Special Economic Zones Facing the Challenges of International Taxation: BEPS Action 5, EU Code of Conduct, and the Future

Corporate taxation and particularly corporate tax incentives that jurisdictions introduce in special economic zones have not, until recently, been subject to extensive international regulation. OnlyinthelastdecadehasaregimeofsoftlawstandardsandEuropeanUnion measures with extraterritorial effect been constructed. This article explains how the Base Erosion and Profit Shifting Action Plan developed by the Organisation for Economic Co-operation and Development (OECD) and the European Union Code of Conduct for Business Taxation interact with corporate tax incentives in special economic zones. Empirical evidence from Latin American and Caribbean jurisdictions shows that this emerging international regime began having an impact on special economic zone laws beyond the OECD and European Union Member States. An analysis of ongoing negoti-ations on the further developments of the international tax regime permits the cautious conclusion that the regulation of SEZs may in the future be affected in a more fundamental manner by international norms. Thereby, the article shows that special economic zones’ unilateralism in corporate taxation may be slowly receding in contrast to other areas of international economic governance.


I. INTRODUCTION
In many jurisdictions, companies operating in special economic zones (SEZs) benefit from permanent or temporary corporate tax exemptions or reduced tax rates. Corporate taxes are levied on the profit of corporations. Corporate tax rules are sometimes employed as a tool of economic policy. 1 For example, exemptions or reductions are granted to favour specific economic activities over others or to incentivize investments in certain geographic areas. In SEZs, such exemptions or reductions have been conferred with the objective of attracting investment into the zone.
This article analyses the interaction of corporate tax rules in SEZs with the transnational legal order (TLO) of international taxation 2 that primarily comprises bilateral tax treaties, international soft law standards developed by the Organisation for Economic Co-operation and Development (OECD) and the United Nations (UN), as well as European Union (EU) soft law with an impact on third states. 3 With reference to international trade and investment law, SEZs have been interpreted as 'unilateral economic law', i.e. domestic policies that embody the fundamental values of international trade and investment law while operating in parallel to or in the absence of multilateral or bilateral treaties. 4 Similar to the international trade and investment regimes, the first phase of the TLO of international taxation primarily aimed at liberalizing capital flows through the prevention of double taxation of international transactions. However, its scope remained more limited. 5 The TLO's main components were bilateral double tax treaties and the model tax conventions initially developed by the League of Nations in the 1920s and later by the OECD and the UN. 6 The latter strongly influenced the content of bilateral double tax treaties. These treaties typically constrain the amount of withholding taxes countries can levy on the income of non-residents from sources in their country. They also establish a threshold of connection with a country's economy ('the permanent establishment')-only above which a foreign taxpayer can be considered as falling under the country's full taxing power. However, tax treaties are agnostic in respect to whether, how, or how much tax they levy on domestically incorporated See also definition in the introduction to this special issue by J. Chaisse and G. Dimitropoulos. Unilateral economic law is a 'phase in which states continue to show a respect for the values of international law but in which the institutionalization of these values has devolved from international towards regional and often domestic levels of governance' . 5 The question of whether an 'international tax regime' exists at all has incited much academic debate without leading to a consensus among scholars. See companies or permanent establishments regardless of whether these are owned by foreigners. 7 This has led scholars to qualify the nature of the international tax regime as 'sovereignty preserving' . 8 As a consequence, corporate tax rules in SEZs could not be interpreted as 'unilateral economic law' because they were not related to the objectives of the TLO of international taxation. However, after this early 'sovereignty preserving' phase, the TLO underwent significant transformations beginning at the end of the 1990s and more markedly since the global financial crisis of 2008. 9 Tax evasion by high-net-worth individuals and aggressive tax planning strategies by multinational enterprises (MNEs) became increasingly pressing concerns in many countries in a context of fiscal austerity. 10 This situation led to initiatives through which countries intended to coordinate countermeasures against aggressive tax planning. 11 Its main components are the OECD's project on Harmful Tax Competition that was initiated in 1998, 12 which evolved into the Base Erosion and Profit Shifting (BEPS) Project in 2013 13 and the EU's Code of Conduct for Business Taxation. 14 The latter coordinates tax policy among EU Member States and, since 2016, serves as a basis for the establishment of the EU's 'list of non-cooperative jurisdictions for tax purposes' through which the EU exerts influence on tax policies of third countries.
To some extent, the ability of MNEs to adopt aggressive tax planning strategies was considered as an undesired side effect of liberalization. 15 The prevailing idea in these initiatives is that countries thus need to coordinate certain tax rules to prevent taxpayers from exploiting discrepancies among different jurisdictions' tax laws to decrease their tax burden in a manner that is not intended by legislators. 16 In sum, the transnational legal order of international taxation now attempts to balance the goals of enabling the liberalization of capital flows, on the one hand, and safeguarding countries' capacity to raise tax revenues, on the other hand. This transformation has consequences for corporate tax rules in the SEZs that are analysed in this article, using the example of SEZs in Latin America and the Caribbean (LAC). Despite their soft law nature, the OECD and EU initiatives have had a tangible impact on SEZs in the last years: nine countries of the LAC region changed their SEZ regimes to align with these international tax standards. 17 This suggests that SEZ laws have become embedded in the TLO of international taxation. The fact that some states exceed what was technically required from them in their reforms demonstrates that the standards developed by OECD and EU may often not substantially contradict with the objectives that states of the region pursued through SEZs. However, the authors also discuss whether and how the scope of the TLO of international taxation might be extended in the coming years, potentially spurring changes of a more substantial nature in corporate tax incentives in SEZs. 18 The article further reviews the different types of corporate tax regulations found in SEZs in the LAC region under section II. In section III, the article explains how the standards established by the BEPS Project and the EU Code of Conduct interact with tax incentives typically found in SEZs and assesses their practical impact on the SEZs in the LAC. In section IV, the article looks to the future and discusses how the global anti-base erosion (GloBE) proposal recently published by the OECD and currently debated in the BEPS Inclusive Framework might impact SEZs and economic unilateralism promoted through them.

II. INCOME TAX REGIMES IN LATIN AMERICAN SEZS
This section of the article presents evidence from a database on SEZ tax regimes in LAC compiled by the authors. 19 Currently, an extensive variety of corporate tax regimes that are applicable to SEZs can be observed. Out of 42 countries and autonomous tax jurisdiction in the region, 29 operate SEZs. According to United Nations Conference on Trade and Development (UNCTAD) data, these jurisdictions host 486 out of 5400 of 16 (2016)).
the SEZs worldwide (9%). 20 As illustrated in Table 1, among the LAC countries that have introduced SEZs, most offer a more favourable corporate income tax regime than what is applicable in the rest of the country in order to incentivize corporations to establish their activities in the zones. However, a variety of corporate tax treatments can be observed in the region: A few jurisdictions offer a moderate rate reduction (less than 50% of the generally applicable corporate income tax (CIT) rate). This is the case of the general Colombian SEZ regime and the Brazilian Manaus Free Trade Zone (FTZ). 21 Second, a few jurisdictions such as Aruba, Curacao, and Jamaica offer-or used to offer 20 Own calculation, based on UNCTAD, 'World Investment Report 2019: Special Economic Zones. Web Table  21. until recently 22 -a near-total exemption of income tax. The most widespread types of income tax benefits in LAC, however, are tax holidays of various lengths and full income tax exemptions. Some jurisdictions have more than one SEZ regime-sometimes with different types of benefits. Costa Rica, for example, offers a shorter tax holiday (8 years) in SEZs located in the Greater Metropolitan Area 23 than in SEZs outside the area (12 years). Colombia offers benefits that are more generous to SEZs established in the city of Cúcuta. A last type of benefits includes a switchover to gross instead of net taxation: In Paraguay, for example, SEZ firms can elect to be taxed at 0.5% of their turnover. Finally, some of the jurisdictions of the region offer no corporate income tax benefits in SEZs at all-Bolivia and Argentina, for example. The Bahamas and Cayman Islands do not levy a corporate income tax, which is why there cannot be a more favourable corporate income tax treatment in their SEZs than in the country in general.
The above categorization generally relates to the situation in 2020 24 and does not show how corporate income tax benefits in the region have developed over time. Longitudinal data on the development of the regulatory regimes of SEZs are scarce. Based on the literature on economic zones, however, it can be assumed that tax benefits have always been part of the regulatory package. 25

III. THE IMPACT OF BEPS ACTION 5 AND COCG ON SEZS
Since the end of the 1990s, aggressive tax planning and enabling policies and practices of states have featured prominently on the agendas of international organizations and have led to the OECD's initiative on Harmful Tax Competition 26 as well as to the establishment of the EU Code of Conduct on Business Taxation. 27 Both organizations have sought to establish standards to assess if the tax regime of a country can be considered as 'harmful' in the sense that it may contribute to the erosion of other countries' tax bases. Most aggressive tax planning structures rely on low-or no-tax jurisdictions: For example, a company that invests in a subsidiary in another country can set up an intermediary financing company in a tax haven (typically without any significant business activities) to defer taxation on income from the investment. Similarly, intellectual property rights may be transferred to a tax haven country to receive royalty income from subsidiaries in other countries tax free. 28 As shown in section III, many countries offer low tax rates or tax exemptions for business conducted within SEZs. Hence, a company established in an SEZ can fulfil a similar function as a tax haven company in an MNE's tax structure. Thus, SEZs have also become the object of the OECD's and EU's initiatives for tackling aggressive tax planning.
The following sections detail how the regimes on harmful tax practices of the OECD (sub-section A) and the EU (sub-section B) relate to tax incentives in SEZs. Subsection C shows how SEZs in LAC have been evaluated by these regulatory initiatives, and sub-section D assesses how countries have responded to evaluations.

A. SEZs in light of BEPS Action 5
The OECD's work on harmful tax practices was in its infancy in 1998 and confined to its Member States and to 'tax haven' jurisdictions during the 2000s. The latter group included jurisdictions from the Caribbean; however, assessments focused on whether these countries' tax systems could be classified as a 'tax haven' in their entirety. Moreover, in the beginning of the 2000s, the OECD's work was refocused on aspects of tax information exchange and bank secrecy and, therefore, reforms asked from jurisdictions in the region concerned this type of regulations and not the preferential tax regimes in SEZs. However, the scope of the OECD's work was extended both thematically and geographically with the endorsement of BEPS Action 5 as one of four 'minimum standards' in 2015 and the creation of the BEPS Inclusive Framework (IF) in 2016. All 139 jurisdictions that have joined the IF (as of February 2021) have committed to complying with the BEPS minimum standards.
The standard of BEPS Action 5 mandates that countries must not resort to 'harmful tax practices' and establishes a review process of tax regimes. Thereby, it is based on the criteria of the 1998 Report on Harmful Tax Competition that already discussed which features of tax regimes could be considered as 'harmful' but puts a stronger focus on the issue of 'economic substance' . 29 Any preferential tax regime of a country that provides for a low tax rate or a (temporary) tax exemption and can be accessed by mobile types of businesses (such as, for example, financial services, holding companies, etc.) is in scope of a review. 30 Subsequently, the Forum on Harmful Tax Practices (FHTP), a subsidiary body of the OECD's Committee on Fiscal Affairs, assesses the 'harmfulness' of the IF members' tax regimes that fulfil the scoping criteria. In addition, IF members can suggest a review of a non-member's regime, or non-members can submit their own regimes for review. 31 29 In the 1998 report, four key factors and eight ancillary indicators were used to evaluate whether a preferential regime that is available for mobile business income (such as income from the provision of intangibles and financial services) can be regarded as harmful.  The regime is in scope, meets the low or no effective tax rate criterion, and implicates one or more of the criteria, but an assessment of the economic effects shows that the regime is not having a harmful impact in practice. The regime is subject to a yearly monitoring process by the FHTP and, when changes are identified, the FHTP can reconsider the conclusion.

Potentially harmful
The regime is in scope and meets the low or no effective tax rate criterion, and features of the regime implicate one or more of the criteria. However, an assessment of the economic effects has not yet taken place to decide whether the regime is (actually) 'harmful' .

Not harmful
The regime is in scope but does not have any features that implicate any of the criteria.

Out of scope
The regime does not grant tax benefits to geographically mobile activities Source: BEPS Action 5 report. Ibid.
Harmfulness is established if, in addition to the factors mentioned above, the regime is 'ring-fenced' from the domestic economy 32 and/or if it is available to taxpayers with insufficient economic substance (such as qualified employees and fixed assets). 33 However, the FHTP would only label the regime as harmful if an economic analysis reveals that the regime has detrimental economic effects in practice, i.e. that it effectively attracts mobile activities that lack economic substance. 34 If the latter is not the case, the regime would only be considered as 'potentially harmful but not actually harmful' but would be subject to continued monitoring. Table 2 summarizes the different possible outcomes of a review as stated in BEPS Action 5. that are potentially harmful and provides a framework within which Member States can commit themselves to follow the principles of fair competition' . 35 Notwithstanding a few differences, the resulting criteria to assess the harmfulness of preferential tax regimes were similar to those established in the OECD's Harmful Tax Competition Project. 36 The code is monitored by the EU Code of Conduct Group (COCG), which consists of high-level representatives of Member States and the European Commission. 37 While similar to the OECD initiative, the EU Code of Conduct applied only to EU Member States in its beginning. Yet, it already included the vision that 'associated territories' should be included in the scope of the reviews. The SEZ regime of Aruba, for example, was reviewed and amended in the early 2000s due to Aruba's status as an associated territory of the Netherlands. 38 In 2008, however, the EU introduced the EU Standard of Good Governance in Tax Matters, which mandates EU Member States to promote transparency, exchange of information, and fair tax competition in third (non-EU) countries. The standard is intended to be inserted in free trade agreements that the EU concludes with other jurisdictions. 39 To assess whether countries comply with the 'fair tax competition' criterion, the EU Commission relies on the criteria developed by the code of conduct. 40 In 2016, the EU further extended the territorial scope of the code of conduct via its Communication on its External Strategy for Effective Taxation. This communication establishes a listing process of jurisdictions that do not comply with the standards of good governance in tax matters. 41 Thereby, the code of conduct becomes de facto applicable to the entire world, and the COCG began reviewing tax regimes globally, including SEZ regimes in the LAC region. However, while the OECD has sought voluntary commitment to the BEPS Action Plan from jurisdictions around the world, the EU rather uses its market power and the reputational disadvantages that a country could incur when being placed on an EU 'blacklist' to induce jurisdictions to comply with the regulatory regime. 42 While the criteria that the EU uses to evaluate the 'harmfulness' of a tax regime are similar to those of the OECD's Forum on Harmful Tax Practices, a noteworthy difference lies in the importance that both organizations accord to the 'economic assessment' of the reviewed tax regimes. While the OECD BEPS Action 5 Report clearly states that a regime would only be regarded as harmful if it effectively attracts, for example, shell companies that contribute to the erosion of other countries' tax bases (see sub-section A), EU documents refer to the use of economic factors and impact data only as a 'guiding principle' . 43 This discrepancy could be one source of divergent assessments of SEZ regimes.

C. Impact on SEZs in LAC
At its onset, the work of both the OECD and the EU had only limited impact on countries of the LAC region, with Mexico being the only OECD member state from the region until Chile's accession in 2010. As explained above, the territorial scope of both initiatives was extended with the creation of the BEPS Inclusive Framework and the EU's list of non-cooperative jurisdictions. As of February 2021, 29 out of 42 jurisdictions of the LAC region are members of the BEPS Inclusive Framework. 44 In their review processes, the OECD's FHTP and the COCG do not focus on individual SEZs but on the corporate tax aspects of SEZ regulations that can potentially relate to one, several, or all SEZs in each country. Worldwide, 60 SEZ regimes in 32 countries have been reviewed by the FHTP, and 77 zone regimes in 46 countries have been examined by the COCG. 45 Sometimes, these two organizations work in a complementary manner while, in other instances, the same regime is assessed by both organizations, which has led to divergent views on the harmfulness of a certain tax regime in a few cases. 46 As shown in section III, many countries in Latin America provide tax exemptions, tax holidays, or low tax rates to companies in SEZs that signal they are potentially in scope for a review by the FHTP and COCG. Taken together, the SEZ regimes of 15 countries in the LAC region have been evaluated since the publication of BEPS Action 5 in 2015 and the extension of the EU's review process in 2016. Using the criteria outlined in sub-sections A and B, the FHTP and COCG considered the SEZ regimes of nine countries as harmful, specifically, Antigua and Barbuda, Aruba, Belize, Costa Rica, Curacao, Panama, Saint Lucia, Trinidad and Tobago, and Uruguay. In addition, at the time of publication of the latest report of the FHTP, the SEZ regimes of the Dominican Republic and Jamaica were still under review. The SEZ regimes of Brazil, Grenada, Paraguay, and Peru were found to be 'not harmful' or 'out of scope' of the review (see Table 3).
Both organizations prompted the evaluated jurisdictions to amend the regimes considered as harmful, sometimes resorting to political pressure. In an interview given in a Costa Rican newspaper, Pascal Saint-Amans, the director of the OECD's Centre for Tax Policy and Administration, made Costa Rica's entry into the OECD conditional on the amendment of its SEZ regime. 47 The Council of the EU sent correspondence to the assessed jurisdiction, urging them to undertake changes in order to be removed from the EU's list of non-cooperative tax jurisdictions. 48

D. Country responses to reviews
In response to the reviews by the FHTP and the COCG, all nine countries introduced amendments to their SEZ laws. This indicates some degree of effectiveness of the OECD and EU initiatives in curbing SEZ unilateralism. However, there is some diversity in the strategies that various countries adopted to comply with the reviews.
For regimes designed to attract intellectual property (so-called 'patent boxes'), the OECD developed a model substance requirement, the 'nexus approach' . Its essence is that a country should only allow firms to benefit from a preferential regime in proportion to the amount of the work to develop the intellectual property actually carried out in the country. 49 The OECD guidance contains a formula to calculate the nexus. For preferential regimes that countries offer to other types of mobile economic activities (e.g. financial or headquarter services), the guidance is less precise. The requirement is, in essence, that countries must not allow companies to benefit from the regime if the 'core income generating activities' of the company are not undertaken in a country, e.g. if an MNE establishes a subsidiary responsible for the firms' investment in a preferential regime, but the people who take the relevant investment decisions work in another country. In practice, the OECD translates the concept of 'core income generating activities' as 'having an adequate number of full-time employees with necessary qualifications and incurring an adequate amount of operating expenditures to undertake such activities' . 50 A few countries in the LAC region introduced the minimum requirement to comply with reviews by proposing a substance requirement modelled on the OECD recommendations in their SEZ law. Some jurisdictions made amendments that are more significant than strictly necessary to comply with the reviews, for example, excluding 47 Marvin Barquero, 'OCDE exige reforma a zonas francas para dar acceso al pais a ese bloque', La Nación, 18 September 2017, https://www.nacion.com/economia/politica-economica/ocde-exige-reforma-a-zonasfrancas-para-dar-acceso-al-pais-a-ese-bloque/RFM3UHK2RNHBTG273WXB2EP7MY/story/ (visited 22 July 2020). 48 While the correspondence is not entirely publicly available, a compilation of the response letters received is publicly available.  some or all 'mobile' economic sectors entirely from the benefits of the SEZ tax regime. 51 Other jurisdictions, however, introduced similar amendments but also measures that could be interpreted as compensatory-for example, by making a certain type of benefit applicable to all taxpayers (whether located in an SEZ or not). Costa Rica, for example, introduced a so-called 'strategic eligibility index for service companies' . 52 The aim of the index is to reserve access to the regime for companies that pay a certain remuneration to employees in Costa Rica and commit to a certain amount of investment in fixed assets there, thus ensuring that companies have a sufficient degree of substance. 53 In addition, it only allows activities considered as being strategic for the country's economic development to benefit from the SEZs' incentives. By excluding financial services from the list of potential beneficiaries, some typically 'mobile activities' are prevented from accessing the regime. 54 Trinidad and Tobago introduced changes of a more fundamental nature as well: The country closed its 'free trade zones' regime to new entrants altogether, and the government announced the introduction of a new 'Special Economic Zones' regime that should contain, among others, substantial activity requirements and less beneficial corporate income tax benefits (time-bound and dependent on the size of a firm's investment). 55 Saint Lucia, on the other hand, attempted to compensate its change of the SEZ law with the introduction of another benefit in its general tax law. Mobile activities, such as financial services, were removed from the activities that companies may perform within the SEZ, and a requirement for new companies to furnish information about economic substance (such as the number of employees working within the SEZ and their qualifications) was introduced. At the same time, however, Saint Lucia changed its general corporate income tax system from a worldwide to a territorial system. This means that, after the reform, all foreign-sourced income of companies that are resident in Saint Lucia is exempt from tax. 56 The theoretical consequence is that businesses without substance formerly operating in the free zone might simply relocate outside the zone since 51 It should be pointed out that these additional changes might not be causally related to an OECD or EU review but might have been undertaken for other (domestic) reasons. 52 'Índice de Elegibilidad Estratégica para las Empresas de Servicios' Reforma Ley Régimen de Zonas Francas Asegurar cumplimiento de los estándares internacionales establecidos por la organización para la cooperación y el desarrollo económico en el marco inclusivo del plan de acción de lucha contra la erosión base imponible, Asamblea the general tax system of the country is then de facto equally beneficial for any business that earns primarily foreign income. On a theoretical level, this demonstrates an unwillingness to repeal unilateralist policies. The EU COCG, however, corresponded with Saint Lucia requesting the repeal of this reform as well. 57 At the time of writing of this article, Saint Lucia's reaction to this request is not yet public. Curaçao, on the one hand, introduced fundamental reforms to its 'E-Zone' regime, limiting access to the regime to companies that trade in goods and thereby restricting access for 'geographically mobile activities' such as trade in services. Moreover, in 2020, it closed the corporate income tax benefit for new companies altogether and will phase out the benefit for existing companies until 2022. 58 On the other hand, similar to Saint Lucia, Curaçao introduced a general exemption for income earned abroad. 59 Despite these variations among countries, the fact that no country ignored the FHTP and COCG reviews can be considered as first evidence that countries do not maintain SEZ policies that contradict with the international tax regime. While there is no comprehensive study on the impact of FHTP and COCG on SEZs at the global scale to the authors' knowledge, the reports released by both organizations indicate that countries in all regions of the world have made reforms in response to reviews. 60 It should be noted, however, that a few of the LAC countries identified in section II that offer corporate tax incentives in SEZs have not yet been assessed by the FHTP or the COCG at the time of the writing of this article. Moreover, the economic effects of the modifications to SEZ incentives enacted by countries are still unclear. In principle, if all companies operating in an SEZ had 'economic substance', i.e. employees, assets, etc., before the introduction of a substance requirement in the law, such a requirement would not have any economic effect since all companies would already comply with the new legislation. In the case of Costa Rica, as laid out above, the attraction of mobile economic activities with little substance was probably never an objective of the SEZs. In such a case, the objectives of the OECD and the EU are not substantially inconsistent with the objectives of the country.
On the other hand, if many 'shell companies' without substance benefited from a regime, the effect of a reform on these companies might be sizeable. However, since the primary characteristic of shell companies is precisely a lack of economic substance, the overall impact on the main economic indicators of an SEZ (such as employment, production, assets, etc.) is likely to be minimal as well. In this scenario, negative consequences would be expected for the corporate and legal services industry allows multinational companies to artificially shift profits in that country (for example, through the absence of any substantial activity requirements that the company needs to fulfil to access the benefits of the regime). 67 While an increasing number of reports by academics, international organizations, and civil society organizations highlighted the redundancy of many tax incentives and their role in a 'race-to-the-bottom' among jurisdictions, 68 the OECD resumed discussions on tax competition in general. 69 However, the idea of regulating tax competition at the international level remains politically controversial. The OECD's discussion documents mention the need to 'shield developing countries from pressure to offer inefficient tax incentives' as a reason to address tax competition. 70 Yet, not all tax policy stakeholders in developing countries share this understanding. On the contrary, establishing low tax rates or introducing special incentives is often viewed as necessary for developing countries to attract foreign investment. 71 Hence, the inclusion of so-called 'tax sparing' clauses in double tax treaties has been considered by some stakeholders as a policy favouring developing countries. 72 The logic of a tax-sparing clause is the exact opposite of the GloBE rules' logic. The purpose of the clause is to ensure that a residence country grants a company a credit for tax that would have been due under the source country's generally applicable tax system but was not actually collected because the company was granted a tax incentive by the country of source. 73 Due to the controversy surrounding the desirability of tax competition for real investment, it is still unclear what version of the GloBE proposal would be adopted by countries. In its least effective version, it would contain wide 'substance-based carve outs' and a low minimum tax rate. In this case, the GloBE rules would primarily be an extension of the BEPS Project and would not signify a fundamental departure from the idea that a state is free to impose the tax rate that it considers appropriate-provided that the features of its tax system do not facilitate profit shifting. The main difference is that, under the Harmful Tax Practices regime as described in section III of this article, 'economic substance' of companies is defined and monitored by the state enacting the tax incentive. Under the weak version of the GloBE rules, the assessment of substance is instead undertaken by the home state of the investor and through a unified formula (based on the company's employees and tangible assets). 74 In its most efficient version (without substance-based carve outs and a high minimum tax rate), however, it would create strong incentives for states to adapt corporate tax incentives provided in SEZs (and elsewhere), regardless of whether these encourage profit shifting. Considering the last discussion paper circulated by the OECD, it seems unlikely that a strong version will be adopted any time soon. 75 Nonetheless, the authors will discuss how a strong version would affect SEZs, given that the TLO of international taxation may be developed in this direction in a more distant future.

B. The GloBE rules' potential impact on tax rates in SEZs
Despite the many uncertainties about the future of GloBE, it is possible to reflect on several potential consequences for SEZs worldwide. Even in its strongest version, the GloBE proposal would not 'legally' require any country to eliminate tax regimes that provide for a low rate such as those of many SEZs. However, it would provide a strong 'economic' incentive to the country to increase its tax rate up to the minimum rate as can be explained with a hypothetical example: if a country previously granted a tax exemption or a tax holiday to companies operating in an SEZ with the goal of attracting foreign investment, foreign investors' home countries would begin taxing the income of the subsidiaries operating in foreign SEZs up to the minimum rate. As a result, the tax advantage of investing in the SEZ would be neutralized to some degree. The likely consequence would be that the country operating the SEZ would increase the tax rate up to the globally determined minimum rate rather than forego the tax revenue. Consequently, an increase of corporate tax rates levied in SEZs would be witnessed. 76 This indicates that, in the GloBE Project, countries continue to rely on 'coordinated unilateralism' rather than on binding commitments through multilateral treaties. 77 The legality of SEZ unilateralism is thus not questioned. Yet, the effectiveness of tax incentives in SEZs as a policy tool might be affected in a more fundamental manner, possibly leading to their decline. The prospects of a global adoption of a strong version of the GloBE rules are currently dim. However, the GloBE discussions are representative of a general tendency towards more international embedding of corporate tax policies and, therefore, a pushback against economic unilateralism in corporate taxation.

V. CONCLUSION
Unlike in the international trade and investment regimes, unilateralism has been the norm regarding many aspects of corporate taxation. The TLO of international taxation has remained agnostic for a long time regarding corporate tax rates and corporate tax incentives provided by countries. As evidence from LAC shows, countries have often used corporate tax incentives to attract investment into SEZs. In the context of increasing globalization, concerns about the potential role of such incentives in aggressive tax planning schemes have grown. Therefore, the TLO of international taxation has expanded through soft law standards developed by the OECD and coordinated unilateralism by the EU, which have begun exercising an influence on SEZs all around the world. When these organizations have concluded that the tax incentives in a country's SEZs might facilitate the erosion of other countries' tax bases, they have demanded reforms. In LAC, the SEZ laws of approximately one-third of countries with SEZs have been found incompatible with OECD and EU standards. To formally comply with the international standards, these countries introduced requirements or limitations preventing companies without sufficient economic substance from accessing the tax benefit. Reports published by the OECD and the Council of the EU suggest that, beyond Latin America, many SEZ regimes around the world have been impacted. 78 However, the impact of these reforms should still be qualified as limited in that the most important features of tax incentives in SEZs (such as the tax rate) are not affected. Yet, deliberations on a global minimum tax have been taking place since mid-2019 within the OECD Inclusive Framework, which suggests that further reforms are possible. An agreement on a strong version of such a minimum tax is not likely to be reached in the short term; however, the mere existence of such discussions indicates that a stronger international agreement might be reached in the medium to long term. Such a reform would provide strong disincentives for countries to maintain special regimes with very low tax rates.