2.1 In Europe

Since the very beginning, the EU has pursued the dream of tax harmonisation. Indeed, various proposals were discussed as early as 1962 (1962: Neumark report; 1970: Van den Tempel report; 1975: proposal for a directive on the alignment of tax rates between 45% and 55%).

But it became clear that its attempts were premature, and in 1980 the Commission laconically stated that the attempt at harmonisation was probably doomed (COM(80) 0139) and so it concentrated instead on targeted measures to complete the internal market.

Three proposals were soon adopted: the Merger Directive (90/434/EEC – now 2009/133/EC), the Parent Companies and Subsidiaries Directive (90/435/EEC – now 2011/96/EU), and the Arbitration Procedure Convention (90/436/EEC).

The fate of the proposal for a directive on a common system of taxation on interest and royalty payments between parent companies and subsidiaries in different member states illustrates the often-protracted nature of negotiations with member states: despite revision and a favourable opinion from the European Parliament, the Commission withdrew it because it failed to reach agreement in the Council. A new version appeared in 1998 as part of the ‘Monti package’, which was finally adopted as Directive 2003/49/EC, just before the enlargement.

Meanwhile, in 1991, the Ruding Committee of independent experts was established. Its report recommended a programme of actions to eliminate double taxation, harmonise corporate tax rates and ensure full transparency of the tax concessions offered by member states to promote investment. The Commission made many proposals, all of which were later withdrawn.

In 1996, the Commission launched a new approach to taxation. In the field of company taxation, the main result was the Code of Conduct for Business Taxation, which was adopted as a Council resolution in 1998. The Council also established a Working Group (“WGCC”) to examine cases of unfair business taxation.³ This denoted another parallelism with the OECD’s work, which at that time was dedicating special attention to harmful tax competition.

In 2015, the Council granted the WGCC a new mandate to extend the scope of the EU Code of Conduct and update the criteria. The additional tasks that the WGCC was asked to take on included: (a) issuing guidance on the application of the substantial activities criterion to non IP-regimes (headquarters regimes, holding company regimes, distribution and service centre regimes, banking and insurance regimes, financing and leasing regimes, shipping regimes and fund management regimes) in line with the “modified nexus approach” agreed regarding IP regimes (benefits should be available only when the core income-generating activities are undertaken by the qualifying taxpayer); and (b) revising past EU guidelines on transfer pricing in line with the BEPS Project. The results are expected by the end of 2019.

In 2001, the Commission carried out “an analytical study of company taxation in the European Community” (SEC(2001) 1681). The accompanying Commission notice (COM(2001) 0582) noted that the main problem companies faced was having to adapt to different national regulations in the internal market.

The Commission proposed several approaches to the problem of providing companies a common tax base for their EU-wide activities: home state taxation, an optional common consolidated tax base (CCTB), a European company tax and a compulsory, fully harmonised tax base. In 2004, a working group was set up, and the results of its work were incorporated into a Commission proposal (COM(2011) 0121). The proposed “common consolidated corporate tax base” (CCCTB) meant that companies would benefit from a system with a central contact point to which they could submit their tax returns and refund claims. They would also be able to consolidate all their profits and losses made in the EU. But Member States would retain full responsibility for setting their own rates of corporate tax.

In April 2012, the European Parliament adopted its legislative resolution on this proposal. But no action ensued.

In June 2015, to give fresh impetus to the negotiations in the Council, the Commission came up with a strategy for relaunching the CCCTB proposal in 2016. The Commission opted for a two-step process, separating the common base and consolidation elements with two interconnected legislative proposals: one on a common corporate tax base (CCTB) and one on a common consolidated corporate tax base (CCCTB). The revamped proposals, once adjusted to take account of the OECD’s work, was also designed to address tax avoidance by closing regulatory gaps between the national systems and thus put a stop to common tax-avoidance arrangements.

In March 2018, the Commission proposed two Council directives designed to ensure fair taxation of companies offering digital services.⁴

Much better results have been achieved in pursuing fair taxation, tax transparency and measures to combat tax avoidance. After the 2008 financial crisis, attention turned to combatting tax avoidance and to the equitable taxation of companies. Increased transparency was seen as one way of achieving this, as evidenced in the Tax Transparency Package of March 2015, which included the Council Directive on the automatic exchange of information on tax rulings between member states (Directive (EU) 2015/2376) and the notice on tax transparency to fight tax evasion and avoidance (COM(2015) 0136).

In 2015, the Commission adopted an action plan for a fair and efficient corporate tax system across the EU (COM(2015) 0302), with provisions for reforming the corporate tax framework to combat tax abuse, ensure sustainable revenue and support an improved business environment in the internal market.

In April 2016, the Commission proposed an amendment to Directive 2013/34/EU on the disclosure of income tax information by certain types of companies and branches. The proposal requires MNEs to publicly disclose some parts of the information submitted to tax authorities.

In June 2017, the Commission proposed new transparency rules for intermediaries (e.g., consulting firms, banks, lawyers, and tax advisors) that design or sell potentially harmful tax schemes, following a request for a legislative proposal in the European Parliament’s resolution (TAXE 2).

The Council then adopted the proposal in May 2018 and, in December 2017, published the first ever EU list of non-cooperative jurisdictions. The list is updated regularly.

EU rules on State aid have proved to be another effective tool to combat tax avoidance. Aid exists when a Member State uses public resources to promote certain economic activities or protect national sectors.⁵ This aid can distort competition by favouring certain companies over their competitors.

In 1988, the Commission published the Notice on the Appli­cation of State Aid Rules to Measures Relating to Direct Business Taxation, emphasising its determination to apply these rules rigorously in the field of taxation. The Commission has since issued a signifi­cant number of regulations, frame­works and notices regarding fiscal State aid.

In assessing fiscal aid, it is crucial to ascertain whether the measure: (a) grants an economic advantage, and (b) is selective. An economic advantage exists when the beneficiary’s net financial position is improved because of the measure. In terms of fiscal State aid, an economic advantage may be granted through different forms of reduction (temporary or permanent exemptions) of an undertaking’s tax burden (fiscal holidays, tax credits, reduced tax base, reduced tax rates, accelerated deprecia­tion, etc.) and by reducing the tax base, the tax rate applied or the tax due.

Selectivity exists when a Member State applies more lenient tax rules to a particular company or to companies of a particular cluster. Selectivity may be “de jure”, i.e., when it results directly from the law, or de facto, i.e., when the law is not selective per se, but the effects of the measure are such that it favours a cluster of businesses.

The WGCC’s work and State aid investigations have much in common, especially as to their scope of attention. Harmful tax competition generally takes the form of reverse discrimination, as it treats foreign undertakings better than domestic ones, whereas State aid typically favours local undertakings. But several exceptions to both rules exist. One area of quasi-matching overlap is that of tax rulings in which the focus was on their transparency.

The issue was seen to be so important that EU member states adopted a directive on the mandatory automatic exchange of information regarding cross-border tax rulings and advance pricing agreements (Directive 2015/2376). However, this did not completely remedy the problem. For instance, the information included in an individual tax ruling does not typically provide the full picture regard­ing tax planning strategies adopted by an MNE or a group of companies. This perceived loophole was remedied with the approval of DAC6 (Directive 2018/822), regarding the man­datory automatic exchange of information in the field of taxation in relation to reportable cross-border arrange­ments.

More recently, as a follow-up to the gigantic OECD’s BEPS exercise, the Council approved two important directives known as Anti-Tax Avoidance Directive I and II (“ATAD I and II”).

ATAD I established a minimum standard for member states to implement in five areas:

• General Anti-Abuse Rule (GAAR): arrangements whose main purpose is to obtain a tax advantage that circumvents otherwise applicable tax provisions and are non-genuine when considering all relevant facts and circumstances.

• Interest deductibility limits: limits to the deduction of exceeding borrowing costs to 30% of taxable earnings before interest, taxes, depreciation and amortisation (EBITDA), with several optional provisions such as grandfathering of existing loans and carryforward and carryback options.

• Controlled foreign company (CFC) rules: rules aimed at attributing certain income to a parent taxpayer when a subsidiary (or permanent establishment) is more than 50% controlled by the parent company and the CFC is in a lower tax regime.

• Exit taxes: rules applicable in circumstances in which a member state imposes a tax corresponding to the market value of the transferred assets, at the time of exit, minus their value for tax purposes.

• Hybrid mismatch rules: rules applicable when a taxpayer arrangement results in a double deduction or a deduction without income inclusion due to differences in the legal characterisation of a financial instrument or entity.

ATAD II introduced additional requirements over and above the original ATAD package, establishing rules to neutralise hybrid mismatches when at least one party is a corporate taxpayer in a EU member state, thus expanding ATAD to third countries. ATAD II also addresses hybrid permanent establishment (PE) mismatches, hybrid transfers, imported mismatches, reverse hybrid mismatches and dual resident mismatches.

2.2 At the OECD

The scope of tax work at the OECD has also changed over time.

For more than 50 years, the OECD Centre for Tax Policy and Administration dedicated almost exclusive attention to double tax treaties and transfer pricing guidelines.

Toward the end of the last century, the focus moved to tax transparency and — albeit slowly due to strong political resistance — to aggressive tax competition.⁶ Issues like the exchange of banking information among tax administrations and the combat of international tax evasion became priorities for the CTPA.

Tax evasion, tax transparency and aggressive tax competition are global issues that require an inclusive approach. This was why in 2002 the OECD formed the Global Forum on Transparency and Exchange of Information for Tax Purposes (Global Forum), with 150 participating countries.

In 2002, the OECD created the Forum on Tax Administration (FTA), bringing together tax commissioners from 51 OECD and non-OECD countries. The FTA is a forum through which tax administrators can identify, discuss and influence relevant global trends and develop new ideas to enhance tax administration, tax compliance and tax collection around the world.

The creation of the FTA gave a significant boost to the cooperation of tax enforcers of OECD countries, which proved effective in the exchange of information and the setting of common actions to reduce international tax avoidance by MNEs.

After the financial crisis, many raised the criticism that globalisation enabled large MNEs to avoid paying their fair share of tax by exploiting gaps and mismatches in tax rules and artificially shifting profits to low or no tax jurisdictions, where little or no economic activity occurs.⁷ This was no longer tolerable.

The mounting aversion against this phenomenon was at the base of the G20/OECD initiative to address the problem of Base Erosion and Profit Shifting, which rapidly became a truly inclusive effort involving 125 countries and jurisdictions.

The Inclusive Framework on the OECD/G20 Base Erosion and Profit Shifting (BEPS) Project established in 2016 brought together more than 115 jurisdictions to ensure implementation of the BEPS action items and alignment of tax with where value is created. The OECD/G20 BEPS Project focuses on addressing the tax challenges of the digitalisation of the economy. This is an area in which the collaborative approach of the Inclusive Framework is needed to secure a long-term, consensus-based solution.

In its efforts, the OECD partnered with the G-20 to ensure effective and immediate legislative adoption of the new standards. The pillars of this partnership are: (a) enhancing tax transparency; (b) addressing tax avoidance; (c) promoting tax policies for strong, sustainable and inclusive growth; and (d) supporting tax as a tool for development.

The need to preserve the integrity of domestic tax basis and to eliminate the opportunities for international tax evasion and avoidance ignited the current era of unprecedented international cooperation.

3.1 Corporate tax rates

Until the mid ’90s, Italy was moving in the opposite direction from that of other industrialised countries. Due to budgetary constraints, at that time its overall corporate tax rate was among the highest in Europe; corporate income tax was set at 37%, but businesses were also subject to a 16.2% local income tax and a 0.72% asset tax. The combined tax rate was near 55%, one of the highest in the world.

The statutory CIT rate was still 37% in 2000: by 2017 it had dropped 13 percentage points, making Italy one of the OECD countries that experienced the sharpest decrease in the statutory rate since 2000.

In terms of contributing to revenue, CIT reached its peak in 2007 (3.1% of GDP), decreased rapidly thereafter and is nowadays slightly above 2% of GDP. This decrease reflects the nominal rate reduction and the sloppy performance of the domestic economy.

Throughout the years, the rate reduction in Italy was paired with a series of gradual changes to the design of corporate tax.

The decrease in CIT standard rates did not result in an equally sharp decrease on CIT revenue. This reflects two factors: (a) CIT revenue depends not only on rates but also on the basis; when rates decline and the basis grows, revenue tends to remain stable; and (b) corporate tax revenue is largely driven by the economic cycle.

The following chart illustrates this phenomenon in Italy.

3.2 Elimination of economic double taxation and other selective measures the determination of/for determining the tax basis for corporate tax

In 1998, Italy implemented a selective reduction of corporate taxes, aimed in particular at reducing the debt-equity tax bias, i.e., the tax distortion between debt and equity financing.

The main changes were: (a) enacting a restricted version of the Allowance for Corporate Equity (“ACE”) called the Dual Income Tax system (“DIT”); and (b) abolishing local taxes and replacing them with a regional tax (IRAP) on an enterprise’s added value, at that time at 4.25%. The 1998 tax reform also entailed an important revision of the rules on corporate reorganisations, with the aim of making the tax system neutral for all reorganisation patterns.

The Italian DIT system worked as a dual rate schedule, in which the profits were divided into two portions. The first approximated normal profits, representing the opportunity cost of new investments with equity financing, and was taxed at 19%. The second portion was residually computed by deducting normal profits from total profits and was taxed at the ordinary rate of 37%.

This mechanism aimed to guarantee tax neutrality as to the choice between debt or equity financing and between productive factors. Critics of the DIT system objected that the dual rate system applied only on the incremental equity (and not on the entire equity base of the company) and that the effective tax rate for each company was not transparent and stable.

The DIT regime remained in existence until 2004, when the second important corporate tax reform was enacted with Legislative Decree No. 344/2003, which introduced the new corporation tax called IRES. One of the policy choices behind IRES was to simplify and level the tax rules on corporate profits. Instead of selective measures to favour specific enterprises or business models, it proposed a general and equal reduction of tax rates for corporate profits.

Apart from this policy twist and the different names used,⁸ IRES has a very different fundamental structural feature from its predecessor (“IRPEG”). The full imputation system to avoid economic double taxation of corporate profits was changed to a “participation exemption” system, similar to what some other OECD countries implemented in the 1990s (e.g., Finland, France and Germany).

In the IRPEG model, recipients of dividends distributed by an Italian resident company were obliged to report that dividend in their yearly tax return but entitled to an imputation credit. The credit amount was a percentage of the dividend received and was first added to the recipient’s taxable income and then deducted from the tax due on the grossed-up taxable income. If the credit amount exceeded the gross tax on the shareholder’s total income, the shareholder was entitled to a refund of the excess. However, the IREPG was introduced in 1973, whereas the imputation credit became law only four years later in 1977.

This mechanism was such that the tax paid by the company was not finally acquired by the tax authority when paid by the company, as the tax authority could be obliged to fully or partly refund it to the shareholders. In essence, IRPEG was an advance payment by the company for the taxes that the shareholders owed on that dividend.

This model had several inefficiencies.

First, it was extremely complex. This was mainly because profits distributable to shareholders very often do not represent profits on which the distributing company is to pay corporate tax.

For instance, if a company was entitled to a tax holiday, when it distributed profits to the shareholder granting a credit for unpaid taxes was far too generous. Apart from tax breaks, the amount of distributable profits in a given year is greater (or lower) than the profits that the company has been (or will be) taxed for. This because numerous differences exist — whether temporary or permanent — between pre-tax accounting profits and taxable profits. Moreover, an efficient system needed to be designed to allow a company to pass on a tax concession to its shareholders, if the legislator so wished.

The imputation system had to take these occurrences into consideration and regulate them properly. In the “IRPEG era”, this was achieved in two different ways, equally complex.

In 1983, a new law required the distributing company to calculate and pay an “equalisation tax” (maggiorazione di conguaglio) when it distributed profits to its shareholders. The equalisation tax was meant to equate the corporate tax paid by the company to the imputation credit available to the shareholders. The regulations on the equalisation tax were extremely complex. In addition to its inherent complexity, the system was such that the equalisation tax could only approximate the imputation credit available to the shareholders but never match it with exact precision. This because some categories of shareholders were not entitled to the imputation credit because they were tax exempt (e.g., a pension fund) or liable to tax by way of a final withholding tax at source (which is the case for all non-resident shareholders). In all such cases, the nature of the equalisation tax changed: despite being designed to ensure that the imputation credit for shareholders did not exceed the taxes paid by the company, it resulted in an unjustified additional tax burden on distributed profits.⁹

The system was slightly modified in the 1998 tax reform. The company was requested to keep track of the CIT it suffered in each tax year. This amount was put in a “basket” (“Basket A”) that also included the tax that the company did not pay under specific tax rules that, by law, should have given right to a refundable imputation credit for shareholders even though no actual tax was paid. The unpaid taxes that the law did not want to generate a refundable tax imputation credit for shareholders were put in another basket (“Basket B”). When the company distributed a dividend to its shareholders, it also distributed portions of Basket A and B imputation credits, both of which the shareholders could use to offset tax due on the dividends. However, if the imputation credit exceeded the shareholders’ tax liability, only Basket A credit was refundable.

The system was made somewhat simpler and most inefficiencies of the equalisation tax removed, but its structural weaknesses remained.

Indeed, the imputation tax system was structurally unable to work properly in cross-border situations (i.e., when a company and a given shareholder reside in different jurisdictions). Additionally, it addressed the economic double taxation only on dividends — not on capital gains.

As to its structural inability, the imputation system appeared to work perfectly in purely domestic situations. But a shareholder resident in Italy was eligible to use the imputation system for Italian-sourced dividends only. When dividends arrived from abroad, no credit was available. But why should Italy have granted credits for taxes that the distributing company paid to another tax authority? And how could the taxes paid by that company be calculated?

No dividend tax credit for foreign-sourced dividends resulted in a higher taxation of equity investments made abroad. This discrimination resulted in an in-built bias in favour of domestic investments. This was not tolerable, not to mention that it conflicted with EU fundamental freedoms.

But EU complications (and possible discriminations including outside the EU) existed also in cases in which the distributing company was in Italy and the shareholder was abroad. Indeed, an Italian resident shareholder was entitled to the credit, but a non-resident shareholder was not. This discrimination was possibly tolerable as it was inherently coherent with the fundamental principle of Italian taxation, but it remained a discrimination.

As mentioned, another issue was that, when available, the imputation system eliminated the economic double taxation on distributed profits, but not on capital gains.

The rationale for aligning the tax regime of dividends and capital gains on shares is rather simple. The tax exemption of distributed profits eliminates economic double taxation. But when a company retains its after-tax profits, the value of its shares increases correspondingly. If a shareholder sells the shares before the dividend is distributed, the right to collect it tax-free goes to the purchaser, and the seller therefore requests an increased sale price. That constitutes a capital gain that has the same exact nature as the dividend and must therefore be taxed in the same way. Even when the capital gain does not correspond to profits already generated by the company, it represents one of the following (or both): (a) the existence of hidden reserves that would sooner or later be generated and thus taxed; or (b) the company’s ability to generate profits in the future, which would be subject to tax. In either case, the origin and nature of the capital gain remain the same.

The above considerations were the main drivers of the 2004 reform, whose pivotal change was replacing the imputation tax system with a participation exemption system — i.e., the system now in force in almost every country worldwide.

The 2004 reform moved the centre of gravity of corporate taxation from shareholders to the company and aligned the corporate tax model to that prevailing since the start of the century. As mentioned, from 1992 eliminating economic double taxation of corporate profits between companies residing in two different EU member states required either a participation exemption or an imputation credit system. All EU Member States (including Italy) chose the participation exemption, but Italy maintained the imputation system for purely domestic situations. The two systems could not co-exist.

The change from the imputation system to the dividend received participation exemption entailed several other important changes to Italian corporate taxation.

First, the exemption had to be extended to capital gains. The exemption for capital gains on shares required that tax losses on shares, whether incurred or accounted for as write-downs, had to be non-deductible.

This meant that a corporate shareholder could not make effective use of the (tax) loss incurred by its participated company. Indeed, when a subsidiary incurs an accounting loss, this generally results in a tax loss, which can be carried forward to offset the subsidiary’s future taxable profits. Before the participation exemption was introduced, the parent company could make immediate use of that loss by writing down the value of its shareholding and deducting the accounting loss from its taxable base.¹⁰ As that possibility no longer existed, it had to be replaced with another corrective mechanism.

This was one of the reasons for introducing the tax consolidation system,¹¹ one of the other very important changes of the 2004 reform. The optional tax consolidation group aims to provide a business the same overall tax treatment whether it operates as a single entity or as a group. A tax consolidation group thus eliminates the impediments to the most efficient use of tax attributes (tax losses, tax credits, etc.) belonging to different entities of the group.

The 2004 corporate tax reform was harshly criticised by the then opposition parties. Specifically, the introduction of the participation exemption was considered a sort of unjustified gift to plutocrats, which could enjoy no tax on their large gains from selling shares. The reality is that replacing the imputation system with the participation exemption increased (rather than diminished) yearly revenue from corporate taxation by some EUR 3.5–4 billion.¹²

The 2004 corporate tax reform also abolished the DIT system and introduced a mechanism (known as the “thin capitalisation rule”) that limited the deduction of excessive interest expense. Instead of correcting the debt bias by encouraging the company’s capitalisation through incentives, the new system penalised excessive leverage from shareholders.

Finally, the 2004 reform reduced the CIT rate to 33%.

Since then, the system has remained relatively stable, with several minor amendments introduced every now and then, the most important of which are the following.

In 2008, the thin capitalisation rules were abolished and replaced by a generally applicable ceiling to the deduction to net interest expense exceeding 30% of EBITDA. Interest expense that exceeds the threshold can be carried forward to future tax years or made available to other affiliated entities within a consolidated tax group. The corporate tax rate was reduced to 27.5% and the taxable base increased by prohibiting taxpayers from deducting accelerated and anticipated depreciation allowances.

In 2011, an amended form of Dual Income Tax was reinstated, renamed as "Aiuto alla Crescita Economica" (‘a help to economic growth’), whose acronym (ACE) mimics that of Allowance for Corporate Equity.

The main difference between the 1998 DIT and the 2011 ACE is that the 1998 DIT allowed a portion of the corporate profits (those representing the opportunity cost of new investment financed with incremental equity) to be taxed at a lower rate. Conversely, the 2011 ACE allowed the deduction of the opportunity cost of qualifying incremental equity, whereas the taxable base was subject to the same proportional tax rate.

In 2017, the nominal rate was further reduced to the current one of 24%. In 2018, the ACE relief was once again abolished.

3.3 Measures to combat tax evasion and tax avoidance

Tax evasion through tax havens and offshore bank secrecy is something of a national sport in Italy, especially since currency exchange controls were removed.

This comes as no surprise, as not only is Italy surrounded by countries who make others’ tax evasion one of their main GDP contributors (Switzerland, Austria, MonteCarlo and Liechtenstein), but it even hosts one within its border (San Marino).

Due to this territorial peculiarity, Italy has always been at the forefront of the fight against international tax evasion and avoidance.

Indeed, as early as 1991, Italy introduced its first rule limiting the deduction of costs and expenses deriving from transactions with related entities in black-listed countries. At that time, very few other countries had similar measures. The rule was in force for 25 years, undergoing various amendments to widen its scope of application.

Curiously, the measure was abolished in 2016, when the international response to base erosion and profit shifting reached its peak.

Italy introduced anti CFC rules back in 2000, shortly after the 1998 OECD Report “Harmful Tax Competition: an Emerging Global Issue” was published. Their scope of application was then extended to affiliated companies (i.e., companies in which the Italian taxpayer held 20%–50% of the stock) in 2004, something unheard of anywhere else in the world.

In 2010, CFC legislation was again widened to include EU subsidiaries. And just when international consensus was about to be reached and widespread introduction of CFC rules suggested, those in force in Italy for 25 years were relaxed. Likewise, the CFC rules applying to affiliated companies were abolished in 2016.

As seen above, Italy introduced limitations to the deduction of interest expense as early as in 2004 and modified them in 2007 in a way that closely resembles those mandated as the European minimum standards by the ATAD Directives.

Italy was also ahead of international trends regarding hybrid financial instruments, having introduced a specific rule in 2004. The rule states that the tax exemption for foreign-sourced dividends applies to the extent that in the taxpayer’s country of residence, the dividend is not a deductible cost item, even partially. The rule aims to prohibit one of the several cases of “deduction-not inclusion” mismatches addressed by BEPS and ATAD rules.

Indeed, the BEPS project included recommendations on addressing preferential ruling regimes, hybrid mismatch arrangements, abuse of double tax treaties, the avoidance of permanent establishment status, and similar arrangements.

To ensure/guarantee uniform minimum standards of adoption of the OECD recommendations within Europe, between 2016 and 2017 the EU Council approved two directives (ATAD I and II) to implement most BEPS recommendations.

The ATADs address the following anti-BEPS actions: (a) limitation for the deduction of interest expense; (b) entry and exit taxation; (c) CFC; (d) general anti-abuse rules; and (e) hybrid mismatches.

As mentioned, Italy had rules governing each of the above topics for many years before the ATADs came into force, some of which mirror those suggested by the ATADs.

The first measure (on interest expense) limits the deduction of net interest payments if they exceed 30% of the corporate taxpayer’s EBITDA. Italy introduced a system almost identical to the EU standard already in 2007. It can be reasonably assumed that ATAD I took the Italian rules and experience as a model to design the EU-wide measure.

Indeed, the implementation of the EU ATADs standards required only the following marginal changes to the existing rules:

1. EBITDA must now be quantified based on the tax values, i.e., reflecting the corporate income tax adjustments applied to the EBITDA computed from an accounting perspective. The former rules were based on accounting EBITDA;

2. net interest expense exceeding 30% of EBITDA, not deducted in a year, can now be carried forward for five years. Under the former rules, it could be carried forward for an unlimited time; and

3. interest expense subject to limitation now includes interest capitalised on the cost of purchased assets, and interest on trade payables.

The same is true for entry and exit tax rules, i.e.; the rules addressing the tax consequences of transferring a company’s fiscal residence to or from Italy. The issue was well known and addressed in Italy. The new ATAD rule essentially envisages the possibility to spread tax payments due over five years; Italy previously allowed six years.

With regard to CFC, Italy introduced its first CFC rules in 2000 and amended them several times (in 2004, 2010 and 2015). The ATAD CFC rules differ from the Italian ones in some respects, none of which are of major significance. However, the need to implement ATADs enabled Italy to reshape and streamline its pre-existing CFC rules as follows:

1. CFC rules now apply also to Italian permanent establishments of non-resident persons if the PE effectively holds controlling interests in non-resident companies;

2. the notion of control was broadened to include cases in which the Italian person is entitled to more than 50% profit participation rights in the foreign entity;

3. a CFC exists if the following two conditions are met:

   1. the foreign entity’s effective tax rate is lower than 50% of that applicable if the entity were tax resident in Italy; and

   2. more than one third of the foreign entity’s revenue is “passive income” (dividends, interest, royalties, capital gains, etc.); and

4. the CFC rules do not apply if the entity carries out a substantive economic activity, with adequate staff, equipment, assets and premises.

As to tax law abuse, Italy decided not to transpose the general anti-abuse rule (“GAAR”) contained in the ATADs as it held that the existing GAAR was fully in line with the EU standards.

The evolution of the public response to tax avoidance in Italy is quite interesting. Tax practitioners and commentators always highlight the traditional distinction between tax planning (legal, however unpopular) and tax evasion (illegal, however widespread). Tax avoidance is somewhere in between. Curiously, today’s wider public view tax avoidance as when companies (especially large ones do not pay what they “ought” to pay, which they consider far worse than tax evasion by “normal” people.

Until the beginning of the ’90s, it was accepted in the tax profession, tax administration and tax courts that tax laws were to be interpreted literally and strictly, and that tax provisions so construed had to be applied to transactions that were themselves to be analysed in a very legalistic way.

Tax laws contained numerous loopholes ready to be exploited against high tax rates. Tax avoidance merchants created increasingly sophisticated tax avoidance schemes. As the scale and artificiality of these schemes increased, powerful concerned voices started to question, then criticise and finally overturn the legal basis underlying these schemes.

Italy introduced its first quasi-general anti-avoidance rule in 1997. It had two limitations: (a) it applied only to income taxes, and (b) it did not cover facts occurred before its effective date. In the first decade of the new century, Supreme Court judges were asked to check the legality of widespread cases of dividend stripped and dividend washing carried out before 1997 (i.e., when they were not subject to any express anti-avoidance legal scrutiny). Their response was very surprising: the Grand Chamber of the Supreme Court ruled that the Italian Constitution has implicitly prohibited tax avoidance since 1948, when the Constitution was enacted.

Hence, dividend stripping and dividend washing trades were held unlawful, even if carried out before anti-avoidance legislation came into force.

Apparently, until then no-one in Italy had noticed this astonishing peculiarity of the Italian Constitution. Certainly not the Parliament, which was unaware that the Constitution already contained a GAAR when it introduced the first quasi-general anti-avoidance rule in 1997. In any event, the Italian legislator approved a new, fully-fledged GAAR in 2015, which closely resembled that contained in the ATADs.

The ATADs minimum standards on hybrid mismatch arrangements are extremely complex, and Italy decided to implement them without any substantial change. Most of these anti-hybrid rules are brand new also in Italy, with one notable exception. As mentioned, as part of the 2004 tax reform, a rule was introduced to deny the tax exemption of foreign-sourced dividends, which entitled the distributing company to deduct the payment in its country of residence. For all intents and purposes, this was a hybrid mismatch preventing rule. It was in force in Italy for some 15 years before the ATADs and for 10 years before the amendment by Council Directive 2014/86/EU of Art. 4(1)(a) of the EU parent-subsidiary directive, which now stipulates that the Member State of the parent company is to tax received profits distributions to the extent that the profits are deductible by the subsidiary.

3.4 Web Tax

The 2015 BEPS Action Report 1 is entitled “Addressing the Tax Challenges of the Digital Economies”. Since its publication, the many public policy challenges of digitalisation have been discussed to help policymakers better understand digital transformation and develop and implement a resilient framework that fosters a positive and inclusive digital economy and society.

Four years on, and consensus on such formidably complex issues (such as new taxation nexus and profit allocation, users’ participation, marketing intangibles, and fractional apportionment to significant economic presence) is far from being reached.

In this period, Italy experimented with not one but two Italian ways of taxing digital profits, neither of them yet effective.

The 2019 Budget Law introduced a tax on digital services which replaced the previous measures envisaged in the 2018 Budget Law, which did not enter into force as the implementing decree was never enacted.

The new digital service tax is meant to apply to digital transactions performed — individually or at group level — by taxable persons established in or outside of Italy in the course of their business. The deadline for issuing the implementing legislation was 30 April 2019, which passed uneventfully. The hope is that nothing new happens until the EU agrees on something.