4. Carry-forwards and carve-out

286. This Chapter sets out two adjustments that may be made to an MNE Group’s top-up tax calculation for a particular jurisdiction.

(a) The first adjustment described in Section 4.2 allows an MNE to carry-forward losses incurred or excess taxes paid in prior periods into a subsequent period in order to smooth-out any potential volatility arising from the mix of taxes imposed under local law or resulting from timing differences. This adjustment is intended to ensure that Pillar Two does not result in the imposition of additional tax where the low ETR in a jurisdiction in a particular period is simply a result of the timing of the imposition of covered taxes on items of GloBE income or differences in the timing of the recognition of income under financial accounting and local tax law.

(b) The second adjustment described in Section 4.3 is a formulaic substance-based carve-out which is intended to exclude a fixed return for substantive activities within a jurisdiction from the scope of the GloBE rules. Excluding a fixed return from substantive activities focuses the GloBE rules on “excess income”, such as intangible-related income, which is most susceptible to BEPS risks.

287. Section 4.4 describes the methodology to be used in calculating a jurisdictional ETR and top-up tax in light of the adjustments for timing differences and the application of the formulaic substance based carve-out.

288. Temporary differences are differences in the timing of the recognition of income and expense under financial accounting and tax accounting rules. They are not differences in the types of income or expense allowed in the calculation of net income. Instead, they are differences in the proper time for including those items in the calculation of net income.

289. Temporary differences can be the sole cause of a low ETR at the beginning of the temporary difference and a high ETR upon reversal, and vice versa. They have an effect on the periodic measurement of the ETR but do not affect the average ETR over the life of the entity.

290. Temporary differences should not give rise to permanent tax liability under the GloBE rules. Therefore, a mechanism to address the effects of temporary differences on the ETR and the GloBE tax liability are necessary.

291. While the discussion focuses on the application of the income inclusion rule, similar principles can be applied in the context of the undertaxed payments rule. Generally, the loss carry-forwards and local tax carry-forwards will be taken into account to determine the jurisdictional ETR under the undertaxed payments rule as well. However, a tax credit similar to the IIR tax credit will not be available to recoup taxes paid in prior years due to the undertaxed payments rule.

292. Temporary differences can be addressed under the GloBE rules using a carry-forward approach or a deferred tax accounting approach. From a time value of money perspective, deferred tax accounting generally is more favourable to taxpayers because it leaves in place the benefits of immediate expensing of assets, accelerated depreciation, and other tax deferral mechanisms that are commonly adopted by jurisdictions to encourage capital investment and for other reasons. The basic carry-forward approach with some modifications can preserve the most significant of those benefits but does not wholly align with the tax deferral benefits allowed in every local jurisdiction.

293. The principal policy risk of deferred tax accounting, however, is that it relies on a taxpayer’s estimate of future liability for tax in a subsidiary jurisdiction to determine its current liability under the GloBE rules. The carry-forward approach, in contrast, relies on actual tax liabilities existing at the time tax liability under the GloBE rules is determined.

294. The carry-forward and deferred tax accounting approaches both impose some compliance and administration burdens. On the surface, the deferred tax accounting approach appears to be simpler because taxpayers use deferred tax accounting for financial purposes already. However, differences between the policies of the GloBE rules and financial reporting will likely mean that modifications would need to be made to certain deferred tax accounting outcomes in order to adapt deferred tax accounting to the GloBE rules. These modifications would add a significant amount of additional complexity and may lead to uncertainty. Although maintaining memorandum accounts in respect of carry-forwards is somewhat burdensome, it is a familiar exercise for both taxpayers and tax administrations. Accordingly, while there appear to be some advantages with a deferred tax accounting approach the members of the Inclusive Framework do not consider that such an approach would serve as an appropriate mechanism to address timing differences. They do consider, however, that financial information on deferred tax accounting could play a useful role in the development of simplification options described below in Section 5.

295. The basic operation of the carry-forward approach relies on two carry-forwards and a credit to address the various effects of temporary differences on the GloBE tax base and ETR computation. Both carry-forwards would be tracked using memorandum accounts maintained on a jurisdictional basis. The credit is tracked using memorandum accounts for each Parent of the MNE Group that applies an IIR.

296. Under the carry-forward approach, the covered tax expense included in the numerator of the ETR computation for a particular tax year includes only the amount of taxes reported as due and payable in the tax returns of Constituent Entities filed with respect to their income for that tax year. Thus, the covered taxes in the numerator for a tax year include the taxes paid in respect of that year both during the year and with the tax return for that year to the extent those taxes are payable with respect to the income for that tax year. However, any covered taxes reported in a tax return that is not paid, for example due to an administrative practice of allowing loss carry-forwards in the administrative assessment of tax rather than as part of the self-assessment in the tax return, is not included in the numerator of the ETR fraction. The covered taxes in the numerator do not include any amount of tax expense that is accrued for financial accounting purposes and that is not reported in the tax returns filed with respect to income of the relevant tax year, for example, uncertain tax positions, contested tax liabilities, and deferred tax liabilities. Finally, any amount of tax that is reported in the tax returns with respect to the income for a year that is not paid within a certain period (for example 2 years) should be treated as a reduction in covered taxes in the subsequent year.

297. Limiting the covered taxes expense in the numerator of the ETR fraction to taxes reported in the tax returns for the relevant year could, in some situations, result in an MNE Group being subject to both the IIR or UTPR and the STTR in respect of a single transaction. For example, a Constituent Entity could be resident in a jurisdiction that is generally low-taxed and thus likely subject to a Withholding Tax (WHT) under the STTR on payments received from another Constituent Entity. A timing difference between the accrual of the income and the payment, however, could result in an ETR below the minimum rate in the year of accrual, which would result in a top-up tax liability in the payer’s jurisdiction under the IIR or UTPR. Assuming the income is actually received the year after, the Constituent Entity would be subject to the STTR as well when it received the income because it is resident in a low-taxed jurisdiction. However, if the income accrual and receipt had occurred in the same tax year, the WHT imposed under the STTR would have been taken into account to compute the ETR of the jurisdiction, which could have raised the ETR above the minimum tax rate and prevented application of the IIR or UTPR. The IIR tax credit can adequately address this situation. However, in the absence of a similar mechanism available under the UTPR, frequent instances of double taxation could occur on transactions subject to a withholding tax.

298. The GloBE rules ameliorate this mismatch between the income accrual and tax payment by allowing accrual of withholding taxes for income that are expected to be paid shortly after the year in which the related income accrues. Specifically, any withholding tax accrued by a Constituent Entity for financial accounting purposes on an item of income other than a distribution from another Constituent Entity that will be paid within 12 months following the end of the taxable year in which the Constituent Entity accrues the related item of income for financial accounting purposes may be included in the covered taxes expense for the year in which the income is accrued. Under this rule, a Constituent Entity that knows it will be liable for withholding tax under an STTR in the year following the accrual of the relevant income will be able to include the accrued WHT levied under the STTR in its covered tax expense for purposes of the jurisdictional ETR computation, with the likely result that the IIR/UTPR will not apply. The rule not limited to withholding taxes imposed under an STTR rule. It does not apply, however, to withholding taxes that are accrued on distributions from other Constituent Entities.

299. In some jurisdictions, loss carry-forwards are limited to a number of years. Imposing a time restriction on the use of carry-forwards can reduce the compliance and administration burdens associated with maintaining the relevant memorandum accounts. Other jurisdictions allow loss and other tax attribute carry-forwards to be carried forward for an unlimited time period.

300. The GloBE loss carry-forward is crafted so that it is effectively unlimited in duration. The GloBE rules apply to a wide range of industries and business sectors. These industries have business cycles of different lengths, with some industries, such as the mining industry, experiencing very long business cycles. Some MNE Groups or Constituent Entities of an MNE Group that are profitable over the business cycle may be profitable in some years and not profitable in other years during the cycle. Unlimited carry forward of losses ensures that MNE Groups will not be subject to tax under the GloBE rules on more than their economic income due to an expired loss carry-forward.

301. The ability to use excess local taxes to create an IIR tax credit in respect of IIR tax paid in prior years and the local tax carry-forward are limited in duration, however. This limitation does two things. First, it effectively treats a long-term deferral as a permanent difference for purposes of the GloBE rules. In other words, if a carry-forward attributable to a timing difference expires before the timing difference resolves itself, the timing difference will produce the same result under the GloBE rules as if it were a permanent difference. Second, it limits the period for which taxes that are imposed at a rate above the minimum rate can be used to shield income taxed below the minimum rate from GloBE tax liability. However, the IIR tax credit is not limited in duration, which eliminates, or significantly reduces, the possibility that IIR tax paid due to a timing difference will result in taxation under the GloBE rules.

302. The loss carry-forward is needed to prevent taxation in excess of economic income under the GloBE rules. Financial accounting does not have a loss carry-forward concept. The financial accounting income for each year is determined without regard to any losses incurred in prior years. The annual GloBE tax base computation starts with financial accounts and thus may also include income in excess of the economic income over a period. The loss carry-forward is allowed as a deduction to prevent taxation in excess of economic income.

303. The loss carry-forward under the GloBE rules is relatively straight-forward and will be familiar to many taxpayers and tax administrations. A loss carry-forward is created for a jurisdiction in any year in which the expenses taken into account in computing the jurisdictional GloBE tax base, including prior losses carried forward from previous years, exceed the amount of income taken into account in computing the jurisdictional GloBE tax base. Losses can be carried forward indefinitely under the GloBE rules. A loss arising in a jurisdiction may be carried back to the same extent that it is carried back under the rules of that tax jurisdiction. Consistent with the jurisdictional blending model, loss carry-forwards arising in a jurisdiction can only be used as a deduction in the computation of the adjusted GloBE tax base in that same jurisdiction. A loss carry-forward is only used to reduce the GloBE tax base if the jurisdiction has an ETR below the minimum tax rate determined without regard to the loss carry-forward.

304. The loss carry-forward under the GloBE rules will not generally be expected to align with rules governing loss carry-forwards in the local jurisdiction. Rather, the GloBE loss carry-forward is intended to prevent taxation in excess of economic income and recognises that an MNE should not be subject to tax under the GloBE rules on the mere recovery of prior period losses. The loss carry-forward allowed under GloBE may be inconsistent with the rules governing loss carry-forwards in the jurisdiction in which the loss arises. For example, the tax rules in the local jurisdiction may place a limitation on the amount of loss that may be carried forward or the time period for which it can be carried forward. Alternatively, the local jurisdiction may allow losses to be carried back and used against tax liabilities arising in prior years, which may generate a tax refund. Under those circumstances, the loss carry-forward for GloBE purposes may be a different amount than the loss carry-forward for local tax purposes. In addition, losses in one subsidiary may be used to offset the income of another subsidiary in the same jurisdiction in the computation of the jurisdictional GloBE tax base, while a similar type of offset may not be allowed under local tax rules. In that case, one subsidiary could have a loss carry-forward for local tax purposes even though the loss was fully deducted in computing the GloBE tax base for the jurisdiction. No adjustment is necessary to address these differences because the GloBE tax base is not trying to mirror the local tax base and any GloBE tax consequences resulting from the differences will be addressed by the local tax carry-forward and the IIR tax credit described below.

305. The carry-forward rules are designed to smooth the ETR of the jurisdiction over a period of time, irrespective of whether fluctuations in the ETR arise from temporary or permanent differences. The GloBE rules apply when the ETR in a jurisdiction is below the minimum tax rate. Temporary or permanent differences between the local tax base and the GloBE tax base may cause the ETR in a jurisdiction to be above or below the minimum tax rate in a particular year. Over time, the temporary differences will reverse and various permanent differences may have opposite effects on the ETR. The rules effectively smooth the ETR for the jurisdiction over a period of time by allowing a shareholder to recoup GloBE taxes paid previously or avoid paying GloBE taxes in the future whenever the ETR in the jurisdiction exceeds the minimum tax rate. Specifically, whenever the tax in a jurisdiction exceeds the minimum tax rate, a shareholder that previously has paid IIR tax in respect of the jurisdiction is allowed to create an IIR tax credit to the extent the previously paid IIR tax has not already been treated as an IIR tax credit. An IIR tax credit can be used to satisfy a current or future IIR tax liability with respect to any jurisdiction. If the shareholder has not previously paid IIR tax in respect of a jurisdiction, tax paid in that jurisdiction in excess of the minimum tax rate on the aggregate GloBE tax base is treated as a local tax carry-forward that the shareholder can use in the computation of its ETR for that jurisdiction in a future taxable year, which may increase the ETR of the jurisdiction up to the minimum rate in that year.

306. There will be many cases in which the tax rate in a jurisdiction exceeds the minimum rate by an amount sufficient to avoid GloBE tax liability even after taking into account a permanent difference. By incorporating mechanisms that take into account the effects of temporary and permanent differences on the computation of income and tax liabilities over a period of years, the rules neutralise the consequences stemming from application of the annual accounting concept under the GloBE rules. Although the rule is primarily aimed at temporary differences that reverse over time, it also ameliorates the effect of permanent differences because their effect on GloBE tax liability may also be due to their timing. For example, equal and offsetting, but otherwise unrelated, permanent differences would have no effect on the GloBE tax liability if they arise in the same tax year but would produce a GloBE tax liability if they arose in separate tax years. While this is not a temporary difference, it does have a timing aspect and is therefore appropriately addressed by a carry-forward of excess local taxes. In addition, rules designed to separate the amount of excess local taxes attributable exclusively to temporary differences would be very complex and unduly complicate the GloBE rules.

307. Both the IIR tax credit and the local tax carry-forward hinge on excess taxes paid in respect of a jurisdiction. The rule defines excess taxes as the amount of covered taxes reported as due and payable in the tax returns of the Constituent Entities in the jurisdiction filed with respect to the income of a particular tax year, in excess of the minimum tax rate on the aggregate GloBE tax base for that jurisdiction for that year. If the GloBE tax base computation for a jurisdiction results in zero income or a loss for a year, any local tax paid in respect of the year would be excess taxes for that year and would be included in a local tax carry-forward. This situation could arise where, for example, the local tax base denied certain deductions that were recognised under the GloBE tax base. The definition of excess taxes is intended to ensure that the ETR is computed based on the taxes actually paid in respect of the relevant year. It does not include income taxes that are accrued, for example based on the likely disallowance of an uncertain tax position, but that are not reflected as due in the tax return filed for the year.

308. Under the IIR tax credit rule, in any year in which there are excess taxes in respect of a jurisdiction, the shareholder first looks back to see if it previously has paid IIR tax in respect of that jurisdiction.1 If the shareholder has paid IIR tax in respect of the preceding taxable years (the lookback period), an IIR tax credit is created. The IIR tax credit created for a year is equal to the lesser of the excess taxes for the year and the IIR tax paid during the lookback period that has not already given rise to an IIR tax credit. The IIR tax credit is illustrated in Annex, Example 4.2.1B. In most cases it is expected that the IIR tax credit will be available to reduce IIR tax liabilities arising in the year the credit is created or any year thereafter.2 Although an IIR tax credit results from a payment of IIR tax and a subsequent payment of excess taxes in the same jurisdiction, IIR tax credits can be used to reduce an IIR tax liability arising with respect to any jurisdiction. The use of the IIR tax credit in respect of IIR tax liabilities arising in respect of other jurisdictions is illustrated in Annex, Example 4.2.1C.

309. It is possible, given the design of the rules, that a parent with an accrued entitlement to an IIR tax credit will not have, and is not expected to have in the foreseeable future, an IIR tax liability to offset that IIR tax credit against. The mechanics of the IIR and the IIR tax credit are novel and their operation will depend on the structure and operations of the MNE and the point in the ownership chain where the IIR is applied. Rather than creating incentives for an MNE Group to restructure its operations simply in order to take advantage of an unused IIR tax credit, tax administrations could contemplate introducing a mechanism that allowed the MNE Group to offset such credits against other domestic tax liabilities of the Parent or another Constituent Entity of the MNE Group that is resident in the Parent’s jurisdiction due to difficulties in recovering the IIR tax credit that could persist over time, for example, where that credit had not been used to offset an IIR tax liability within a reasonable period of years following the period in which the credit arose.3

310. To effectively address timing differences, the local tax carry-forward period and the lookback period for creating an IIR tax credit need to be long enough for the timing difference that caused the original IIR tax or the carry-forward to reverse. The period in which a timing difference will reverse varies based on the timing rules of each tax jurisdiction. Nonetheless, the GloBE rules should establish a period that will adequately cover the period required for most timing differences to reverse without creating significant compliance and administrative burdens of tracking and verifying excess taxes claimed in respect of years long past. The GloBE rules will also include specific adjustments to minimize the most significant of timing differences arising in most businesses, i.e. timing differences attributable to immediate expensing and accelerated depreciation. In light of the overall design of the rules and the economic cycle of most businesses, a period of seven years may, depending on other design features of the rules, be a reasonable period for both the lookback period and carry-forward period. Further consideration will be given to whether extended time limits, including for the transitional carry-forward rules, are appropriate in certain industries with long economic cycles.

311. Under the local tax carry-forward rule, the excess taxes for a jurisdiction that do not create an IIR tax credit create a local tax carry-forward. A local tax carry-forward may be included in the ETR computation in any subsequent year in which the local tax paid by the Constituent Entities in the same jurisdiction falls below the minimum tax rate on their aggregate GloBE tax base. The local tax carry-forwards can only be used to compute the ETR for the jurisdiction in which they arose and are used in chronological order. Local tax carry-forwards are reduced by the amount used to increase the local tax to the minimum tax rate in any year. The local tax carry-forward is illustrated in Annex, Example 4.2.1A.

312. If a Parent is potentially subject to IIR tax liability with respect to multiple jurisdictions in the same year, it may allocate its IIR tax credits, if any, among those jurisdictions as it chooses. However, if there are multiple Parent’s in the same MNE Group applying an IIR, a Parent may not use another Parent’s IIR tax credit.

313. The IIR tax credit may be used against a Parent’s IIR tax liability arising in respect of any jurisdiction in the period the credit arises or a subsequent period.4 At first blush, the ability to credit the IIR tax in one jurisdiction against IIR tax liabilities arising in respect of another jurisdiction may seem tantamount to worldwide blending. It is not the same, however. An IIR tax credit only arises when IIR tax is paid in respect of a jurisdiction, and subsequently, tax is paid in that same jurisdiction in excess of the minimum tax rate. Conceptually, a shareholder becomes eligible for a credit of IIR tax paid in respect of a jurisdiction when it can demonstrate that the tax was paid in respect of a temporary difference in the jurisdiction that reversed after the tax payment. The carry-forward approach uses the payment of tax in excess of the minimum tax rate as a proxy for the reversal of a temporary difference.5

314. If the timing of the income under the local tax base matched the timing of income under the GloBE tax base, there would have been no IIR tax liability in the first instance and no need for a subsequent IIR tax credit. From the taxpayer’s perspective, the credit mechanism is functionally equivalent to a refund of IIR tax previously paid. However, instead of getting a cash refund from the tax administration and separately paying another IIR tax liability, the shareholder uses the credit to pay that tax liability. The fact that the credit can be used to satisfy an IIR tax liability arising in respect of a different jurisdiction does not change the analysis. Eligibility for the IIR tax credit is always rooted in a temporary difference arising and reversing with respect to the same jurisdiction or the interaction of two offsetting permanent differences in the same jurisdiction. In contrast, allowing excess taxes paid in respect of the income in one jurisdiction to create an IIR tax credit for IIR tax paid in respect of another jurisdiction would be tantamount to global blending.

315. There are a number of events that could trigger the application of the GloBE rules to an MNE Group for the first time. Where an MNE Group already has revenues in excess of the revenue threshold, it will become subject to the GloBE rules once they have been introduced into the domestic law of a jurisdiction in which the MNE Group operates. Smaller MNE Groups, however, will become subject to the GloBE rules for the first time if they grow their revenues above the threshold, either organically or as a result of a merger or acquisition.

316. At the point an MNE Group becomes subject to the GloBE rules, it will be required, under a jurisdictional blending approach, to compute the ETR on its income in each jurisdiction where it operates and compare it to the agreed minimum tax rate. Failure to take appropriate account of operating losses that the MNE Group has suffered in the period or periods immediately prior to becoming subject to the GloBE rules could, however, result in a distorted picture of the MNE Group’s tax position in that jurisdiction and may subject the MNE Group to taxation in excess of its economic profit. For example, a Constituent Entity may have incurred operating losses in the years immediately prior to the MNE Group becoming subject to the GloBE rules. Frequently, the operating losses of the Constituent Entity will also be recognised for local tax purposes and these losses may be eligible to be carried forward and be available to reduce taxable income arising in a future period in the same jurisdiction. Ignoring the effect of these prior period losses could result in an immediate GloBE tax on profits arising in subsequent periods despite the fact that, the local tax jurisdiction is otherwise a high-tax jurisdiction and that the income subject to charge under the GloBE rules, represents, from the MNE Group’s perspective, a recovery of recent losses. Failure to take appropriate account of pre-regime losses could therefore result in the MNE Group being overtaxed, by converting what was essentially a timing difference into a permanent difference based on the mere fact that the MNE Group was brought within the scope of the GloBE rules after those losses arose.

317. A similar transition-related issue in relation to timing differences that straddle the applicability date of the GloBE rules. Of particular concern are those timing difference that result in the acceleration of income and hence taxes paid prior to an MNE Group being subject to the GloBE rules, which then reverse after the MNE Group is subject to the GloBE rules. These situations may arise, for example, when local law taxes pre-payments of contractual fees upon receipt rather than over the term of the contract or prohibits deductions for estimates of future bad debts or warranty expenses (i.e. reserves for bad debts or warranty expenses). Absent a corrective rule, the result would be a lower GloBE ETR in the year(s) of reversal and thus potential GloBE top-up tax in those years, despite the fact that the local tax jurisdiction is otherwise a high-tax jurisdiction. Similarly, timing differences that defer tax on income arising before the GloBE rules apply would, absent a special rule, reduce the GloBE tax liability on GloBE income arising within the GloBE applicability period.

318. The Inclusive Framework on BEPS considers that transition rules are needed to minimize the GloBE tax liability effects of losses and timing differences that straddle the applicability date of the GloBE rules. The most accurate approach to addressing these transition issues would be to identify an applicable start date for the transitional period and require MNE Groups to compute an opening balance of its loss carry-forward and local tax carry-forward as if the GloBE rules had applied during the transitional period. However, applying the GloBE rules retroactively to prior years could be unduly complex and burdensome. A simplified method could be developed that could produce results that reasonably approximate this approach with less complexity and administrative burden. Further technical work will be necessary to develop a workable solution that provides for appropriate outcomes without imposing undue compliance or administrative burdens.

319. In addition to rules dealing with what happens when an MNE Group enters the GloBE regime for the first time, further rules are required to address the situation where Constituent Entities join or leave an MNE Group that is already subject to the GloBE rules. These transactions may be taxable or non-taxable and may include, for example, acquisitions of the equity or assets of Constituent Entities, or acquisitions by one MNE Group of the equity of the Ultimate Parent Entity of another MNE Group, or a spin-off by one MNE Group of Constituent Entities that following the spin-off become another MNE Group. The ability to transfer the benefit of carry-forwards from one MNE Group to another can be expected to be more restricted under the GloBE rules than ordinary tax rules because the GloBE carry-forwards are determined and applied on a jurisdictional basis. Thus, under the GloBE rules the carry-forwards are strongly associated with the MNE Group as a whole rather than with a Constituent Entity. Nevertheless, there are cases where part of the consideration for the sale of a Constituent Entity can include the benefit of certain deferred tax assets such as carry-forward losses or the benefit of taxes paid in advance of income which are available to reduce tax on future income.

320. The ability to carry-over the tax attributes of a Constituent Entity on a sale or disposal is a question of local law design. Many jurisdictions have shareholder continuity rules that prevent a company from carrying forward the benefit of a credit, loss, or other relief where there is a change in control. However, these rules are not comprehensive. They may require a significant change in shareholding before they apply and they would not, for example, typically prevent a company carrying forward the benefit of tax pre-payment or the tax paid on accelerated income. The ability to carry over the target company’s local attributes in a sale transaction could, in the absence of a corresponding adjustment under the GloBE rules give rise to the risk of double, over-or under-taxation.

321. For example this situation could arise where a Constituent Entity such as a company, that is entitled to retain certain accrued tax benefits (such as carry-forward losses), is sold to an independent purchaser. If the target company derives income in a period subsequent to the transfer then that income will be sheltered by the carry-forward losses that are available under local law. Unless the GloBE rules recognise the transfer tax benefits that are permitted for local law purposes then:

  1. a. The seller could effectively double dip on the value of the transferred tax losses through increased consideration for the sale of the target shares and by using the losses that it is treated as retaining under the GloBE rules to shelter other low-tax income

  2. b. The buyer could suffer economic double taxation in the form of an increased purchase price for the target company shares and a tax charge under the GloBE.

322. The Inclusive Framework on BEPS considers that an adjustment to carry-forwards is appropriate when a Constituent Entity is sold outside the MNE Group. One approach under consideration would require the buyer and the seller to adjust the amount of any carry forward losses or excess taxes by the amount of the related deferred tax asset retained by the target company (or inherited by a successor entity, such as in the case of a merger) immediately following the sale. The deferred tax assets retained by the target company may represent a reasonable approximation of the amount of the carry-forwards that left the MNE Group along with the target company. The target company (or successor entity) may be expected to retain (or inherit) a deferred tax asset in transactions involving the sale and acquisition of equity of the company or in non-taxable acquisitions of the assets of the company (such as through a merger). The buyer and seller would be further required to identify those deferred tax assets that are in fact transferred to the buyer under local law and to make a corresponding adjustment to the amount of the carry-forward to recognise the effect of the transfer. Further technical work will be undertaken with regard to adjustments to carry-forwards when a Constituent Entity leaves the MNE Group. The outputs from this work will be incorporated into the model rules to be developed in accordance with Section 10.5.1.

323. The ETR for a jurisdiction under the carry-forward approach is determined by dividing the covered taxes in the jurisdiction with respect to that year by the GloBE tax base for the jurisdiction. However, an MNE Group’s liability for covered taxes may increase or decrease after the tax return for the year is filed due to various reasons. This could include a change in the amount of income recognised for local tax purpose due to an examination of the returns by the local tax authority, a review of the returns by the entity’s management or tax advisers, or a loss in a subsequent period that is allowed to be carried back under applicable tax law, or a refund of tax to the taxpaying entity or its shareholder upon distribution of dividends. Increases would normally result in additional tax paid and decreases would normally result in a refund of tax (either in cash to, or as a reduction of another tax liability of, the taxpaying entity or its shareholders) after the IIR tax return for the relevant year was filed. The tax underpayments or overpayments may have impacted the shareholder’s IIR tax liability and the amount of one of its carry-forwards for the corresponding tax year or its IIR tax credits. In other words, if the final tax liability had been correctly determined when the tax return was filed, the shareholder may have paid more or less IIR tax and may have established a larger or smaller local tax carry-forward or IIR tax credits.

324. The GloBE rules incorporate a carry-forward adjustment mechanism to address the effect of post-filing tax increases and decreases in subsidiary jurisdictions on the IIR tax liability and carry-forwards. This mechanism is much simpler than an alternative approach of requiring an amendment of the return to which the adjustment relates (and in some cases intervening year tax returns (i.e. tax years between the date of the adjustment and the date to which the adjustment relates)).

325. Under the carry-forward adjustment mechanism, the effects of a post-filing tax increase or decrease would be taken into account prospectively by adjusting the balances of the relevant carry-forwards when the increase or decrease is determined with finality. For this purpose, “determined with finality” means that the period for disputing the tax adjustment has expired either due to the passage of time or due to an administrative or judicial determination. For example, a payment of tax based on an administrative assessment to avoid additional interest expense while a claim is being litigated is not a final determination, but the judicial determination is a final determination when the time for appealing the decision expires without an appeal. Although the post-filing tax increase or decrease is primarily treated as an adjustment to the carry-forwards that arose in prior periods, the effect of those adjustments will be on the MNE’s current and future tax liability under the GloBE rules.

326. Under the carry-forward adjustment mechanism, a tax decrease or refund in a jurisdiction would be treated as reduction in the amount of the relevant local tax carry-forward. Because local tax carry-forwards are limited in duration, they will need to be tracked based on the year in which they were created. Sometimes, a local tax carry-forward created for a year will have been used to reduce IIR tax liability in a subsequent year before it is discovered that the carry-forward was overstated. Rather than sifting through and unwinding all of the effects of the over-stated carry-forward in subsequent years, a local tax decrease should be treated as:

  1. (a) a decrease in the outstanding balance of the local tax carry-forward for the year to which the refund relates, if any;

  2. (b) a decrease in the outstanding balance of local tax carry-forwards established for subsequent years, to the extent thereof;

  3. (c) and then, a reduction to the tax expense for the year in which the decrease becomes final, to the extent thereof.

327. If the tax decrease exceeds the local tax carry-forwards described in (a) and (b) and the tax expense for the year, the excess should be treated as a current liability for IIR tax because it represents IIR tax that should have been due in respect of a prior year. If a tax decrease or refund does not relate to a specific taxable year, for example because it arises upon distribution of a dividend, the decrease or refund should be treated as a reduction of the current year tax liability to the extent thereof, and then a reduction of excess taxes paid in prior years. Any amount in excess of taxes paid in prior years should be treated as a current liability for IIR tax. Application of the rules in the case of a post-filing reduction in local tax liability is illustrated in Annex, Example 4.2.2A.

328. Under the carry-forward adjustment mechanism, a tax increase in a jurisdiction creates IIR tax credits to the extent of IIR tax paid in the look-back period that had not already given rise to an IIR tax credit. The excess, if any, is treated as a local tax carry-forward in respect of the year to which it relates or the last year in which IIR tax was paid. The period for using such local tax carry-forwards runs from the year in respect of which the carry-forward is established rather than the year in which the tax increase becomes final. Application of the rules in the case of a post-filing increase in local tax liability is illustrated in Annex, Example 4.2.2B.

329. Generally, the adjustments described above are only required with respect to post-filing increases or decreases in local tax liabilities with respect to tax years in which the MNE Group was subject to the GloBE rules or a tax year included in the determination of any carry-forward established in connection with a transition rule.

330. In addition to errors in the computation of annual tax liability, an MNE Group may occasionally make an error in the computation of its profit (or loss) before tax for financial accounting purposes that carries over into the GloBE tax base. For financial accounting purposes, the MNE Group generally must correct material “prior period errors” retrospectively by restating the comparative amounts for the prior period(s) presented in which the error occurred. “Prior period errors” are omissions from, and misstatements in, the consolidated financial statements for one or more prior periods. A prior period error that requires a restatement for prior periods in the MNE Group’s consolidated financial statements should be corrected in the GloBE tax base by including the cumulative difference in income or expense in the GloBE tax base computation for each jurisdiction in the year in which the error is corrected for financial accounting purposes.

331. Furthermore, an MNE Group may change accounting method or principles used in the preparation of its financial statements. The change could be to the treatment of a single item or a category of transactions and could be voluntary, for example, due to an elective treatment of a particular type of transaction, or required, for example, due to a newly promulgated accounting standard. The MNE Group may also change from one acceptable accounting standard to another, such as from U.S. GAAP to IFRS. These changes may result in a cumulative change to the equity of the MNE Group based on the application of the different standards. In such cases, the cumulative change to equity attributable to each jurisdiction should be included in the GloBE tax base computation for the jurisdiction in the year in which the accounting method or principle is adopted for financial accounting purposes.

332. The policy rationale behind a formulaic carve-out based on expenditures for payroll and tangible assets is to exclude a fixed return for substantive activities within a jurisdiction from the scope of the GloBE rules. The use of payroll and tangible assets as indicators of substantive activities is justified because these factors are generally expected to be less mobile and less likely to lead to tax induced distortions. Conceptually, excluding a fixed return from substantive activities focuses GloBE on “excess income”, such as intangible-related income, which is most susceptible to BEPS risks. Furthermore, a carve-out based on expenditures for payroll and tangible assets should help to shield low-margin businesses from what would otherwise be disproportionately negative outcomes under the GloBE as a result of expenditure based tax credits and other forms of government subsidy based on expenditure, such as government grants.

333. The carve-out will only benefit those MNEs with operations in jurisdictions that are taxed at below the minimum rate. However, provided the amount of the carve-out is limited to a modest return (sometimes colloquially referred to as a “routine return”) on expenditures for payroll and tangible assets, then the MNE will not generally be able to use the carve-out to shelter other low-tax returns in a particular jurisdiction. An MNE can increase the amount of the carve-out by shifting more payroll and tangible assets into the jurisdiction, but, all other things being equal, increasing investment in these production factors would result in a corresponding real increase in the routine returns attributable to those factors and will not allow the carve-out to shelter excess returns or returns attributable to other factors such as intangible assets.

334. By acknowledging the contributions of both employees and tangible assets, a combined carve-out for payroll and tangible assets provides for a more level playing field by allowing a meaningful carve-out for MNEs with varying substance profiles, including labour-intensive and asset-intensive businesses. Whereas a carve-out based on a single factor, either payroll or tangible assets, would end up favouring one set of industries over another. Therefore, a combined carve-out provides greater neutrality between different industries.

335. Further consideration will be given, in light of the policy rationale behind the formulaic substance-based carve-out, to the effect of the carve-out on the calculation of the ETR and top-up taxes under the GloBE, particularly whether an MNE group that claims the benefit of the carve-out should be required to make a corresponding and proportional adjustment to the covered taxes. A decision on this tax adjustment will impact on the ETR and the top-up taxes payable under the GloBE and will be considered together with other questions such as the determination of the fixed percentage mark-up to be applied in a formulaic substance-based carve-out.

336. The Sections below set out more detail on the operation of the carve-out, starting with the payroll component and then turning to the tangible asset component.

337. The payroll component of the carve-out removes from the GloBE tax base a fixed return on activities performed in that jurisdiction calculated by reference to the taxpayer’s employment costs. Such a carve-out design recognises a Constituent Entity’s payroll expense as an appropriate proxy for substantive activities carried out by employees of the taxpayer in the relevant jurisdiction. In applying such a carve-out it is necessary to identify relevant employees (eligible employees), the situs of those employees as relevant for jurisdictional blending, and the relevant payroll expenses of those eligible employees (eligible payroll costs).

338. For the purposes of the payroll component of the carve-out, eligible employees includes all employees of the MNE, including part-time employees. Eligible employees also includes independent contractors participating in the ordinary operating activities of the MNE, which is both consistent with country-by-country (CbC) reporting and avoids what would otherwise be a difficult line-drawing exercise of distinguishing an employee from an independent contractor.9 Independent contractors include only natural persons and may include natural persons who are employed by a staffing or employment company but whose daily activities are performed under the direction and control of the Constituent Entity. Independent contractors do not include employees of a corporate contractor providing goods or services to the Constituent Entity.

339. The payroll component of the carve-out is computed on a jurisdictional basis focused on where the actual activity is performed. The CbCR rules, which are based on the residence jurisdiction of the Constituent Entity paying the salary, can be used for determining this in most cases. However, where the residence jurisdiction of the Constituent Entity paying the employee’s salary (i.e. the employer) differs from the jurisdiction where the employee’s activities or services are performed, the residence of the employee should be used as an indicator of the place of actual activity10, unless there is strong evidence that the actual activity is performed in another jurisdiction. The ordinary or normal residence of its employees should be information that an MNE can easily access and so this should not constitute a significant additional compliance burden. The approach for determining where eligible employees’ activities or services are performed should be applied consistently across the MNE Group and from year to year.

340. A broad approach for determining eligible payroll costs is used for the payroll component of the carve-out based on a general test of whether the expenditure of the employer gives rise to a direct and separate personal benefit to the employee. Eligible payroll costs include expenditures for salaries and wages as well as for other employee benefits or remuneration such as medical insurance, payments to a pension fund or other retirement benefits, bonuses and allowances payable to eligible employees, and stock-based compensation. Eligible payroll costs also includes payroll taxes (or other employee expense-related taxes such as fringe benefits taxes), as well as employer social security contributions.

341. Consistent with the broad approach for determining eligible payroll costs, the payroll component of the carve-out is based on the total amount of the payroll expenditures in the current year, rather than the amount treated as an expense in the income statement as per financial accounting rules. This also avoids the additional administrative burden of taxpayers sorting payroll costs based on whether it was currently expensed or capitalised for financial accounting purposes.11 It is also consistent with BEPS Action 5 approach to computing qualified R&D expenditures for the purposes of the modified nexus ratio.

342. The tangible asset component of the carve-out is equal to the sum of:

  1. (a) [x]% of the depreciation of property, plant and equipment;

  2. (b) [x]% of the deemed depreciation of land;

  3. (c) [x]% of the depletion of natural resources; and

  4. (d) [x]% of the depreciation of a lessee’s right-of-use tangible asset.

343. The tangible asset carve-out base includes the annual cost of using depreciable property, plant and equipment, land, natural resources, and a lessee’s right-of-use assets that are used in the production of income. Including a broad range of tangible assets in the carve-out base recognises that all such assets are indicative of substantive activities. Moreover, it helps to level the playing field across industries that use varying types of tangible assets in their business. Including leased tangible assets neutralises the difference between owning and leasing assets and recognises that the business decision to own or lease typically has no bearing on the intensity of substantive activities.

344. While the carve-out generally seeks to recognise a broad range of tangible assets, an MNE should not be allowed to generate a larger cave-out by purchasing investment property in a jurisdiction. This risk is particularly relevant as it relates to buildings and land, which are commonly held as investments. To neutralise this risk, buildings and land that are held to earn rental income or for capital appreciation (or both), not owner-occupied; not used in production or supply of goods and services, or for administration; and not held for sale in the ordinary course of business are excluded from the carve-out. This rule is not expected to materially increase complexity or compliance costs because many accounting standards already require that such assets be separately identified and accounted for. For example, in the case of IFRS, investment properties are separately accounted for under IAS 40 – Investment Property.

345. Similarly, an MNE should not be allowed to generate a larger carve-out via tangible assets whose carrying amount, i.e., cost, will be recovered principally through a sale transaction instead of through continuing use in the business. Since such assets are held for sale, not use, they are a poor proxy for substantive activities. Consequently, assets held for sale are excluded from the carve-out. In order to be considered held for sale, the asset must be available for immediate sale in its present condition subject only to terms that are usual and customary for sales of such assets and its sale must be highly probable.12 This rule is also not expected to materially increase complexity or compliance costs because many accounting standards already require that such assets be separately identified and accounted for. For example, in the case of IFRS, assets held for sale are separately accounted for under IFRS 5 – Non-current Assets Held for Sale and Discounting Operations.

346. The Sections below provide additional rules on the individual elements of the tangible asset carve-out, starting with property, plant and equipment.

347. Property, plant and equipment are tangible items that are held for use in the production or supply of goods or services or for administrative purposes and are expected to be used during more than one period. Assets in this category include: buildings, machinery, computers and other office equipment, motor vehicles, furniture and fixtures, and land improvements with a limited useful life. Land is also technically part of property, plant and equipment. However, because land is non-depreciable it is excluded from the definition of property, plant and equipment, for purposes of the carve-out, and considered separately.

348. Property, plant and equipment is initially recognised on the balance sheet at its costs, including its purchase price and any costs directly attributable to bringing the asset to the location and condition necessary for it to be capable of operating in the manner intended by management. After initial recognition as an asset, an item of property, plant and equipment is carried at its cost less any accumulated depreciation and any accumulated impairment losses (referred to as the “cost model”). Depreciation refers to the systematic allocation of the cost of an asset, less its residual or “salvage” value, over its useful life. An impairment loss is the amount by which the carrying amount of an asset exceeds its recoverable amount.13

349. Therefore, calculating depreciation of property, plant and equipment requires three factors be established: (i) the depreciable base of the asset, which is the cost of the asset less its residual value, (ii) the useful life of the asset, and (iii) the depreciation method, such as straight-line, diminishing balance and units of production. For purposes of the carve-out, all three factors are required to conform with those used for the same asset for financial accounting purposes, specifically the financial accounts used to compute the GloBE tax base for the relevant Constituent Entity (referred to as the “conformity rule”). For example, if the parent prepares its consolidated financial accounts in accordance with IFRS, then the factors used to compute depreciation, for purposes of the carve-out, must follow IAS 16 – Property, Plant and Equipment.

350. However, there are three exceptions to the conformity rule. The first relates to the revaluation model, which is permitted by some financial accounting standards as an alternative to the cost model. The second relates to asset sales between GloBE group members. The third relates to self-constructed assets.

(a) Revaluation model. Under some financial accounting standards, including IFRS, an entity can elect either the cost model or the revaluation model as its accounting policy. Under the revaluation model, an asset is carried at a revalued amount, which is its fair value at the date of the revaluation less any subsequent accumulated depreciation and subsequent accumulated impairment losses. Revaluation increases are generally recognised in OCI, rather than profit or loss. Revaluation decreases, on the other hand, are generally recognised in profit and loss. Absent a corrective measure the revaluation model would impact the quantum of the carve-out because depreciation expense is determined based on the revalued amount. This result is not appropriate because revaluation increases/decreases have no connection to substantive activities. Therefore, to eliminate the effect of the revaluation model for purposes of the carve-out, any subsequent incremental increase in depreciation resulting from revaluation increases are disregarded. And, any revaluation loss recognised in profit and loss is treated as additional depreciation in the year of the loss, and the subsequent decreased financial accounting depreciation is included in the carve-out base annually. In both cases, i.e., revaluation increases and decreases, the result of this rule is that the total depreciation charge included in the carve-out base over the life of the asset is the same as what it would have been without the revaluation. Such a result recognises that revaluation increases/decreases have no connection to substantive activities. It also eliminates a key difference across accounting standards: those that allow the revaluation model and those that do not.

(b) Intercompany transaction. Under the GloBE rules, transactions between Constituent Entities in different jurisdictions are not eliminated as they would be in the MNE’s consolidated financial statements. Thus, when an asset is sold between Constituent Entities, the buyer may recognise the asset for purposes of the GloBE rules based on the purchase price of the asset, rather than the carrying value of the asset that is in the consolidated financial accounts. This creates the risk of MNEs engaging in intra-group asset sales designed to “refresh” the carrying value of assets and generate a larger carve-out. To prevent this type of planning, any incremental depreciation resulting from an asset acquired in a related party transaction is not allowed in the carve-out base.14 If, however, the gain on the asset sale is recognised in the GloBE tax base, then the resulting increase in depreciation is regarded for purposes of the carve-out.

(c) Self-constructed assets. The cost of a self-constructed asset includes the labour costs, i.e., payroll, of the employees that constructed it. Absent a corrective measure, the labour-related costs of a self-constructed asset would be counted once in the payroll component of the carve-out and again in the tangible asset component. To eliminate such double-counting, the labour-related costs included in the carrying cost of a self-constructed asset are disregarded for purposes of computing the asset’s depreciation. To avoid a further instance of double counting, the carrying costs of a self-constructed asset excludes the depreciation of assets used in its construction.

351. The depreciation charge of property, plant and equipment is recognised in financial profit or loss, and, by extension, the GloBE tax base, unless it is included in the carrying amount for another asset. For example, depreciation on manufacturing equipment is included in inventory. Eventually, when the product is sold, the depreciation charge becomes part of cost of goods sold. For purposes of the carve-out, depreciation includes both depreciation charges that are recognised in profit and loss in the period it is incurred (often referred to as “period costs”), and depreciation charges recognised in profit and loss in the period when the related product is sold (often referred to as “products costs”). Depreciation that is a product cost is included in the carve-out base in the year it is incurred regardless of when the related product is sold. This rule recognises that the timing of the recognition of the depreciation charge in the income statement has no bearing on substantive activities.

352. Most financial accounting standards, including IFRS, require firms to test property, plant and equipment for impairment when events or changes in circumstances indicate book value may not be recoverable. When an asset is in fact impaired, an impairment loss is recognised in profit and loss and the carrying value of the relevant asset is reduced. The post-impairment carrying value then serves as the revised basis for subsequent depreciation. Consequently, absent a corrective measure, impairments would reduce the quantum of the carve-out amount. This result is not appropriate because impairments are typically caused by deteriorating market conditions, poor management, new competition or technological innovations, and are generally not indicative of a reduction in substantive activities. To eliminate the effect of impairments on the carve-out amount, impairment losses are treated as equivalent to depreciation for purposes of the carve-out, thus included in the carve-out base in the year of impairment, and the subsequent decrease in depreciation for financial accounting purposes is regarded.

353. For purposes of computing the carve-out on a jurisdictional basis, depreciation of property, plant and equipment is treated as having nexus in the jurisdiction of the Constituent Entity (which includes a permanent establishment) that uses the asset. As explained in the right-of-use asset Section below, a lessor is not allowed to include assets that it has leased to customers in its tangible asset carve-out base.

354. Land is technically part of property, plant and equipment, however, unlike other property, plant and equipment, land is non-depreciable. As a non-depreciable tangible asset, land does not naturally “fit” in a depreciation-based tangible asset carve-out. Nonetheless, land is considered an indication of substantive activities and a significant non-mobile factor of production, and, on that basis, is included in the tangible asset carve-out base. The mechanic for doing so is a deemed depreciation charge for land, which requires establishing the same three depreciation factors as are required for depreciable property, plant and equipment, that is: (i) the depreciable base (ii) the useful life, and (iii) the depreciation method.

355. For purposes of calculating the deemed depreciation of land, as relevant to the carve-out, the depreciable base is equal to the original acquisition cost of the land, i.e., without regard to revaluation increases/decreases and with an assumed residual value of nil. The useful life is deemed to be [x] years. The depreciation method is deemed to be straight-line. Taken together, this implies that each year the deemed depreciation amount will be [x]% of the original acquisition cost of the land.

356. As previously provided, land that is an investment property is not included in the carve-out base. All other land is included, including land on which a building rests and land used in an agricultural business. The acquisition cost of land should be computed separately from the building and agriculture produce. Separating the value of land from a building or agriculture produce is not expected to materially increase complexity or compliance costs because many financial accounting standards already account for these assets separately. For example, IAS 16 – Property, Plant and Equipment, provides that land and buildings are separable assets and are accounted for separately. Similarly, IAS 41 – Agriculture Assets, scopes out the land related to agriculture activity, and requires the land be separately accounted for under IAS 16 and IAS 40 – Investment Property.

357. Like property, plant and equipment, land is tested for impairment. In the case of land, an impairment could arise when, for example, the area where the land is located experiences a natural disaster such as flooding, an earthquake or a tornado. If the land is in fact impaired, an impairment loss is recognised and the carrying value of the land is reduced. Consistent with the impairment rule for property, plant and equipment, impairment losses on land should be treated as equivalent to deemed depreciation for purposes of the carve-out, thus included in the carve-out base in the year of impairment.

358. For purposes of computing the carve-out on a jurisdictional basis, land is treated as having nexus in the jurisdiction in which the land is located.

359. Natural resources include oil and gas deposits, timber tracts and mineral deposits. These assets are accounted for similarly to depreciable property, plant and equipment. That is, natural resources are initially recognised at cost, including acquisition, exploration-related, and restoration costs. After initial recognition, the asset is carried at its cost less any accumulated depletion and any accumulated impairment losses, i.e., the cost model. 15 Depletion is the allocation of the cost of natural resources, and has a number of similarities to depreciation accounting. Because the usefulness of a natural resource is generally directly related to the amount of resources extracted, the units of production method is widely used to calculate depletion. Service life is therefore the estimated amount of resources to be extracted, e.g., tons of minerals or barrels of oil.

360. For purposes of the carve-out, the assumptions used to compute depletion are required to conform with those used for the same asset for financial accounting purposes, specifically the financial accounts used to compute the GloBE tax base for the relevant Constituent Entity. For example, if the parent prepares its consolidated financial accounts in accordance with IFRS, then the factors used to compute depletion must follow IFRS 6 – Exploration for and Evaluation of Mineral Resources. However, as with property, plant and equipment, an exception applies with respect to the revaluation model. In particular, any incremental increase in depletion resulting from revaluation increases are disregarded. And, any revaluation loss recognised in profit and loss is treated as additional depletion in the year of the loss, and the decreased financial accounting depletion determined after the revaluation is included in the carve-out base annually. Additionally, the depletion charge should be computed without regard to restoration-related costs, which are not incurred until after the natural resource has been extracted from the site.

361. Depletion, being a product cost, is included in the cost of inventory, just as the depreciation on manufacturing equipment is included in inventory. The depletion charge is recognised as cost of goods sold in the income statement when the inventory is eventually sold. Consistent with the rule provided for property, plant and equipment, depletion is included in the carve-out base in the year it is incurred regardless of when the inventory is sold. This rule recognises that the timing of the recognition of the depletion charge in the income statement has no bearing on substantive activities.

362. Natural resources are tested for impairment under financial accounting rules. In the case of a natural resource, an impairment could arise when, for example, exploration for and evaluation of mineral resources in the specific area have not led to the discovery of commercially viable quantities of mineral resources.16 If the asset is in fact impaired, an impairment loss is recognised in profit in loss and the carrying value of the relevant asset is reduced. The post-impairment carrying value then serves as the revised basis for subsequent depletion. Consistent with the impairment rule for property, plant and equipment and land, impairment losses on natural resources are treated as equivalent to depletion for purposes of the carve-out, thus included in the carve-out base in the year of impairment, and the subsequent decrease in depletion for financial accounting purposes is regarded.

363. For purposes of computing the carve-out on a jurisdictional basis, depletion is treated as having nexus in the jurisdiction in which the natural resource is located.

364. A carve-out based on the ownership of tangible assets would lead to a difference between owning and leasing assets. In order to avoid this distortion, the carve-out treats an appropriate portion of the expense of leasing a tangible asset, including buildings and land, in the same way as depreciation of property, plant and equipment.

365. In a lease arrangement a lessee recognises a “right-of-use” asset on its balance sheet representing its right to use the underlying asset and a lease liability representing its obligation to make lease payments. A lessee accounts for right-of-use assets similarly to an owner of property, plant and equipment. Specifically, a lessee initially recognises right-of-use assets based on the present value of the lease payments, and subsequently recognises depreciation and impairment losses i.e., the cost model.17 The depreciation method is typically straight-line and the useful life is generally the earlier of the end of the useful life of the asset or the end of the lease term. The lessee also recognises interest expense on the lease liability.

366. For purposes of the carve-out, the assumptions used to compute depreciation of a lessee’s right-of-use asset must conform with the assumptions used for the same asset for financial accounting purposes, specifically the financial accounts used to compute the GloBE tax base for the relevant Constituent Entity.18 For example, if the parent prepares its consolidated financial accounts in accordance with IFRS, then the assumptions used to compute the depreciation of the lessee’s right-of-use asset must follow IFRS 16 – Leases. However, an exception applies with respect to the revaluation model. In particular, any incremental increase in depreciation resulting from revaluation increases are disregarded. And, any revaluation loss recognised in profit and loss is treated as additional depreciation in the year of the loss, and the decreased financial accounting depreciation determined after the revaluation is included in the carve-out base annually.

367. As with property, plant and equipment, land, and natural resources, a lessee must test the underlying asset for impairment for financial accounting purposes. A leased asset could be impaired for the same reasons property, plant and equipment is impaired: deteriorating market conditions, poor management, new competition, technological innovations, etc. If a right-of-use asset is in fact impaired, an impairment loss is recognised and the carrying value of the relevant asset is reduced. The post-impairment carrying value then serves as the revised basis for subsequent depreciation. Consistent with the impairment rule for property, plant and equipment, land, and natural resources, impairment losses on right-of-use assets are treated as equivalent to depreciation for purposes of the carve-out, thus included in the carve-out base in the year of impairment, and the subsequent decrease in depreciation for financial accounting purposes is regarded.

368. The lessor of an asset is not allowed a carve-out in respect of the depreciation on that asset. This rule reflects the fact that the lessor is not actively using the underlying asset to earn income. It is therefore not a reliable measure of substantive activities of the lessor.

369. For purposes of computing the carve-out on a jurisdictional basis, a lessee’s depreciation of a right-of-use asset is treated as having nexus in the jurisdiction of the Constituent Entity lessee that uses the property in its business.

370. The impact of expenditure based tax credits on a taxpayer’s ETR is more significant for low margin businesses. This is due to the fact that low margin businesses have more credits as a proportion of their total income. However, the formulaic substance-based carve-out based on payroll and tangible assets is designed, in part, to accommodate to some extent the provision of tax credits and other incentives for low margin businesses. Low margin businesses have more expenses as a percentage of their income and thus would secure a relatively larger carve-out. The carve-out will therefore limit the impact of GloBE rules on low margin businesses that are entitled to tax credits for local expenditures.

371. The computation of the ETR for each jurisdiction and the top-up tax for each Constituent Entity in the jurisdiction is set forth in the preceding Sections. This Section describes the three-step process to compute the ETR for each jurisdiction and the top-up tax applicable to each Constituent Entity in a low-tax jurisdiction.20 The ETR for the jurisdiction is computed first. Second, if the ETR is below the minimum rate, a top-up tax percentage is calculated. Third, the top-up tax for each Constituent Entity in the jurisdiction is determined. After each Constituent Entity’s top-up tax is computed under the rules of this chapter, the liability to tax is then determined by the application of the operational rules discussed in the following chapters. For purpose of computing the jurisdictional ETR, the income or loss of a Constituent Entity is the total income or loss of the entity, irrespective of whether the MNE Group owns 100% of the entity. If the adjusted GloBE income for a jurisdiction is zero or a loss, there is no GloBE tax liability for the jurisdiction, and any loss is carried forward under the loss carry-forward rules described in Section 4.2.1

372. For purposes of the ETR computation, the adjusted covered taxes are the covered taxes assigned to the jurisdiction pursuant to the rules in Section 3.4.2, except taxes attributable to income excluded as a permanent adjustment from the GloBE income of the Constituent Entities located in the jurisdiction, increased by the lesser of the total local tax carry-forward or the amount of the local tax carry-forward necessary to achieve an ETR that is equal to the minimum rate. The amount of the local tax carry-forward necessary to achieve the minimum rate is equal to the excess of the adjusted GloBE tax base for the jurisdiction multiplied by the minimum rate over the covered taxes assigned to the jurisdiction. The amount of local tax carry-forward included in the adjusted covered taxes of the ETR computation for any year reduces the amount of the local tax carry-forward available in subsequent years.21

373. If, after taking into account local tax carry-forwards, the ETR remains below the minimum rate, the top-up tax percentage must be computed for the jurisdiction. The top-up tax percentage is the excess of the minimum rate over the ETR for the jurisdiction.

374. Finally, the top-up tax for each Constituent Entity that has positive net income for the year in the jurisdiction is computed by multiplying the adjusted GloBE income of each such Constituent Entity by the top-up tax percentage. The adjusted GloBE income of the Constituent Entity is equal to the entity’s income for the year reduced by its share of:

  1. (a) the current year losses of other Constituent Entities resident or located in the jurisdiction;

  2. (b) the loss carry-forward for the jurisdiction; and

  3. (c) the carve-out determined for the jurisdiction.

375. The current year losses, the loss carry-forward allowed, and the carve-out for a jurisdiction are allocated to Constituent Entities proportionally based on their net income. Specifically, current year losses, the loss carry-forward allowed, and the carve-out amount determined for the jurisdiction are allocated to a Constituent Entity with positive net income based on the ratio of that entity’s net income to the total net income of Constituent Entities in the jurisdiction that have positive net income for the year.

References

[1] OECD (2015), Transfer Pricing Documentation and Country-by-Country Reporting, Action 13 - 2015 Final Report, OECD/G20 Base Erosion and Profit Shifting Project, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264241480-en.

Notes

← 1. IIR tax paid in respect of a preceding taxable year includes IIR taxes that were paid or satisfied through the use of a pre-existing IIR tax credit.

← 2. Further consideration will be given to permitting IIR tax credits to reduce other tax liabilities of the shareholder.

← 3. If an IIR tax liability is offset against other domestic tax liabilities, this should not reduce the MNE’s ETR in the jurisdiction

← 4. It could not be claimed in connection with a liability in another jurisdiction arising pursuant to the other jurisdiction’s application of the undertaxed payments rule.

← 5. The references to temporary differences in this paragraph are not meant to limit the creation or use of an IIR tax credit to situations involving a temporary difference.

← 6. While all fixed percentage mark-ups are indicated with an [x], that should not be interpreted to mean the mark-up necessarily has to be the same for each item.

← 7. Further consideration is required to determine if the tangible asset component should be reduced to account for debt-financed property.

← 8. Further consideration is required to determine the deemed useful life of land for purposes of the carve-out.

← 9. See page 34 of Action 13 Report (OECD, 2015[1]): “In the tenth column of the template, the Reporting MNE should report the total number of employees on a full-time equivalent (FTE) basis of all the Constituent Entities resident for tax purposes in the relevant tax jurisdiction. The number of employees may be reported as of the year-end, on the basis of average employment levels for the year, or on any other basis consistently applied across tax jurisdictions and from year to year. For this purpose, independent contractors participating in the ordinary operating activities of the Constituent Entity may be reported as employees. Reasonable rounding or approximation of the number of employees is permissible, providing that such rounding or approximation does not materially distort the relative distribution of employees across the various tax jurisdictions. Consistent approaches should be applied from year to year and across entities.”

← 10. If an employee’s employment activities and services are not performed in the jurisdiction of the employer, it is likely that the employee generally performs a substantial portion of their employment activities and services in the jurisdiction in which they are resident.

← 11. For example, the direct labour costs of manufacturing related employees are capitalised into work-in-process inventory, then finished goods inventory and subsequently recognised as part of cost of goods sold, which may not be until a year or more after the payroll expenditure was initially incurred.

← 12. IFRS 5 – Non-current Assets Held for Sale and Discontinuing Operations.

← 13. IAS 16 – Property, Plant and Equipment.

← 14. This rule also applies to assets sold to third parties as part of a structured back-to-back transaction.

← 15. Under some financial accounting standards timber tracts are accounted for the same as other natural resources, i.e., cost model. However, under IFRS, specifically IAS 41 – Agriculture, “biological assets”, which includes timber tracts, are valued at their fair value less estimated costs to sell, with changes in fair value included in profit or loss. For purposes of the carve-out, a deemed depletion charge for timber tracts must be derived using the cost model.

← 16. IFRS 6 – Exploration for and Evaluation of Mineral Resources.

← 17. Some financial accounting standards require lessee’s to distinguish between “operating leases” and “finance leases”. Other standards, including IFRS, have a single lessee accounting model which requires a lessee to recognise assets and liabilities for all leases with a term of more than 12 months, unless the underlying asset is of low value.

← 18. For the financial accounting standards requiring lessee’s to distinguish between “operating leases” and “finance leases”, the rental costs for ”operating leases” can be treated as depreciation costs for the calculation of the carve-out base for the GloBE rules.

← 19. As noted in Section 4.3, further consideration will be given to the effect of the carve-out on the calculation of the ETR and top-up taxes under the GloBE, including whether an MNE group that claims the benefit of the carve-out should be required to make a corresponding and proportional adjustment to the covered taxes.

← 20. Attributing a portion of the top-up tax to the each low-tax Constituent Entities Entity is necessary in situations where the profits made by some of these low-tax Constituent Entities are subject to the UTPR or are subject to IIRs applied by separate Parents, for example, due to split-ownership. As explained under sections 6.3 and 7.2 this can be the case if an IIR applies to the profits made by some (but not all) Constituent Entities located in the low-tax jurisdiction. For example, if the Ultimate Parent Entity is not subject to an IIR but a sub-holding parent entity that owns some but not all low-tax Constituent Entities located in a given jurisdiction is subject to an IIR, another Parent’s IIR or another CE’s UTPR may apply with respect to the other Constituent Entities in the jurisdiction.

← 21. In the event that a UTPR taxpayer has a UTPR tax amount that is carried-forward from a prior year (see Section 7.7.4), such amount should not be taken into account for purposes of the ETR computation.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.