On the 6th of August, 2020, a drafting committee of the UN Committee of tax experts submitted a proposal for taxing income from automated digital services. This is the newest saga in the ever unfolding effort to tax income of digital companies or the digitalised economy, if you prefer. The OECD's so called unified approach is stuck in the power politics of the continents. Impatient of prolonged deliberations amongst the disparate 100+ members of the so called Inclusive Framework member countries (IF), the OECD Secretariat officials came up with their own version to solve the issue and called it the unified approach and then asked the IF member countries/jurisdictions to endorse the same. The OECD proposal is so complex that it baffles ordinary people and the proposal has been more or less roundly criticized by all and has satisfied none.

Many organisations/countries are now moving forward with their own proposals and one does not really know whether any consensus can be arrived at and if so, when. The UN tax committee, on its part had in 2017 already decided to work independently on the tax consequences of the digitalised economy. A sub-committee was also formed but till recently there has not been much headway. Traditionally, the UN tax committee ends up endorsing more or less the OECD point of view with some changes in favour of the developing countries.

This time, however, things seem to be taking a somewhat different turn. The OECD's unified approach runs on two pillars- pillar 1 and pillar2. Pillar 2 concerns a minimum global tax that MNC companies are expected to pay, failing which the country of residence gets a right to tax a minimum % of profit. There is nothing much in pillar 2 for countries that are not capital exporting unless there is agreement that source countries will also get a right to disallow base eroding payments that are not adequately taxed. Apparently, the work on Pillar 2 does not require consensus and the work is progressing on a parallel track.

The Pillar one proposal finally recognised that digitalised companies earn a lot of revenue in market countries without the need for any physical presence therein. The concept of physical presence being the cornerstone of the current international distribution of taxing rights, can be easily exploited due to fragmentation of activities amongst group companies following the single entity concept and parking IP rights in subsidiaries in tax favoured jurisdictions. Therefore, the profit attribution in terms of the existing transfer pricing rules was also not sufficient to attribute any meaningful taxable profit to source countries. Non-profit organisations have been spearheading the campaign for a change in the status quo in this regard for some time especially after the launch of the BEPS project.

Under Pillar 1, the OECD finally agreed to some changes in the current dispensation of international taxation of cross border business profits in the digitalised business environment. To a limited extent, it also deviated from the single entity concept and its dogmatic opposition to any deviation from the application of the arm's length standard.

According to the OECD proposal, market (source) countries will be entitled to three elements of profit - Amounts A, B and C. Out of these, only amount A constitutes a new taxing right in the changed circumstances and is really the most important element. Amounts B and C that market countries will also be able to tax, will be determined following the traditional arm's length approach and the physical presence requirement of the old system and have been put in as part of the Pillar 1 proposal apparently for reducing disputes but one suspects that these were actually put in to put pressure on the developing countries to accept the OECD's insistence of mandatory binding arbitration- a classic manipulation tactic. It seems that the developing countries have fallen for the bait although mandatory binding arbitration will now be replaced with a system of mandatory binding dispute resolution with a strong peer review process to monitor implementation. According to Pascal Saint Amans, the standard required will be very close to the mandatory binding arbitration.

But according to the OECD proposal, only certain types of digitalised businesses will be in scope for the additional right (Amount A) to accrue - those offering standardized and automated digital services - search engines, social media platforms, online advertising, cloud computing etc. These would have affected mostly the US based digital companies. Therefore, under the US pressure, it was decided that the other type of businesses that will be affected are the so called consumer facing businesses - those deriving income from sale of goods or provision of services for the personal use of the consumers as opposed to use for business purposes. Thus, the scope of application of the new rules itself will be rather limited.

According to the OECD's own impact assessment, the combined effect of amounts A.B and C of the Pillar 1 is minimal and there will be only a slight increase in total revenue. However, in order to get this pittance, countries will have to pass through a number of mind-boggling and complex thresholds, profitability percentages and other tests.

In terms of the July 2020 paper released by the OECD, the departure from the current practice is that the calculation will be at the group level rather than separate entities of the MNC group and the allocation is on a fixed % level rather than through the ALP application. However, there are at least 6 steps for arriving at the amount A that the market jurisdictions are supposed to get.

The first is a turnover test - apparently at euro 750 million euros as in country by country reporting. Only if the group turnover exceeds this threshold, then further calculation will be required. The second filter is the activity test . As mentioned earlier only automated digital service companies or consumer facing businesses will be affected. In case a company has mixed activities, then the activity test will be applied and only ADS/CFB activities will be carved out.

Then there is a de minimis revenue test and only the in-scope revenues from aggregated ADR/CFB activities exceeding an as yet undefined amount will be taken into account for calculating the Amount A liability.

After that, a business line profitability test will be applied and if the profitability is above a certain undefined percentage, then only we will move to the next stage. But even this is not enough.

There will then be a de minimis test on aggregate residual profit . If such aggregate residual profit exceed an agreed amount, we will come to the final stage of determination of amount A, that is the nexus test in each market and according to the OECD proposal, this will not be only on the basis of revenue.

After all this rigmarole, we will finally get the amount A which will then be distributed to various market jurisdictions using an allocation key which is also to be worked out.

Then there will be amounts B and C out of which amount B is a baseline return to market jurisdictions in respect of marketing and distribution activities in the source states. This is not really an issue arising out of digitalised economy. Nevertheless, it is considered here apparently in the interest of reduction of litigation around the issue. If the source state claims that the marketing and distribution activities in its jurisdiction is above the baseline functions as defined, it will be entitled to an additional Amount C whose scope is still under discussion. But, the OECD is already concerned about possible overlap between the amounts, particularly when there is already a taxable presence of the group in the source State in the form of PE or subsidiary and steps will be taken that there is no double counting. The OECD documents do not specifically mention that amounts B and C will apply to businesses that are in-scope for the purpose of calculating amount A.

We may note that when the OECD Secretariat asked for inputs on its unified approach, rather unusually, the UN the tax committee decided to submit its views on the OECD's Pillar 1 proposal. The committee was rather trenchant in its criticism of the pillar 1 proposal. It specifically pointed out deficiencies in the following areas.

With respect to the scope, it questioned the unprincipled choice to target consumer facing business pointing out that the original objective was to deal with tax issues related to digital companies without physical presence.

The Committee pointed out that the economic threshold should be low enough to include a substantial number of MNEs; whereas the revenue threshold of €750 million will unnecessarily restrict the possibility to tax companies that deliver a substantial amount of digital services to developing countries and suggested that lower thresholds be applicable to "regional" MNEs that may be more important for a particular group of countries and also questioned the justification for the carve out for financial services industry.

With respect to the new nexus and country level revenue thresholds, the Committee cautioned against too-high thresholds, as that would prevent developing countries from taxing substantial profits attributable to their markets while supporting the proposal that thresholds should be country specific, taking into account the respective size of the economy.

With respect to the calculation of Amount A, the Committee mentioned that the whole operation would only justifiable if Amount A would attribute a fair portion of the non-routine profits to market jurisdictions. The committee also suggested not to exclude the possibility of applying a fractional apportionment also to routine profits.

In so far as amount B is concerned, the committee while welcoming a baseline return for marketing and distribution activities suggested that it should not be an elective safe harbour for the companies. For amount C, the committee again reiterated the reluctance of developing countries to accept mandatory binding arbitration. The Committee also pointed out the complexity of the proposal and difficulties that will be faced by the developing countries in understanding and implementing the OECD proposal and instead suggested the unified approach be remodelled into a simpler approach through the use of a withholding tax.

After the submission of the comments, Mr. Rajat Bansal submitted his own proposal for the consideration of the full committee. He also criticized the unified approach of the OECD on more or less the same grounds. In particular, he emphasized that all profits being essentially the result of global activities of a firm, there is no sound basis for allocating only the non-routine profits to market jurisdictions.

According to him, even the process of dispute resolution itself on which much emphasis has been given by the OECD will be problematic in as much as "(d)etermination of Amount A through a unitary approach is confined to the residual profits only while leaving transfer pricing methods intact for determining other taxable profits of the same MNES in various jurisdictions through separate entity approach. Amount A is to sit above the ALP driven profits. This will on one hand require the existing methods of determination and dispute resolution to continue and on other hand introduce a completely new method of determination as well as dispute resolution."

As regards the arbitrary cut off of 750 million Euros of group turnover for the applicability of the amount A, he mentions that if an MNE group has say 700 million Euro global revenue with revenue from one market country itself being 600 million euro, Amount A will not apply. On other hand, an MNE group with 800 million euro global revenue with revenues between 50-300 million euros in various market jurisdictions may be subject to Amount A, depending on local revenue crossing local threshold. Thus two different MNE group entities having identical local sales revenue in a market jurisdiction may be treated differently just because of difference in global revenue of the two MNE groups.

Mr Bansal added that elimination of double taxation in respect of Amount A will not be straightforward and it will not be possible to use corresponding adjustment approach of Article 9(2). This is due to Amount A being not premised on there being identifiable transactions between particular group entities. It will be necessary to determine which jurisdiction will have an obligation to eliminate any resulting double taxation and if there is more than one jurisdiction, the quantum of relief to be provided by each. To further complicate matters, Amount A will affect multiple jurisdictions that may not have existing bilateral treaties between them.

According to the OECD proposal, the exclusive filing will be in the ultimate parent jurisdictions but only for Amount A, and in view of unitary determination of consolidated profits of MNE Group as a whole, any dispute between two jurisdictions over Amount A will likely affect the taxation of Amount A in multiple jurisdictions and the existing bilateral mechanisms for dispute resolution will not be workable.

Mr Bansal proposed to restrict the in scope activities to automated digital services only in respect of revenue derived directly from the market jurisdictions, not through a subsidiary or a permanent establishment and the nexus for the taxing right should be deemed in a market jurisdiction only on the basis of local revenue derived, and no other thresholds be kept .

The taxable profits could then be determined by each jurisdiction by applying global profit rate of in-scope activities of MNE Group on the local sales revenue and attributing a percentage of the same to market jurisdiction through a fractional apportionment method.

According to him, the proposal could be implemented through insertion of a new Article may be inserted in the UN Model Convention and could be implemented in the same manner as the MLI. Such a system will give flexibility to countries to opt for the new system voluntarily and the new taxing right will operate only in those countries where such taxation is permitted by the domestic law .

Subsequently, the UN tax committee decided to form an ad-hoc drafting committee comprising exclusively members from developing countries to come up with a concrete proposal for the new Article as outlined by Mr Bansal. Mr Rajat Bansal was also a part of that committee. The others included the following: Carlos Protto (Argentina), Eric Mensah (Ghana), Tunda Fowler (Nigeria), Jorge Rachid (Brazil), Jose Troya (Ecuador), Abdould Moussa (Djibouti), Elfrieda Taamba (Liberia), Margaret Moonga (Zambia), Patricia Mongknonavaonit, (Thailand), Georg Obel (Kenya), D Minh( Vietnam), Marlene Parker (Jamaica). The members were acting in their individual capacity. On 6th August, the said committee submitted its report.

The drafting committee has suggested the adoption of a new Article 12B in the UN Model. We may recall that in 2017, the UN Model incorporated a new fees for technical services Article 12A. The structure of the new Article proposed by the sub-committee is more or less on the same lines as that of Article 12A.

The proposed new Article 12B will allow the source State to tax income from automated digital services paid to a resident of the other Contracting State on a gross basis at a rate to be negotiated bilaterally. Under this Article, a Contracting State is entitled to tax payments for automated digital services if the income is paid by a resident of that State or by a non-resident with a permanent establishment or fixed base in that State and the payments are borne by the permanent establishment or fixed base. Income from automated digital services are defined to mean payments for services provided on the internet or an electronic network requiring minimal human involvement from the service provider. Of course, to avail the lower rate of tax as negotiated, the recipient must be the beneficial owner of the income from such automated digital services.

Where the proposed provision differs from Article 12A or similar provisions is that it also gives an option to the companies to be taxed on a net basis when 30% of the net profit would be apportioned to the source state. The exact provision proposed in this regard is as follows:

"Notwithstanding the provisions of paragraph 2, the beneficial owner of the income from automated digital services referred to in that paragraph may require the Contracting State where the income from automated digital services arises, to subject its qualified profits from automated digital services for the fiscal year concerned to taxation at the tax rate provided for in the domestic laws of that State. For the purpose of this paragraph, the qualified profits shall be 30 percent of the amount resulting from applying the beneficial owner's profitability ratio or the profitability ratio of its automated digital business segment, if available, to the gross annual revenue from automated digital services derived from the Contracting State where such income arises . Where the beneficial owner belongs to a multinational group, the profitability ratio to be applied shall be that of the group or, if available, of the business segment of the group relating to income covered by this Article."

Of course, if the services are provided through a permanent establishment, then also the net basis of taxation will apply.

As per the proposed commentary, the following services are considered to be automated digital services: Online advertising services; Online intermediation platform services; Social media services; Digital content services; Cloud computing services; Sale or other alienation of user data; Standardised online teaching services. Since the definition is exhaustive, the Commentary also proposes that the following items will not be considered as income from automated digital services: Customised services provided by professionals; Customised online teaching services; Services providing access to the Internet or to an electronic network; Online sale of goods and services other than automated digital services ; Broadcasted services including simultaneous internet transmission; Composite digital services embedded within a physical good irrespective of network connectivity ("internet of things").

While some of the exclusions may be justified since there are specific provisions for taxation as in the case of customised professional services, there is no apparent reason for excluding some of the other types as proposed. Besides, we need to remember that the whole issue is that in the digitalised environment, the existing PE or even of fixed base concept is not adequate and status quo has to be disturbed.

As compared to the OECD unified approach under Pillar 1, the UN proposal is much neater and easier of administration particularly for the developing countries that lack much sophistication in tax administration. The million dollar question, however, will be whether the developed countries, particularly the USA will adopt it. But even if it does not adopt the provision, other states can go ahead and incorporate the same. It is at the insistence of the USA that consumer facing businesses were included in the scope of the OECD unified approach and there is logic in applying the same criteria to such businesses. One does not find any apparent reason for excluding such businesses. If the scope of the services is expanded, the proposal will in fact result in more certainty and ease of administration and also result in some gain of revenue to the developing countries in the digitalised environment.

The proposal will now be placed before the plenary session of the committee of tax experts scheduled in October, 2020 and in case, it is approved even with some modifications, it will make the OECD proposal more difficult to achieve consensus.