditure’ was introduced by the US Treasury Department in the late 1960s to denote preferential tax treatments that deviate from the benchmark or standard tax system and are applied to benefit specific groups, activities or economic sectors (see Surrey and McDaniel, 1979). Since 1975, the US has prepared annual TE reports linked to the federal budget, and a growing number of countries worldwide have been following suit (Redonda, von Haldenwang and Aliu, 2023).
3TEs can be applied to income taxes—and can include corporate tax incentives to attract investment—but also to consumption taxes, wealth taxes or customs duties. They can take many forms, including exemptions, reduced rates, deductions, deferrals and tax credits. TEs are legal by definition, though their effectiveness and efficiency as mechanisms for promoting specific public policies are often questionable, as shown by a growing body of evaluations and academic research (James, 2013; Klemm and Van Parys, 2012; Kronfol and Steenbergen, 2020; Myles et al., 2014; Redonda and Axelson, 2021). The Global Tax Expenditures Database (GTED; see von Haldenwang et al., 2023) provides an overview, based on official figures, of the worldwide use of TEs, showing that they are certainly not a minor issue, worldwide average revenue forgone amounting to 3.7 per cent of GDP and a staggering 23.5 per cent of taxes actually collected. Yet to date, little research has been conducted on the role of TEs with regard to IFFs.
- 2 This chapter uses the term ‘source countries’ to refer to countries where IFFs originate, while ‘re (...)
4The present chapter explores conceptual approaches to and empirical evidence of the manifold linkages between TEs and IFFs. It focuses on three kinds of TEs that have recently attracted scholarly attention. First, in countries that suffer from illicit financial outflows (‘source countries’)2 we look at special economic zones (SEZs). SEZs are geographical areas within countries where companies benefit from specific tax regimes, such as exemptions from import duties or corporate taxes (Heitmüller and Mosquera, 2021; UNCTAD, 2019).
5Second, we examine TEs related to the extractive industries. Given the long timelines of extractive projects and the often sizeable up front investments involved, governments frequently use TEs to attract investment to these industries. In many resource-dependent countries, TEs are not only applied through the tax code, but also through exploration or extraction contracts between governments and multinational corporations, which often contain (tax) stabilisation clauses and are characterised by limited transparency (Le Billon, 2011; Otto, 2017; Readhead, 2018).
6A third kind of TEs is sometimes used in countries that receive IFFs, either as final recipients or as ‘transit countries’ (Musselli and Bürgi Bonanomi, 2020). These TEs target intangible assets, in particular through so-called patent boxes (PBs) and other TE provisions for research and development (R&D) (Alstadsæter et al., 2018; Haufler and Schindler, 2022).
7To link these various TEs to the topic of this thematic volume, we base our argument on Carbonnier and Mehrotra’s (2018) distinction between the ‘push’ and ‘pull’ factors driving IFFs. The former refers above all to the existence of conditions that facilitate the illicit outflow of funds from source countries, such as weak governance structures, high levels of corruption, and a lack of oversight, transparency and monitoring. Pull factors, meanwhile, are primarily located in countries that receive such funds, and include financial secrecy and special tax incentives granted to inflowing capital. Obviously, corruption and weak governance structures can play a role in receiving countries as well, for instance by facilitating regulatory capture on behalf of powerful corporate interests (Laffont and Tirole, 1991).
8Related to the three kinds of TEs mentioned above, the chapter also discusses the relevance of bilateral tax treaties, which ‘set boundaries on when and how each country is entitled to tax income earned in one treaty partner by residents of the other, most usually multinational companies’ (Hearson, 2018, 235). Regulations typically refer to withholding taxes on dividends, interest payments, royalties, service fees, etc. that a source country can levy on residents of the treaty partner. In addition, treaties often cover permanent establishment (PE) criteria and the tax treatment of particular types of income such as capital gains, pensions, social security payments and salaries (Hearson, 2018). The Tax Treaty Explorer database counts more than 3,000 such treaties currently in force worldwide (Hearson, 2022). The specific provisions contained in tax treaties are frequently omitted from governmental TE reports. This lack of transparency hinders comprehensive evaluation and facilitates the use of treaties for aggressive tax planning and IFFs (Janský and Šedivý, 2019; Beer and Loeprick, 2021).
9The present chapter identifies numerous ways in which TEs can in fact lead to an increase in IFFs. Generally speaking, the lack of transparency associated with the use of TEs in many countries seems to be a key enabling factor. However, the chapter also shows that, beyond anecdotal evidence, the evidence base of causality statements in this context is rather thin. Endogeneity concerns and a lack of relevant data explain this picture. The remainder of the chapter is organised as follows: Section 2 introduces three main conceptual mechanisms that link the use of TEs to IFFs. Section 3 takes a closer look at TEs in source countries while Section 4 focuses on TE use in recipient countries. Section 5 concludes by discussing the policy implications of the chapter’s findings in light of recent international developments, most notably the Global Anti-Base Erosion (GLOBE) approach introduced by the OECD (OECD, 2022a).
10TEs are sometimes employed as a defensive mechanism in international tax competition, whereby governments feel compelled to grant investors extensive tax holidays and exemptions to counter similar moves from competitor states (von Haldenwang et al., 2021; Waiswa and Rukundo, 2023). This is one of the main drivers of the so-called race to the bottom in corporate taxation (Abbas and Klemm, 2013; Genschel and Seelkopf, 2016). With regard to IFFs, three aspects in particular deserve discussion: (i) the abuse of TEs, (ii) the weakening of public control, oversight and monitoring capacities as a consequence of TEs, and (iii) the lack of transparency created by the combination of and interactions between TEs.
11First, corporate tax incentives are a pull factor by which countries seek to attract capital inflows. As a side effect, however, they may create additional opportunities for multinational corporations to engage in aggressive tax planning. Though TEs are not always and necessarily connected to IFFs, they may be used to facilitate such flows. For example, corporate tax incentives can spur capital ‘round-tripping’, ‘a form of tax-induced regulatory arbitrage that involves moving capital offshore only to bring [it] back onshore again in the guise of FDI [foreign direct investment], so as to benefit from preferential tax treatment’ (Buckley et al., 2013, 120). Round-tripping has been observed in China (Kar and Freitas, 2012), but also in the US (De Simone, Lester and Markle, 2020) and elsewhere (Damgaard, Elkjaer and Johannesen, 2019). The ‘Mbeki report’ (Mbeki et al., 2015, 88–89) lists several types of IFFs that are at least partly based on TE abuse, in particular the overpricing of exports, imports, investments and related-party borrowing.
12A second mechanism involves weakened state control over private sector corporations due to TEs. Companies may use their financial, technological or market power to obtain particular preferential tax treatments, sometimes by means of illicit practices (above all, corruption). This is especially the case for large corporations operating in the extractive sector, where TEs are often granted via bilateral contracts with inbuilt stabilisation clauses and limited transparency (Le Billon, 2011; Mosquera Valderrama, 2021; Readhead, 2018). Such TEs do not per se cause IFFs. A lack of transparency and the excessive use of TEs may, however, result not only in the lowering of the tax burdens of companies, but also in reductions in the oversight and monitoring capacities of the state.
13It can be argued that some types of TEs have a stronger impact on oversight and monitoring than others. For example, tax holidays may, for companies, lead to lower reporting obligations and, for revenue authorities, to lower monitoring responsibilities over a given period. As a consequence, corporations may find it easier to dodge taxes by means of illicit actions. In contrast, tax deferrals and specific loss carry-forward regulations require both companies and revenue authorities to keep track of operations for the period covered by the TE in question.
- 3 Information obtained from the Zimbabwean Revenue Authority (ZIMRA) in the context of the project ‘P (...)
14Also, tax authorities sometimes deliberately lower reporting requirements in the name of simplicity and efficiency. In Zimbabwe for instance, corporations are required to declare the amount of corporate tax incentives granted to them as one single item on their income tax return form. This makes it much more difficult for the authorities to monitor and evaluate the use of individual TEs.3 This mechanism also commonly applies in free zones, where companies may be fully exempt from custom and internal duties as well as other charges. On the one hand, this reduces the filing requirements and compliance costs of companies established within the zone. On the other, the absence of tax revenue makes the monitoring of economic activity in these zones unattractive for revenue authorities, as other monitoring and investigation avenues might render higher returns in terms of tax revenue recovery. Indeed, it can be assumed that tax and customs authorities are more motivated to monitor and enforce tax rules in cases where additional revenues are at stake—such as the control of imports subject to excise tax—than they are to enforce and monitor potentially exempt activities within a free zone.
15In a third mechanism, combinations of TEs and the manifold interactions between them may open additional avenues for corporate or private actors engaged in illicit activities. TEs have the property of making tax systems more complicated and less transparent. As Fuest and Riedel (2009, 46–47) state, ‘there is widespread concern that tax expenditures are more difficult to control, more vulnerable to capture by lobby groups or even corruption and therefore more likely to lead to budget imbalances and governance problems than direct government expenditures’. These characteristics tend to play into the hands of actors engaged in IFFs, because they make it easier for private actors to exploit cross-border differences in tax treatments while at the same time making it more difficult for governments to control the impact of individual TE provisions and initiate countermeasures in cases of TE abuse.
16As shown by the GTED (see Redonda, von Haldenwang and Aliu, 2023), a large number of countries routinely manage 100 or more individual TE provisions. This is the case for several high-income countries, including Australia (228 provisions reported in 2020), the USA (328 provisions), South Korea (373) and Italy (513), upper-middle-income countries such as Brazil (230), Morocco (254) and Türkiye (225), but also some low- and lower-middle-income countries, such as Benin (146 provisions in 2019), Madagascar (247 in 2016) and Senegal (115 in 2014). In addition, elected representatives are often confronted with weak or inconclusive reporting, making it more difficult for them to understand and vote on tax laws, which are regularly tied to government budgets. Again, there is no direct causal pathway from TE complexity to IFFs, but it is fair to assume that the mere proliferation of TEs and the resulting complexity of tax systems might be another enabling factor for IFFs, undermining transparency and the accountability of public finances.
17Against this backdrop, we would expect higher levels of TE use–measured, for instance, in revenue forgone as a percentage of GDP or in numbers of TE provisions—to be positively associated with IFFs, proxied, for instance by data on trade misinvoicing (GFI, 2021) or tax secrecy (Cobham et al., 2021). However, sample bias due to TE under-reporting may compromise the identification of patterns at this aggregate level, as the same factors may drive IFFs and TE under-reporting (see von Haldenwang et al., 2023, 76).
18There is another core feature of TEs that relates them to the practice of IFFs: TEs lower the effective marginal tax rate of firms in a given jurisdiction. According to Carbonnier and Mehrotra (2018, 5), corporate tax-rate differentials between different jurisdictions have been identified as a ‘main driver’ of IFFs. The authors identify two major channels of trade-related IFFs—trade misinvoicing and transfer mispricing. With regard to the latter, illicit behaviour is above all motivated by the desire to exploit differences in taxation levels (Beer, de Mooij and Liu, 2020).
19It should be noted that some of the context factors discussed in this section have also been identified by the OECD project on base erosion and profit shifting (BEPS) initiated in 2013, and in particular by BEPS Actions 5, on harmful tax practices, and 6, on tax treaty abuse (OECD, 2013; 2023). However, while the BEPS agenda addresses important weaknesses of international taxation, the use and abuse of TEs in particular has until recently largely remained outside the focus of the project. It is only in the context of the above-mentioned GLOBE approach that the use of tax incentives for investment is receiving more scrutiny (González Cabral et al., 2023b).
20This section focuses on two kinds of TEs in source countries: SEZs and TEs for the extractive industries.
21Many countries establish SEZs to attract FDI and promote industrial growth. Typically, these zones combine several kinds of TEs, above all exemptions from customs duties and value-added tax (VAT) on imported goods as well as TEs related to corporate income tax (CIT), including exemptions, deferrals and reduced CIT rates. The 2019 World Investment Report counts nearly 5,400 SEZs worldwide, with total numbers still growing (UNCTAD, 2019, 128). In addition, almost 8,400 single-enterprise zones (so-called free points) are to be found across 18 economies, according to the report. The type and number of SEZs varies widely across countries. Some have only a small number of SEZs while others may have hundreds. More than 2,500 such zones exist in China alone, many of them at the provincial level. With regard to the various types, some SEZs may seek to enhance exports of certain products or those of certain sectors, while others aim at attracting FDI to certain geographical areas. Moreover, UNCTAD (2019, 135) observes ‘numerous examples of zones within zones’—so, multilayered structures where specific regimes are embedded in others, thus adding to the overall complexity of TEs.
22Variations in the total numbers, types and patterns of embeddedness of SEZs may contribute to some of the push factors identified by Carbonnier and Mehrotra (2018), such as lack of transparency, lack of oversight and monitoring, and weak governance. The UNCTAD’s World Investment Report 2019 highlights that one of the negative financial impacts of SEZs results from ‘the misuse of zones for illicit financial flows and trade misinvoicing, which can be an important problem in zones with laxer government controls’ (UNCTAD, 2019, 186). More specifically, SEZs can act as enablers or facilitators of IFFs through several channels:
23First, related to the argument regarding TE complexity developed in the previous section, SEZs may act as a ‘black box’ in which businesses often benefit from a wide range of TEs, which makes it very difficult (if not impossible) for governments to assess which TEs are cost effective and which are redundant or even harmful. This makes such governments vulnerable to lobbying and political interference, as shown, for instance, by Daude, Gutierrez and Melguizo (2017) with reference to the Dominican Republic.
24Second, as discussed by Heitmüller and Mosquera (2021), SEZs can result in BEPS. This happens, for example, if a company has a legal entity within the SEZ while the productive structure is located elsewhere, allowing the company to shift profits to the legal entity inside the SEZ and benefit from the preferential tax treatment on offer. In this sense, ‘a company established in an SEZ can fulfil a similar function as a tax haven company’ (p. 478). Based on BEPS Action 5 as well as on the European Union Code of Conduct for Business Taxation, an SEZ can be considered as harmful if there is insufficient economic substance with regard to employees and assets. These rulings and subsequent peer reviews have led several Latin American governments, for instance, to modify their SEZ regimes (p. 486). To the best of our knowledge, there has been, however, only limited research on the impact these modifications had with respect to profit shifting through SEZs.
- 4 When a company resident in a country undertakes activities in another country, there are often mini (...)
25Third, provisions included in tax treaties may interact with tax concessions available under SEZ regimes in source countries. Companies established in SEZs often undertake activities related to international commerce and logistics, such as commodities brokerage, shipping, warehousing, packaging and sorting, and insurance. In some countries, bilateral tax treaties benefit SEZ companies as well, on top of the tax rules of the SEZ itself. This means that companies engaged in merchandise storage, international shipping or services linked to SEZ businesses may, for instance, benefit from certain treaty-based exclusions under PE rulings, resulting in non-taxation of active business income in the source country. In countries where only little resources are available to enforce treaty provisions, tax advisers and companies have additional leeway to exploit these interactions between SEZ regulations and treaty provisions, generating IFFs.4 Not least, treaty-based tax rules in the source country can be directly linked to tax advantages available to companies in countries seeking to attract financial flows if those countries are at the same time the hosts of investor companies and recipients of IFFs (see, for example, Aidha et al., 2019 with reference to the Netherlands and Indonesia).
- 5 It should be noted that this is not only an issue of originating countries. For example, freeports (...)
26Fourth, a report issued by the Financial Action Task Force (FATF) shows that SEZs can be instrumental in facilitating the movement of illegally obtained resources and related financial flows (FATF, 2021, 56). Domestic aggregators, refiners and preprocessing businesses act as ‘supply chain choke points’, where licit and illicit goods are commingled to facilitate the laundering of proceeds from illegal extraction or environmental and human rights crimes (FATF, 2021, 55). SEZs provide a variety of channels that facilitate illicit financial outflows from originating countries, adding layers of transactions on illegally extracted resources, and ensuring no or minimal taxation under secretive conditions (FATF, 2021; FATF and Egmont Group, 2013; OECD, 2022b).5 In Botswana, for instance, SEZs dedicated to diamond processing are subject to preferential tax treatment while ensuring access to the international financial system (Masuku-Chimbombi, 2023).
- 6 While granting incentives to attract foreign direct investment is common practice, the available em (...)
27Countries that are rich in natural resources frequently use TEs to promote sectoral growth. This is particularly true of countries that depend on the inflow of capital and technology to develop the sector in question (Coulibaly and Camara, 2022).6 As observed by Le Billon (2011, 6), ‘contracts are frequently negotiated not with resource companies in their home country but with subsidiaries incorporated in low- or no-tax jurisdictions. This insures companies against tax payments agreed under bilateral tax treaties. Profits routed through the subsidiaries’ low-tax jurisdictions are then passed on to the company’s group, often through the proceeds of high-interest loans, in order to also avoid taxes in the home country’.
28Mining investors may change their behaviour in response to tax incentives as part of their tax planning strategies, significantly increasing the risk of BEPS. According to Readhead (2018), tax incentives offered in the form of income tax holidays, export processing zones (EPZs), royalty-based incentives, withholding tax reliefs on interest services, cost-based incentives and fiscal stabilisation guarantees entail a particularly high risk of behavioural response by mining investors. For instance, if a subsidiary of a mining corporation is set up in an EPZ that offers reduced tax rates, there is an incentive for the mining company (subject to the standard tax rate) to underprice its minerals in order to reduce its taxable income and shift profits to the subsidiary in the EPZ.
29The frequent use of stabilisation clauses in exploration or extraction contracts explains, to a certain extent, the lack of flexibility that governments often face when it comes to modifying the use and design of TEs. IMF et al. (2015, 29) discuss the use of stabilisation clauses and state that ‘such stability provisions […] create an uneven playing field between old and new investors and can lead to significant distortions. Such situations should not last for too long. Government[s] might therefore need to renegotiate existing incentive provisions or provide reasonable, time-bound incentives to new investors’. However, if these stabilisation agreements are not disclosed, they cannot be subject to independent analysis and evaluation (Mosquera Valderrama, 2015).
30Beyond stabilisation agreements, the taxation of extractive activities can be largely determined by secretive contracts between governments and the relevant companies. There might be no indication under domestic law or in public documents as to the taxation rules determined by such contracts. For instance, the Botswana diamond mining sector has its tax rules determined exclusively under contract, and mining contracts in the country are not public (Masuku-Chimbombi, 2023). The complete opacity of diamond mining taxation in Botswana not only raises questions as to whether the country is properly mobilising domestic resources from its natural wealth, it also conceals the sector under a veil of secrecy that may promote corruption and IFFs.
31In such contexts, tax treaties also play an enabling role in IFF outflows derived from extractive activities. Beyond specific PE regulations, discussed in the preceding section, a relevant provision in tax treaties concerns the taxation of capital gains (Article 13 of the UN and OECD model tax conventions; see OECD, 2017; UN, 2017). In general, countries retain the right to tax income generated upon the disposal of assets in their territory, and in particular immovable assets such as subsoil natural resources, which are traditionally tied to national sovereignty. However, under tax treaties signatory countries may relinquish that right indirectly by giving up the right to tax gains on the disposal of shares in companies or other legal vehicles owning domestic immovable property when those shares are bought or sold by a resident of a treaty partner country.
32For instance, according to the Tax Treaty Explorer database (Hearson, 2022), the treaties in force linking Botswana with Mauritius and with South Africa prevent the taxation in Botswana of capital gains from the sale, by Mauritian or South African residents, of shares in a partnership or trust with substantial mining rights in Botswana. Thus, although under Article 31(2) of the Botswana Income Tax Act these gains would be subject to domestic tax (BURS, 1995), provisions contained in these treaties award exceptional treatment to indirect sales of mining rights, which become tax exempt in Botswana and may remain untaxed abroad—expanding opportunities for aggressive tax planning.
33In the context of the mechanisms discussed so far, conflicting interpretations can give rise to disputes, where the tax authorities of source states claim that additional tax revenues are due. In particular, the analysis of treaty disputes indicates that royalties taxation is the most contentious topic, with around 12 per cent of total disputes, while 8 per cent of disputes concern capital gains taxation (Baistrocchi and Hearson, 2017, 1533). With regard to stabilisation agreements, source countries may be reluctant to make changes to the law for fear of legal disputes with companies under contract, or with their competitors (Gehne and Brillo, 2017). Thus, the prospect of international disputes may hinder the efforts of public authorities in source countries to fight IFFs derived from the extractive sector.
34In the context of IFFs, recipient countries are those that absorb financial flows from countries where income has been generated illegally, or from those where it should legitimately have been taxed. Section 4 discusses how selected TEs enable the flow of IFFs into recipient countries. Research has shown that a limited number of offshore financial centres act as transit countries (intermediating investments between countries) or final recipients (attracting and retaining foreign capital) in the global economy (Garcia-Bernardo et al., 2017). While the tax policy orientation of the latter centres is often simple non-taxation, the former commonly implement complex tax regimes offering very low or no tax to companies choosing to channel economic activities through their jurisdiction. Both IFFs related to criminal activities and those associated with aggressive tax planning use the same legal structures internationally, which allow for the decoupling of income or ownership rights from the source of those rights.
35Actors engaged in criminal IFFs seek to separate illegal origins from resulting wealth. For IFFs related to aggressive tax planning, actors want to remove income from the high-tax country where it was generated and reassign it to low- or no-tax jurisdictions. TEs that foster the inflow of IFFs are thus in most cases profit-based preferential regimes, which impose lower or no taxes on mobile business income, with little regard for economic activity on the ground. Typical examples of such profit-based TEs are CIT holidays or exemptions granted on certain types of income. They should be distinguished from expenditure-based TEs, which ensure a degree of consistency between the source of income and taxation and tend to reward reinvestment. Expenditure-based tax regimes may include, for example, deductions from taxable income or accelerated depreciation rules linked to investment activities (Meinzer et al., 2019; UNCTAD, 2022, 6).
36A mechanism that has become particularly prevalent among rich countries involves the use of PBs to boost R&D and innovation. Governments subsidise R&D because the social return on R&D investment can be higher than the private return to the companies funding the R&D, causing underinvestment in R&D from a common good perspective. As described by Köhler et al. (2012), most of the literature focuses on the impact of R&D tax incentives on R&D expenditure (i.e. input additionality), only few studies analysing the effect of R&D tax incentives on innovation and economic activity (i.e. output additionality).
37PBs grant preferential treatment to corporate income earned through intellectual property (IP). The first PB scheme was introduced in France in 1971. More recently, several other (mostly high-income) countries—including Belgium, China, Italy, the Netherlands, Spain, Switzerland and the UK—have also put various types of PB in place, though recent modifications in international taxation rules have affected the ability of governments to implement such regimes (see below). PB regimes vary in the tax-rate reductions they offer as well as in their scope and eligibility criteria. In Europe, the reduced tax rates of PBs (i.e. the tax rate on corporate income for which PB benefits are granted) range from 0 per cent in Malta to 15.5 per cent in France.
38Yet besides their stated policy goal, PBs have been proven ineffective in boosting R&D and innovation. Alstadsæter et al. (2018) use firm-level data for three sectors—pharmaceuticals, cars, and information and communication technology—to estimate the impact of PBs on the patent filing strategies of firms. Their results indicate that PBs have a considerable impact on attracting patents. At the same time, however, the authors conclude that PBs do not have a significant impact on real activity because multinational corporations shift the location of their patents without shifting their research operations. Indeed, transferring IP rights from related entities to affiliates in low-tax jurisdictions is one of the two practices highlighted by González Cabral et al. (2023a) as directly affecting international tax planning. The second of these practices concerns the strategic overvaluation of intangible assets within corporate groups, which allows firms to artificially inflate expenses in high-tax regions, thereby reducing taxable income. According to Neubig and Wunsch-Vincent (2018), the tax-induced mismeasurement of cross-border IP flows could amount to more than 35 per cent of global charges for the use of IP, and the figure could be even higher in high–tax rate countries.
39The typical case in large firms is that the inventor is contractually obligated to assign the patent to his or her employer or a holding company, which could have its official address anywhere, including in jurisdictions commonly identified as tax havens such as the Cayman Islands or Luxembourg. Hence, the main effect of the PB discount on royalties or other IP-related profits is to attract assignees, not inventors. The country gains inventors only indirectly, should the assignee choose to use tax savings to employ more inventors in the country that offered the IP box.
40The nexus approach and its more recent version, the so-called ‘modified’ nexus approach described in Action 5 of the BEPS package, are designed to counter this strategy by introducing eligibility conditions to ensure a clearer link between the income derived from IP and the expenditure incurred to develop the IP (OECD, 2015). R&D is, however, often a complex and lengthy process. Expenditure on R&D may be used for several projects. Identifying the total amounts of R&D expenditure that over the years should be allocated to a given patent is a difficult requirement to monitor. Moreover, the nexus approach would not deter firms from merely shifting their R&D investments from one jurisdiction to another rather than creating additional R&D in the country granting the tax benefit. As a result, the country implementing the PB would see domestic R&D spending go up, but the overall global investment on R&D and, thus, innovation could remain unchanged. As discussed by González Cabral et al. (2023a), the growing disparity between where IP income is reported and where the actual R&D activities occur not only goes against the spirit of most IP legislation, it also deprives innovating nations of their rightful tax revenues.
41Finally, another mechanism worth mentioning—although it is not yet the object of scientific research related to IFFs—involves tonnage tax regimes (see Selkou and Roe, 2002 for a discussion with reference to the UK). These regimes are applied by more than 20 European countries to international shipping and related activities. They assess tax liabilities based on the total tonnage of ships owned or managed by a company, not on profits. Effective income tax rates under these regimes may be as low as 0.5 to 2 per cent (Rivero, 2022), resulting in substantial revenue forgone. Tonnage tax regimes usually apply to the shipping of commodities, and related services such as loading, unloading and sorting, but may also be designed to include activities such as natural resource exploration, or resource extraction through offshore drilling. Such economic activities—whose income generation is largely unrelated to the tonnage of the vessels in question—are, for example, included in the tonnage regimes of Cyprus, Denmark, the Netherlands and Malta (TJN, 2021). Because multinational companies can largely choose where to register ships engaged in offshore resource extraction activities, tonnage tax regimes may lead to an – arguably illicit – under taxation of extractive activities. Aggressive tax planning based on these regimes is facilitated by jurisdictions in which little exploration and extraction is undertaken but large profits are booked. These countries are creating important negative spillovers for other countries while still benefiting from tax revenues charged under tonnage regimes and potentially promoting some growth in ancillary services.
42The present chapter has explored the relationship between the widespread granting of TEs and the existence of IFFs. It has shown that TEs may play an important role as facilitators and enablers of IFFs. The use of TEs has been discussed separately with reference to source countries (where IFFs originate) and recipient countries (where IFFs are received, either as their final destination or in transit to other jurisdictions). Based on the available evidence—which is, it must be said, far from abundant or conclusive—we have focused on three mechanisms in particular: the use of SEZs, the granting of TEs in the extractive industries, and the application of PBs and other provisions that target intangible assets.
43These three mechanisms were chosen in order to highlight specific properties of TE regimes that make them vulnerable to IFFs. They have one thing in common: rather than referring to single TE provisions—such as, for instance, an exemption on import duties for capital goods—each covers several TEs at the same time. For example, SEZs typically combine CIT reductions with import VAT and customs duty exemptions, as well as in some cases reductions in withholding taxes and other tax relief. It is important to note, however, that the TEs discussed in this chapter do not exhaust the universe of available measures. Many more such provisions are used by governments. Insofar as they exacerbate tax-rate differentials between different jurisdictions, they too may constitute drivers of IFFs.
44In addition, the chapter has discussed how bilateral tax treaties may affect the use of TEs for IFFs, in particular by their adding another layer of complexity to the tax system, making it more difficult for governments to monitor the use and prevent the abuse of individual provisions. Hence, while there is at present no direct evidence that links the use of TEs to the overall size or specific patterns of IFFs, this chapter has identified several means by which TEs may enhance IFFs. Secrecy, the complexity of tax systems, and weakened monitoring by the state appear to be core driving forces in this regard.
45Even in the absence of robust evidence, the insights presented here allow us to draw some tentative conclusions. It is clear that transparency, tax certainty, and simplicity are important properties of tax systems and that governments may wish to strengthen them when seeking to protect their tax base. These properties apply to TE regimes as well. Consequently, one line of action would involve reducing the complexity of TE regimes and controlling for unwanted interactions that might trigger IFFs. Further, resource-dependent countries would be well advised to safeguard mining rights when negotiating tax treaties (Readhead and Taquiri, 2021, 20–55). Publishing extractive contracts—for instance under the Extractive Industries Transparency Initiative standard (Murombo, 2022)—could also constitute an important step forward.
46Finally, governments need to align their tax systems with the requirements of international tax standards. The GLOBE proposal developed by the OECD (OECD, 2022a) and the introduction of a global minimum tax of 15 per cent in particular are going to have important repercussions for the use of business-related TEs, even considering that this minimum tax applies only to a constrained number of large multinational corporations and only refers to excess profits—so, to profits in excess of a fixed return on tangible assets and payroll (see González Cabral et al., 2023b for an in-depth discussion). While it is too early to draw any conclusions about the impact of this minimum tax, one of the main outcomes will be the need for countries to revisit the use of CIT-related TEs. It can be expected that TEs granted on profits of companies will likely be more affected than TEs granted on certain business expenditures. Hence, tax holidays and SEZs that lower effective tax rates to 0 per cent should be particularly affected. Since profits-based TEs are more clearly linked to IFFs, as shown above, the expected impact of the GLOBE initiative could indeed be positive for the global fight against IFFs.