The Italian digital tax: some brief reflexions

by Alessio Persiani

The Italian government has recently announced plans to introduce in our legal system a digital tax, namely a tax on non-resident companies operating in the digital economy sector and carrying on an economic activity in the Italian market through communication technologies.

The proposal should be put in the wider scenario of the ongoing international debate on this topic. Reference is made, in particular, to the analysis currently underway with the OECD, where the tax challenges posed by the digital economy constitute one of the crucial points of the larger project aimed at preventing the erosion of the taxable base at an international level and the shifting of profits in jurisdictions with more favorable tax regimes [1].

According to information available as of today, the Italian government, after having initially expressed the intention to wait for the results of the OECD analysis, seems now willing to proceed autonomously, introducing ad hoc tax measures for the enterprises of the digital economy in the Italian legal system.

Based on the information currently available, the provisions of the Italian digital tax – to be included in the 2016 Budget Law the government is currently drafting and to be applied as from year 2017 – should be largely shaped after the contents of the bill recently proposed in the House of Deputies by MPs Quintarelli and Sottanelli and named “Rules on the prevention of online tax avoidance”.

The main content of this bill is represented by an amendment to the definition of permanent establishment provided for by Article 162 of the Income Tax Act (Testo Unico delle Imposte sui Redditi, hereinafter TUIR) [2]. In this respect, it is useful to remember that the existence of a permanent establishment constitutes the minimum requirement to levy Italian corporate income tax on the Italian-sourced income of a non-resident company. According to the Quintarelli-Sottanelli bill, an Italian permanent establishment should be considered existent in any case (regardless, therefore, of the circumstance that the requirements of the “traditional” notion of permanent establishment are not met) “if a non-resident company has a continuous presence of online activities, for a period not shorter than six months, so as to generate in the same period payment flows directed to it […] for an amount not lower than five million Euro”.

In addition, and in order to induce the non-resident company to spontaneously declare the existence of an Italian permanent establishment, the bill provides for a 25 per cent withholding tax on payments directed to non-resident companies for goods and services purchased online, requesting Italian financial intermediaries to act as withholding agents. Such withholding tax – and this clearly sheds light on the intention of inducing the non-resident companies to declare the existence of an Italian permanent establishment – does not apply if and insofar as such non-resident companies have an Italian permanent establishment pursuant to Art. 162 TUIR.

This is not the place for an in-depth analysis of the bill, given that the final version of the provisions of the Italian digital tax may well be different, even significantly. Anyhow, it seems appropriate to highlight certain doubts that the Quintarelli-Sottanelli proposal raises; doubts related, in particular, to the relationship between domestic tax provisions and provisions included in the double tax conventions Italy concluded with the other countries of the international community. In this respect, it is worth remembering that, on the basis of the well-established case law and major scholars’ approach, the provisions included in double tax conventions take precedence over domestic rules conflicting with them on the basis of the lex specialis principle. As the notion of permanent establishment is provided – not only under the aforementioned Article 162 TUIR – but also under all double tax conventions concluded by Italy, an amendment of that notion limited to the domestic provision will likely prove to be ineffective, since the more favorable (for the taxpayers) notion provided for by double tax conventions will prevail over the new Art. 162 TUIR by virtue of the aforementioned lex specialis principle. In other words, the “new” and wider concept of permanent establishment of Art. 162 TUIR will be relevant in a very limited number of cases, id est  for the taxation of the Italian-sourced income of foreign companies resident in countries which have not concluded a double tax convention with Italy [3].

By the same token, also the provisions of the final withholding tax of 25 per cent on payments directed to foreign companies for goods and services purchased online raise doubts on their compliance with the provisions of double tax conventions. If – as it seems preferable on the basis of a number of arguments related to the Italian income tax system considered as a whole – the income the non-resident entity derives from its Italian economic activities qualifies as business income, it is taxable in Italy only if and to the extent that an Italian permanent establishment exists. However, and by definition, an Italian permanent establishment is absent in this case [4]. Moreover, even accepting the qualification of such income as “other income”, it seems that the referred withholding tax does not comply with the distributive rules provided under double tax conventions, as Article 21 of the OECD Model Convention – which the double tax conventions concluded by Italy are shaped after – attributes taxing rights on “other income” exclusively to the state of residence of the recipient.

In light of this, if Italy actually intends to levy tax on the Italian-sourced income of the enterprises of the digital economy through an enlargement of the territorial scope of its income tax (id est, through a widened notion of permanent establishment) it would be appropriate to pursue this approach in the context of the BEPS project currently underway at OECD level. Only this way can result in the simultaneous change of the provisions dealing with the notion of permanent establishment included in all the double tax conventions concluded by Italy [5] and, therefore, in the adoption of a tax measure really affecting the non-resident companies operating in the digital economy.

[1] Reference is to the “Base Erosion and Profit Shifting” (BEPS) project, as formalized in the OECD document “Addressing Base Erosion and Profit Shifting”, Paris, 2013. The first step of such a large project (“Addressing the Tax Challenges of the Digital Economy “) is constituted by the analysis of the distinctive features of the companies operating in the digital economy and the identification of the most suitable instruments for levying taxes on those enterprises in the various states in which they operate.

[2] Approved by Presidential Decree December 22, 1986, n. 917.

[3] As mentioned in the text we are dealing with a very limited number of cases, id est those countries Italy qualifies as “tax havens” and which has not concluded a double tax convention with. None of the larger companies (so called big players) of the digital economy (Google, Facebook, Apple, Amazon) would fall in such cases, since they are tax resident in states that have entered into a double tax convention with Italy (e.g., Ireland, Luxembourg, Netherlands).

[4] As mentioned, the withholding tax is provided in order to induce non-resident companies to spontaneously declare the existence of a permanent establishment in Italy and it does not apply in case such an Italian permanent establishment exists. Therefore, and by definition, the withholding tax apply if and insofar as a permanent establishment is absent.

[5] The technical solution to achieve such a result could be constituted by the conclusion of a multilateral treaty among all OECD countries intended to amend the provisions on permanent establishment contained in all the bilateral double tax treaties these countries have signed among them.

21 September 2015