Journal of Australian Taxation
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Journal of Australian Taxation |
|
THIN CAPITALISATION: ISSUES ON THE “GEARING RATIO”
By Antony Ting[*]
The new thin capitalisation regime was introduced in 2001, with the stated objective of preventing multinational entities allocating an excessive amount of debt to their Australian operations. Though it has a more comprehensive scope than the old regime, it may not be effective in achieving its policy objective. For instance, the new regime changes the basis of the statutory gearing ratio limit from “debt to equity” to “debt to asset”. The change can have the possibly unintended effect of allowing more debt than the government would like to allow. The adoption of “asset” in the basis also leads to mismatch problems between accounting and tax concepts. The “safe harbour debt” amount may be distorted as a result.
There are flaws in the design of the tests of the gearing ratio. Definition of and the policy design around “cost free debt capital” are particularly problematic and create absurd results. The “worldwide gearing” test is designed in such a way that it undermines the attainment of the regime’s policy objectives. Furthermore, its treatment of interest-free loans creates loopholes and renders the test conceptually incomparable with “adjusted average debt”.
The original thin capitalisation regime in Australia was introduced in Australia in 1987.[1] In the Review of Business Taxation (“the Ralph Report”) it is noted that the regime was “not fully effective at preventing an excessive allocation of debt to the Australian operations of multinationals”, as it covered only “foreign
related party debt and foreign debt covered by a formal guarantee”.[2] One of the recommendations contained in the Ralph Report was that “the thin capitalisation regime be strengthened by applying it to the total debt of the Australian operations of a foreign multinational investor”.[3] It was also recommended that the regime be extended to cover Australian multinationals that invest abroad, as they “can be in a position to allocate excessive debt to their Australian operations”.[4]
The government adopted the recommendations and introduced the new thin capitalisation regime in 2001, with the fundamental objective of “preventing multinational entities allocating an excessive amount of debt to their Australian operations”.[5]
“Gearing ratio” is the central concept of the thin capitalization regime. Its definition, and the tests designed around it, are pivotal in determining whether an entity is regarded as allocating excessive debt to Australia. This article analyses how the new regime defines “gearing ratio” and the implications of changing the basis from “debt to equity” to “debt to asset”. The problems of adopting the accounting concepts of “asset” and “liability” in the tax law will be identified and reviewed, followed by a detailed examination of the different tests on gearing ratio. It is posited that in several areas the new regime may not be effective in achieving its policy objectives. The analysis will focus on the general rules applicable to companies. Special rules covering financial institutions and groups of entities, and specific control rules in relations to partnerships and trusts will not be covered.
To facilitate the discussion on gearing ratio, it is useful to revisit briefly how it was defined under the original thin capitalisation regime. The maximum allowable foreign debt of a company was equal to the amount of “foreign equity product”.[6] “Foreign equity product” was defined as “the amount ascertained by multiplying the foreign equity ... by 2”.[7] “Foreign equity” was in turn defined basically as the sum of share capital, accumulated profits and asset revaluation reserves.[8] In other words, interest expenses of a company would not be disallowed under the regime if its foreign debt was not more than two times its foreign equity. This was commonly referred to as a “debt to equity” ratio of 2:1.
Under the new thin capitalisation regime, interest expenses of a company may be disallowed if its debt exceeds its “maximum allowable debt”, which is defined, inter alia, to be its “safe harbour debt amount”. The “safe harbour debt amount” is in turn defined, inter alia, as basically three quarters of “assets net of non-debt liabilities” of the company.[9] The government does not offer any explicit reason for the change from a statutory gearing limit based on “equity” to one that is based on “assets”. It simply states that the new formula “reflects the policy of a maximum gearing ratio for debt to equity of 3:1”.[10] The government seems to believe that the two bases offer the same result. In fact, they do not. For instance, assume Company A has share capital of $1 and loan of $99 to fund its purchase of equipment of $100. Based on a “debt to equity” ratio of 3:1, the maximum allowable debt calculated based on “equity” will be $1 x 3, ie $3. However, under a basis of “assets” as stipulated in the new regime, the maximum allowable debt will be $100 x ¾, ie $75.[11] The two bases can produce vastly different results. This is because an increase in debt automatically pushes up the value of “assets” (in the form of cash or other assets), which results in an increase of the maximum allowable debt.[12]
A variation of the example can further highlight the issue. Assume Company A increases its loan to $999. Its assets become $1,000. Under “equity” basis, the maximum allowable debt remains to be $3. Under “assets” basis, the amount becomes $1,000 x ¾, ie $750. In other words, under the “assets” basis, Company A can always deduct interest expenses attributable to an amount of debt equal to 75% of its “assets”, even though the equity remains at $1.
More specifically, for the same gearing ratio of 3:1, under the “equity” basis, interest expenses attributable to all debt in excess of three times the equity will be fully disallowed. Under the “asset” basis, interest expenses will be disallowed only when debt exceeds ¾ of “assets”, regardless of the actual level of equity of the company. In light of the fact that the policy objective of the thin capitalisation regime is to prevent “multinational entities allocating an excessive amount of debt to their Australian operations”,[13] the “equity” basis would appear to be more effective than the “assets” basis.
The Ralph Report contains the recommendation that the statutory gearing limit “be set at a ratio of 3:1 debt to equity”.[14] This recommendation was justified on the basis that the “existing thin capitalisation regime uses a gearing ratio of 2:1, but this ratio only includes foreign related party and parent-guaranteed debt ... Recognising that total debt is being included in the rules, ... the Review recommends a ratio of 3:1 be adopted”.[15] It appears that the Ralph Report’s recommendation entailed an increase of the gearing ratio, but not a change of the basis from “equity” to “assets” to calculate the amount of excessive debt. Interestingly, the government seems to follow the same logic. For instance, the government explains the concept of gearing as follows:
An entity’s debt to equity funding can be expressed as a ratio. For example, a ratio of 3:1 means that for every dollar of equity the entity is funded by $3 of debt. This is also known as gearing.[16]
Based on this definition of “gearing”, it states that the thin capitalisation regime “will limit the amount of debt ... where an entity’s debt to equity ratio exceeds certain limits”.[17] The limit is breached under the safe harbour debt test when “the amount of debt ... is greater than that permitted by the safe harbour gearing limit of 3:1”.[18] The government also compares directly the statutory gearing ratios of the old and new regimes as if they were exactly the same thing.[19] This concept of gearing receives the critical twist when the government explains why the safe harbour debt amount is computed under the “assets” basis:
The safe harbour debt amount is determined by multiplying the ... assets of the entity by three-quarters ... This amount represents the maximum amount of debt that the entity can use to fund its net assets ... Similarly, the remaining one quarter represents the minimum amount of equity that the entity must have to satisfy the equity requirements of the rule. This safe harbour funding mix reflects the policy of a maximum gearing ratio for debt to equity of 3:1.[20]
The fallacy of this statement stems from the last sentence, which is true only when the debt to equity ratio of a company is exactly 3:1. As illustrated in the above examples, when a company has excessive debt (ie its gearing is over the 3:1 ratio), the “assets” basis allows more debt than the “equity” basis. Conversely, when the company has a gearing lower than the 3:1 ratio, the “assets” basis allows less debt than the “equity” basis.
It appears that the rationale behind the Ralph Report’s recommendation (and possibly also the government) was simply to relax the original statutory gearing limit from 2:1 to 3:1. It is not clear whether the government consciously changed the basis from “equity” to “assets” with full understanding of the actual consequences, or it conveniently chooses to follow the established thin capitalisation rules of New Zealand.[21] It is possible that the legislation was drafted in such a way that the maximum allowable debt limit is relaxed not only due to a change of gearing ratio from 2:1 to 3:1, but also inadvertently due to the change of bases. This suspicion seems more probable when the issue is reviewed under the historical perspective. The statutory limit on gearing ratio under the old thin capitalisation regime was tightened from 3:1 to 2:1 in 1997. It seems that the government does not intend to reverse that trend and relax the rules, except for the specific recommendations of the Ralph Report. If that is the case, the legislation may fail to reflect accurately the policy intention, thus violating the drafting standard of “policy transparency” suggested in the Ralph Report.[22]
Tax law does not define “assets”, but intricately defines “debt”.[23] This mismatch of definitions can be problematic when they need to be compared with each other under the thin capitalisation regime. The meanings of “assets” and “debt” will be analysed in detail in the following parts. For the present purpose, it is sufficient to know that “assets” has its meaning under accounting rules, while “debt” takes its meaning under the tax law (hereafter referred to as “debt (tax rules)”).
Conceptually, in the comparison of “assets” with “debt (tax rules)”, the following mismatches can occur:
(a) Debt under accounting rules (hereafter referred to as “debt (accounting rules)”), but not debt (tax rules) or “equity interest” under tax law (hereafter referred to as “equity (tax rules)”).
Such item will inflate the value of “assets” without increasing the value of debt (tax rules). For instance, under accounting rules, the present value of future lease payments under a finance lease (as defined under accounting standards) is recognised both as an “asset”
and a “liability”.[24] To facilitate discussion, assume the present value is $100. This $100 of liability recognised in the financial accounts is unlikely to be equity (tax rules), as for instance the lease does not give rise to a return from the company that is “contingent on the economic performance” of the company.[25] The $100 is also unlikely to be debt (tax rules), as a lease is in general excluded from the definition of “debt interest”, unless it falls under the stipulated exceptions (eg a hire purchase arrangement).[26] Assuming the exceptions do not apply to the finance lease, the $100 “liability” would not be “debt interest” either. As a result, “assets” of the company increase by $100 without a corresponding increase in debt (tax rules).
In this case, the thin capitalisation calculation fortunately should still work properly, as the $100 “liability” is likely to be “non-debt liabilities” (as defined in tax law) of the company. “Non-debt liabilities” are deducted from “assets” in the calculation of safe harbour debt amount.[27] It is defined as, among other things, liabilities of the company excluding equity (tax rules) and debt (tax rules).[28] Therefore, the $100 “liability” will cancel out the $100 increase in “assets”. However, even under this scenario, the interest element embedded in the lease payments would escape the thin capitalisation rules. When a company needs to acquire an asset, it may face a choice between direct purchase by raising a loan or a finance lease. The thin capitalisation regime appears to be biased towards the latter, though the two options are very similar in economic substance. This
effect may violate the “investment neutrality” principle under the Ralph Report.[29]
(b) Debt (accounting rules), not debt (tax rules) but equity (tax rules).
Such item will inflate “assets” without affecting debt (tax rules). An example would be the finance lease discussed above, but structured in such a way that it falls under the definition of “equity interest”. The government originally believed “the risk of manipulating the thin capitalisation rules by raising equity would be minimal as issuing shares is a protracted and costly exercise”.[30] However, it forgets that the debt/equity rules under Div 974 create exciting “flexibility” for taxpayers to structure financing arrangements into either “debt interest” or “equity interest”. In particular, “some financial instruments commonly regarded as debt are now classified as equity although they do not necessarily have the features of a traditional equity instrument ... these types of equity interests can be readily moved in and out of entities at around the time that assets and debts are measured for thin capitalisation purposes ...”.[31] The government recently recognised the problem and proposes to address it with a new concept of “excluded equity interest”. Basically, “excluded equity interest” is an equity interest of a company if either:
• both the company and the holder of the interest are subject to thin capitalisation rules but have different valuation days, or
• only the company is subject to the thin capitalisation rules and the interest is a short-term arrangement (ie on issue for less than 180 days).[32]
Such interest is proposed to be deducted from “assets” in computing the safe harbour debt amount.[33] It is doubtful if such measure will be effective in dealing with the problem. First, it is unlikely to cover the finance lease example discussed above, as the term of the lease most likely exceeds 180 days and the financier can readily be an entity not subject to the thin capitalisation rules. Second, the measure assumes that if both issuer and holder of the interest are subject to thin capitalisation rules and have the same valuation days, the equity interest in question will always be “deducted from the assets of the holder thereby reducing its maximum allowable debt”.[34] Unfortunately, that assumption is wrong. Under the definition of “associate entity equity”,[35] only an equity interest issued by “associated entity” will be deducted from “assets” in the safe harbour debt amount calculation. The definition of “associate entity” of a company basically requires a holding of 50% direct control interest or the associate being accustomed to follow instructions of the company in respect of its profits distribution or financial policies on its assets, debt capital or equity capital.[36] As such, an equity interest (other than a direct control interest) issued by a subsidiary to a fellow subsidiary would not be “associate entity equity”, and thus would not be deducted from the “assets” of the latter company. The transaction will still inflate the “assets” of the former company, even if the two companies have the same valuation days. The same result may even be achievable between unrelated companies through the issuance of such “equity interest” under a round robin arrangement. Conversely, “excluded equity interest” may penalise innocent taxpayers, as discussed below.
(c) Equity under accounting rules, but not equity (tax rules).
The definition of “excluded equity interest” may catch genuine long-term equity injection. Assume Company A has 30% shareholding in Company B as genuine long-term investment, and both companies are subject to the thin capitalisation rules. The 30% equity will most likely be “equity interest” under the tax law. However, if the two companies have different valuation days for reasons other than manipulating the value of “assets” in Company B, the value of the 30% equity will invariably be wiped out from the “assets” of Company B under the “excluded equity interest” rule. This is a wrong result. The rule focuses wrongly on “valuation day”, instead of “temporality” which should be the critical criterion in distinguishing a genuine quasi-equity from a short-term tax-motivated injection. A more sensible approach will be to exclude the equity only if the capital injection is temporary and the objective is to manipulate the safe harbour debt amount. For instance, New Zealand disregards any “temporary ... increase in the value of asset ... if the ... increase has a purpose or effect of defeating the intent and application of [its thin capitalisation rules]”.[37]
Another example of mismatch between accounting and tax rules is the definition and valuation of “assets”. This will be discussed in detail in the following Part. Nevertheless, the above discussion may be sufficient to cast serious doubt on the effectiveness of comparing an accounting concept with a tax concept. Some of the above problems would not exist if the thin capitalisation rules were designed to compare “comparable” items, such as “equity interest” and “debt interest”, which are both defined under the tax law and is designed to be mutually exclusive.[38]
The starting point to calculate the safe harbour debt amount is the value of assets of a company.[39] The tax law does not define the term “assets”, but it does require that the valuation of assets “must comply with the accounting standards”.[40] It is not clear from literal reading of the law whether that implies that “assets” would take its accounting meaning. The government originally stated that “the term assets adopts its normal legal meaning”.[41] Under this position, mismatches between accounting and legal concepts are inevitable. For instance, under accounting rules, a finance lease is recognised as an asset of the lessee, not the lessor. However, the plant or equipment under the finance lease would be recognised as “asset” of the lessor under general legal meaning. This would stand oddly with the tax law requirement that its valuation has to be done in accordance with accounting rules.
The ATO takes a different view from the government. It considers that “the term ‘assets’ for the purposes of Div 820 carries its accounting meaning”.[42] This is a pragmatic approach to avoid the mismatch problems discussed above. Unfortunately, some taxpayers may have already restructured their financial affairs based on the government’s initial view, and may have to redo it all over again under the ATO’s position.[43] The government finally acknowledges the issue and proposes to amend the tax law so that “the accounting standards are to be used to determine the definition of an asset ...”.[44] The confusion and ambiguity have been caused by the failure to properly define “assets” in the tax law in the first place. In this respect, the legislation violates the drafting standard of “clarity of rules” in the Ralph Report.[45]
The proposed clarification to adopt accounting meaning of “assets” may remove some of the mismatch problems, but the accounting rules have their own issues. “Assets” of an entity are defined under the Statement of Accounting Concepts 4 (“SAC 4”) as “future economic benefits controlled by the entity as a result of past transactions or other past events”. Not all “assets” falling under this very wide definition are recognised in the financial statements. SAC 4 states that an asset would be recognised when and only when:
• it is probable that the future economic benefits embodies in the asset will eventuate, and
• the asset possesses a cost or other value that can be measured reliably.
This important limitation on the recognition of “assets” is rightly adopted in the proposed amendment to the regime.[46] However, the recognition criteria under the accounting rules are contentious. The term “probable” means that the chance of the future economic benefits arising is “more likely rather than less likely”.[47] It is likely that, with the reliance of the thin capitalisation regime on the rules, accountants will be under even more pressure than before to recognise “assets” in the financial statements. Research and development costs (including product development costs and mineral exploration costs) are good examples. The Australian Accounting Standards allow companies to capitalise such costs as “assets” if “future benefits related to the costs are expected to be greater than the costs beyond any reasonable doubt”.[48] Companies can and do have different interpretations of the term “beyond reasonable doubt” and end up with variations on their capitalisation policies.[49] Recognition of assets can also change over time. An expenditure that fails the recognition criteria at one point in time may qualify for recognition as an asset at a later date. Confirmation of the existence of a valuable mineral deposit would be an example.[50]
More examples of contentious recognition of “assets” can be found in the 30 June 2000 balance sheet of the famous HIH, including:[51]
• “Deferred acquisition costs” of $304 million: the amount represents sums expended in the past to acquire another business. The rationale of recognising it as an “asset” was that it would be gradually “written off” over time as its “benefits” were deemed to have been used up.
• “Goodwill” of $475 million: a contentious item whose value relies heavily on valuations.
• “Future tax benefits” of $228 million: this item was criticised to have no “real world manifestation”.[52] “Future tax benefits” in general has been condemned as “a peculiar artefact arising from the Accounting Standard on ‘tax effect accounting’ ... [that] has been described, by informed commentators, as ‘hocus pocus’ and as leading to ‘accounting fictions’”.[53]
“Reinsurance recovery receivables” of $1,820 million: the auditor’s report directs attention to the “inherent uncertainty” of realising all of the receivables. Despite the recognition rules in SAC 4, the amount was still recognised as “assets” in the balance sheet, presumably because the auditor agreed that the collection of the receivable is “probable”.
The above discussion reveals that the accounting rules on definition and recognition of “assets” could entail another can of worms. Adopting the accounting rules for “assets” would import these contentious issues into the tax system. It is not clear whether or not the government was conscious of these issues when making the decision to adopt the accounting rules.[54]
The tax law requires the valuation to comply with accounting standards.[55] Historical cost is the basis, but Australian companies are allowed to depart from that amount by revaluing non-current assets (ie assets that are expected to have benefits for more than a year into the future).[56] Accounting standards state that each class of non-current assets must be measured on either (a) the cost basis or (b) the fair value basis (ie arm’s length value). The standards also require that when an asset is revalued upwards, all assets within the same class of assets should be valued at the same time on a consistent basis. There is no similar requirement for a downward revaluation of a particular asset, but a downward revaluation is compulsory when its carrying amount is greater than its recoverable amounts (ie its net realisable amount upon continued use and subsequent disposal).
An obvious potential abuse is on the “recoverable amounts”. For instance, provision of doubtful debts will reduce the value of “assets”. A company would prefer to minimise the provision, for various reasons including thin capitalisation purposes. Deciding the amount of the provision is an art, not a science, and ultimately depends on the judgment of the company’s management and accountant. Another issue on valuation is the government’s proposed amendments to the regime to relax the restrictions on revaluation. On one hand, the government states that valuation is “a crucial component of the thin capitalisation provisions” and warned that the “general anti-avoidance provisions of Part IVA of the ITAA36 may apply where these values are manipulated ...”.[57] On the other hand, it proposes to relax the existing restrictions on revaluation of assets in the following ways:
(a) Waive requirement for expert valuer.
Existing tax law requires all revaluation of assets to be made by an expert valuer “whose pecuniary or other interests could not reasonably be regarded as being capable of affecting the person’s ability to give an unbiased opinion in relation to that revaluation”.[58] Employees or directors of a company would in general fail to meet the criterion. Under the proposed amendments, that requirement will be waived if a company is required to prepare financial statements under Australian law and the revaluation is reflected in those statements.[59] The government does not give any explicit reason for the change.[60] In any case, that could be an important concession to the valuation rules. There is no requirement under accounting standards for revaluations to be made by independent valuers. The accounting standard only requires that the company has to disclose in its financial statements “whether the revalued carrying amount has been determined in accordance with an independent valuation”.[61] In other words, company directors or employees, even if they are not expert valuers, can revalue the assets, provided they do so in accordance with the general guidelines stipulated in the accounting standards. The proposed change would imply a significant increase of flexibility for taxpayers on valuation of assets.
(b) Internal expert valuer.
The government also proposes to relax the requirement on revaluation that is not reflected in the financial statements. Under the proposed amendment, such revaluation can be made by an expert valuer who is also an employee of the company, provided an independent external expert valuer has reviewed and agreed to the methodology of the valuation.[62] Again, there is no explicit reason given for the proposed change. Presumably, the objective may be to reduce compliance costs of the taxpayers. In any case, the proposed change again implies a higher risk of manipulation by taxpayers on values of “assets”.
(c) Revaluation of individual assets.
The government proposes to override the accounting standards and allow a company to revalue only one asset without revaluing the whole class of assets, provided that “no asset in the asset class has fallen in value”.[63] No reason is given for the proposed relaxation. Presumably, the government may intend to reduce compliance costs of taxpayers, who under the proposed rule would not need to revalue all assets in the same class. However, the taxpayer is still required to
ensure (and presumably to keep proper documentation) that all other assets in the asset class have not fallen in value. Furthermore, as discussed in (a) above, valuation may be done by directors or employees of the company if the revalued amount is reflected in the financial statements. It is doubtful if the actual saving in compliance costs would be significant and could justify the override of the accounting standards, which are designed to ensure that the valuations of assets are prudent and not manipulated by “selective” valuation of individual assets.
It can be appreciated that valuation of assets is another difficult issue. Furthermore, “window dressing” of the financial statements at year end is a well known practice. The Commissioner is empowered to substitute a value for an asset “having regard to the accounting standards”, if a company has overvalued it. It will be interesting to see how the Commissioner will operate under that provision.[64] Provided the company and its accountant have followed the guidelines in the accounting standards to make their valuation, the Commissioner may find himself in a difficult position to argue a different value under the same set of standards.
A more fundamental issue on valuation is whether revaluation should enter into the thin capitalisation formula in the first place. Revaluation has always been counted under Australian thin capitalisation rules. For instance, under the old regime, asset revaluation reserves were included in the definition of “foreign equity”.[65] It is natural that the new regime likewise includes revaluation in the value of “assets”.[66] However, the basic premise of the thin capitalisation regime is to avoid excessive deductions rising from debt by comparing the amount of debt against a benchmark. The benchmark chosen under the new regime is, among other things, based on the value of “assets”. Conceptually, the value of “assets” can be set as the historical cost, or the fair market value, or any other readily available figure, provided the multiplier ratio (which is currently at three quarters) is adjusted accordingly. In the Ralph Report it is suggested that one of the national taxation objectives of the business taxation system should be “promoting simplification and certainty”.[67] Based on such criteria, and given the above issues and uncertainties embodied in the revaluation of assets, one may argue that a historical cost basis would be much more certain and simple to operate. Such basis arguably also provides a conceptually more accurate outcome. Increase in “assets” due to revaluation in fact is an unrealised amount. It can easily vanish if the value of the asset depreciates the next day. The amount is not “locked up” in the entity like “equity”. It can hardly represent or be equivalent to actual equity provided by the investors in the entity. The figure becomes real equity only when the asset is actually disposed of and the gain realised. There is no surprise that revaluation of assets is not permitted in the US or Canada under their accounting standards. The US equivalent thin capitalisation rule is based on “the adjusted basis [of assets] for purposes of determining gain”.[68] That basically refers to the costs of the assets. The US rules are regarded as “very effective in dealing with high levels of gearing”.[69] It appears that a historical cost basis for valuation of “assets” is a viable and effective alternative to the problematic “fair value” basis.
The definition of “safe harbour debt amount” is based on the value of “assets” less, among other things, the value of “non-debt liabilities”.[7] As such, the concept of “non-debt liabilities” is potentially another source of manipulation. The term is defined, among other things, as “liabilities” of an entity, other than its debt
interest on issue, its equity capital or “a provision for a distribution of profit if the entity is a corporate tax entity”.[71]
The term “liabilities” goes through a similar fate as “assets” discussed above. It is not defined in the tax law, but is required to be valued according to accounting standards.[72] The government originally relied on its legal meaning, namely “presently existing obligations that an entity owes to another entity”.[73] It even emphasised that non-debt liabilities “must be a present obligation ... For example, a requirement to pay long service leave does not become a present obligation until the relevant employee takes the leave accrued”.[74] Such legal meaning contradicts sharply with the accounting rules for the definition and recognition of liabilities. Under SAC 4, “liabilities” are “future sacrifices of economic benefits that the entity is presently obliged to make to other entities as a result of past transactions or other past events”. Prima facie, the definition seems to rely on the same concept of “present obligation”. However, the accountants have a broader interpretation for the term. Liabilities under accounting rules can be legally owed debts, but they “can also be estimates of future payments based on past agreements, such as those arising from promises of future benefits to employees for long service leave”.[75] In other words, many provisions made in the financial statements for estimates of future payments under past agreements would not be “liabilities” under the government’s interpretation.
The ATO was again caught between the ambiguous legislation and the contentious interpretation of the government. It again bravely took the position that “liabilities” should take their accounting meaning.[76] This is a sensible and pragmatic approach in removing the mismatch between the different definitions of the term under legal and accounting contexts and the valuation requirement under accounting standards. In the process of determining its position on the issue, the ATO also considered the alternative position that “whilst assets should have an accounting meaning, liabilities should have a modified accounting meaning under which employee entitlements and other accounting provisions should be excluded”.[77] Fortunately, the ATO does not adopt this position. Going back to the finance lease example, under such position, the lease would be recognised as “assets” but not “non-debt liabilities” under thin capitalisation regime. The safe harbour debt amount would be inflated wrongly. This discussion illustrates again the inherent problems of comparing an accounting-defined item against a tax-defined item.
The government finally acknowledges the mismatch issues and proposes to amend the legislation to formally adopt the accounting meaning of “liabilities” for the thin capitalisation regime.[78] Unfortunately, the issues would be resolved only after all the confusion and unnecessary costs incurred by some taxpayers in restructuring their financial affairs.[79]
The thin capitalisation regime compares “adjusted average debt” with the greatest of “safe harbour debt amount”, “arm’s length debt amount” and “worldwide gearing debt amount” (for outward investors only). Any excess will trigger a disallowance of “debt deduction”. The major issues on these items are discussed below.
Definition of “adjusted average debt” is, subject to various adjustments, based on the “debt capital of the entity that gives rise to debt deductions of the entity for that or any other income year”[80] “Debt capital” is defined as “debt interests issued by the entity”.[81] “Debt interest” in turn is defined to include interest-free debt.[82] Conceptually, as thin capitalisation regime focuses on excessive deduction arising from debt, only debt that gives rise to deduction should be counted in “adjusted average debt”. The definition of the term seems to follow this logic, but unfortunately may be flawed on two fronts:
(a) “Debt deduction”.
This term is widely defined as an otherwise deductible cost incurred in relation to a debt interest, including not only interest, but also “any amount directly incurred in obtaining or maintaining” the debt.[83] This provision is deemed necessary to catch “disguised” interest payments in the form of borrowing costs. However, it may inadvertently catch genuine borrowing costs incurred on interest-free loans. Such loans would then be counted as “debt capital that gives rise to debt deduction”. In other words, the “adjusted average debt” is wrongly inflated by these loans that do not give rise to any interest deduction or its equivalent. The taxpayer may argue that the interest-free loan is in fact quasi-equity, and the borrowing costs are of a capital nature, thus not deductible.[84] Under that position, the borrowing costs may escape from the definition of “debt deduction”.
Nevertheless, a prudent taxpayer may prefer to avoid any such costs on its interest-free loan in the first place.
(b) Debt deduction for “any other income year”.
The definition of “adjusted average debt” for an income year catches debt that gives rise to debt deductions “for that or any other income year”. The policy may wrongly penalise genuine commercial transactions. For instance, assume Company A, a wholly owned subsidiary of a foreign parent company, has $10 share capital, $90 loan from its parent company and $100 loan from a local bank to finance its purchase of assets of $200. Interest is payable on the loans at 10% per annum. In Year 1, “adjusted average debt” is $190, while “safe harbour debt amount” is $200 x ¾, ie $150. Therefore, assuming the arm’s length debt amount is not more than $150, out of the total interest paid of $19, only $15 is deductible under the thin capitalisation regime.[85] However, assume the parent company decides not to charge interest on the loan from Year 2 onwards, as it has to support Company A that is in financial difficulties. Under the definition of “adjusted average debt”, the $90 loan is still counted, even though in substance it has already become quasi-equity, and Company A does not and will not claim any more interest deduction on the loan. “Adjusted average debt” is still $190, and “safe harbour debt amount” is same as Year 1, ie $150. Therefore, out of the actual interest expenses of $100 x 10%, ie $10, only $7.9 is deductible.[86]
The $90 loan should be treated as quasi-equity, and should not be counted in “adjusted average debt”. In that case, the latter would be $100, which is less than the safe harbour debt amount of $150. The whole interest of $10 paid to the bank would therefore be deductible. This should be the correct outcome under the regime.
The government tries to justify the policy by an example.[87] A company may pay “all interest payments in relation to a particular debt” in Year 1. In the absence of such policy, the loan would be treated as quasi-equity for all subsequent years. The government claimed that “[t]his is inappropriate” without giving explicit reasons why that is unacceptable. Going back to the above example on Company A, assume that the company pays all the interest upfront in Year 1. No matter how much interest it pays in Year 1, the maximum interest deduction it can claim is always limited to $15. In fact, due to the way the formula is designed on an annual basis, it would not make sense for a company to make a huge lump sum interest payment in one year. The government’s concern is valid only if the company expects that it can satisfy the gearing ratio tests in Year 1 but not in later years. However, even under that scenario, the taxpayer can easily circumvent the provision by substituting the loan with a new loan in Year 2, subject to the exposure to the general anti-avoidance provisions in Pt IVA of the ITAA36. Similarly, the grieved Company A in our example may easily avoid the undesirable outcome by substituting the $90 loan with a new interest-free loan in Year 2. The difficult financial positions of Company A may serve as a strong ground against any challenge from the general anti-avoidance provision.
Moreover, the policy ignores the actual nature of the $90 loan, which has changed from an interest-bearing loan to quasi-equity. It also fails to keep the proper focus on the primary target of the whole regime, namely “excessive deduction due to debt”. It is doubtful that the provision would effectively achieve the policy intention of the thin capitalisation regime in any case.
In this regard, the old thin capitalisation regime seems to be more reasonable. A debt is counted only if “interest is or may become payable” in respect of the debt.[88] The ATO rightly took the position
that a debt will not be counted during a period “for which interest can never be charged”.[89] If interest-free period ceases, the debt will be counted from the end of that period.[90] The rules kept proper focus on “excessive deduction arising from debt” and arguably produced more equitable outcomes. In the design of the new regime, the government unfortunately seems to lose the focus on the fundamentals, but becomes preoccupied with anti-avoidance considerations.
“Cost free debt capital” is an addition to the amount of “adjusted average debt”.[91] The term is defined basically as, among other things, a debt interest not giving rise to any “debt deduction” at any time and either:
• both the issuer and the holder of the debt interest are subject to thin capitalisation rules, and they adopt different valuation days; or
• only the issuer is subject to thin capitalisation rules, and the term of the debt interest is less than 180 days.[92]
Intuitively, one would wonder why an effectively interest-free loan (which is in substance quasi-equity) should be added to “debt” in the thin capitalisation calculation. The government’s policy intention is essentially anti-avoidance. It recognises that treating interest-free debt as quasi-capital has the effect that “the borrower’s assets ... increase without any corresponding increase to its adjusted
average debt. This creates the opportunity for the safe harbour debt amount calculations to be manipulated by providing an interest free loan and then repaying it shortly after the borrowing entity’s valuation day”.[93] This is a genuine concern. Unfortunately, the adjustment ends up in the wrong place. As the effect of the abuse is an increase of “assets”, one would naturally think the adjustment should reduce “assets”. Miraculously, the adjustment in the tax law is to increase “debt”. This has different interesting effects for inward and outward investors.
(a) Inward investors
Assume that Company A, a wholly owned subsidiary of a foreign parent company, has share capital of $10 and loan of $90 (at 10% interest per annum) to fund its purchase of assets of $100. Assume further that the “arm’s length debt amount” is not more than $75. Then “adjusted average debt” is $90; “safe harbour debt amount” is $100 x ¾, ie $75. Deductible interest is therefore $7.5.[94] Now, assume the foreign parent injects $100 “cost free debt capital” into Company A. The “adjusted average debt” becomes $100 + 90, i.e. $190; “safe harbour debt amount” becomes $200 x ¾, ie $150. Deductible interest becomes $7.1.[95] This is smaller than the amount of deduction (ie $7.5) that Company A should have if the “cost free debt capital” does not exist. The adjustment of “cost free debt capital” therefore has a penal effect for inward investors.
The result would be different if the adjustment is made to “assets” instead of “debt”. In that case, “adjusted average debt” is $90; “safe harbour debt amount” is $(200 - 100) x ¾, ie $75. Therefore, this approach exactly cancels out the injection of “cost
free debt capital” and fulfils the policy intention.[96] One may argue that a penal effect is justified as a determent to anti-avoidance transactions. However, bearing in mind that the provision may inadvertently catch genuine interest-free loans (as discussed below), it is doubtful if the penal effect is justified. The penal effect is more questionable when it is compared with the “rewarding” effect on outward investors as discussed in (b) below.
(b) Outward investors
Assume that Company B, an Australian company with a wholly owned offshore subsidiary, has share capital of $10 and loan (from an independent bank) of $90 (at 10% interest per annum) to fund: (i) share investment in the subsidiary of $20 and (ii) purchase of other Australian assets of $80. Assume also that the subsidiary has no loan. Then “adjusted average debt” is $90; “safe harbour debt amount” is $(100 - 20) x ¾, ie $60.[97] “Worldwide gearing debt amount” is $73.2.[98] Assuming the arm’s length debt amount is not more than $73.2, the deductible interest would be $7.3.[99]
Now, assume Company B receives $100 “cost free debt capital”. Then “adjusted average debt” becomes $190; “safe harbour debt amount” becomes $(200 - 20) x ¾, ie $135. “Worldwide gearing debt amount” becomes $172.[100] Deductible interest is therefore $8.1.[101]
This is greater than $7.3 in the base case. The figure is even greater than the amount of interest that Company B would have incurred if all of its Australian assets are fully funded by loans, namely, $80 x 10%, ie $8. The absurd effect, partly due to the wrong addition of “cost free debt capital” to “adjusted average debt” and partly to the problematic formula of the “worldwide gearing debt amount” (which will be discussed in Part 3.3 below), completely defeats the policy objective of “cost free debt capital”. This result is also at odd with the penal effect for inward investors.[102]
If the adjustment is correctly made to “assets” instead of “debt”, then “adjusted average debt” is $90; “safe harbour debt amount” is $(200 - 20 - 100) x ¾, ie $60. Furthermore, provided the formula of “worldwide gearing debt amount” is properly adjusted to exclude “cost free debt capital” from both “worldwide debt” and “worldwide equity”, this approach would again cancel out exactly the effect of the injection of “cost free debt capital” and fulfils the policy intention nicely.[103]
Another issue relating to “cost free debt capital” is in the disallowance formula for debt deduction. The disallowance ratio is defined as:
[the excess of “adjusted average debt” over maximum allowable debt] to
[the sum of “cost free debt capital” and debt capital that gives rise to debt deduction in any income year].[104]
The government adds “cost free debt capital” to the denominator because the amount “is treated as debt”.[105] This is wrong in the following respect:
(i) As discussed above, treating “cost free debt capital” as debt is conceptually wrong. The amount should be deducted from “assets” instead.
(ii) As the amount is added to the denominator, less debt deduction will be disallowed. This contradicts squarely with the policy intention.
(iii) The formula is to calculate the proportion of total “debt deduction” to be disallowed. Conceptually, the denominator of the disallowance ratio should be the total amount of debt that gives rise to the “debt deduction”. Interest-free loan should not be counted in the denominator.
“Cost free debt capital” seems to be a perplexing issue for the government. A substantial portion of the amendments to the regime enacted in 2002 was devoted to the issue.[106] Unfortunately, the adjustment ends up in the wrong place and creates the above anomalies. Furthermore, the definition of the term is also problematic and can catch genuine interest-free loans. For instance, if both the issuer and the holder of such loans are subject to the thin capitalisation rules, and they have different valuation days, the loans would be “cost free debt capital” and treated as debt. This is regardless of whether the loan is genuine long-term finance provided as quasi-equity. It is doubtful if the “valuation day” is a relevant test at all. The government’s true target for the measure is “short-term interest-free debt funding, [which] may potentially be used to inflate the safe harbour debt amount of the entity. It is the ease with which such interest free funding may be provided and removed that causes concern”.[107] It appears that the “valuation day” test fails to keep the correct focus on “temporality” as the critical criterion in distinguishing between genuine quasi-equity and short-term injection of interest-free loan. Different valuation days of the issuer and the holder would not, and should not, affect the quasi-equity nature of a long-term interest-free loan. In this regard, again the New Zealand’s approach to tackle temporary fluctuation of “assets” seems more sensible and effective.[108]
Another issue with the definition of “cost free debt capital” is a technical error. The law requires that, for the issuer of the debt interest, “the result of applying the formula in s 820-37 to the entity for the income year is less than 0.9”.[109] The formula in that section should in fact apply only to outward investors. The requirement inadvertently has the effect of excluding from the definition interest-free debt issued by some inward investors. Clearly this is against the government’s intention.[110] The legislation thus violates the drafting standard of “policy transparency” in the Ralph Report.[111] The government recently proposed to amend the legislation to remove the error.[112]
“Worldwide gearing” test is a privilege to Australian multinational investors.[113] The policy intention is to allow such investors to “gear the Australian operations up to 120% of the worldwide gearing ratio” and thus “accommodates businesses with a high level of gearing in Australia that need equity to expand offshore”.[114]
Debt deduction of an Australian multinational investor would not be disallowed even if it fails the “safe harbour debt amount” test, provided its “adjusted average debt” is not higher than its “worldwide gearing debt amount”.[115] The latter item is defined basically as, among other adjustments, “assets less non-debt liabilities” times 120% of “worldwide gearing ratio”.[116] The last item is defined as the ratio of “worldwide debt” to “worldwide equity”. “Worldwide debt” is defined basically as the debt interests issued by the Australian multinational investor and its controlled foreign entities, excluding inter-entity debts.[117] “Worldwide equity” is similarly defined in the same subsection. The design of the definitions gives rise to the following issues.
(a) Australian assets fully funded by debt.
To facilitate the discussion, go back to the example of Company B in Part 3.2(b) above. There is an Australian multinational investor with a wholly owned offshore subsidiary, has share capital of $10 and bank loan of $90 (at 10% interest rate) to fund equity investment of $20 in the subsidiary and purchase of other Australian assets of $80. As calculated above, “worldwide gearing debt amount” is $73.2, if the subsidiary has no loans; deductible interest is $7.3. Now assume the overseas subsidiary has $200 loans. “Worldwide gearing debt amount” becomes $77.8.[118] Deducible interest becomes $7.8.[119] This is the result even if the $200 loans are short-term interest-free loans. The calculation illustrates that the maximum allowable debt amount approaches the amount of Australian assets of $80 as “worldwide debt” increases. More drastic results occur if Company B itself increases its loans by short-term interest-free loans.[120] In that case, Company B can claim an interest deduction that is greater than the amount of interest it would have incurred if all of its Australian assets are fully funded by interest-bearing loans.
The result defeats the policy objective of preventing an Australian multinational investor from allocating “excessive amounts of debt to its Australian operations to maximise deductions for interest”.[121] Going back to the base case of Company B, one may argue that conceptually the maximum allowable debt to fund the Australian assets of $80 should be $80 x 90/100, ie $72. The amount is based on the premise that Australian assets and overseas assets of the company should be funded equally by the same gearing ratio. This result would seem to be in more compliance with the policy objective.
(b) Loophole for short-term interest-free loans.
The definition of “worldwide debt” does not exclude short-term interest-free loans, making the item an easy target for manipulation. As shown in (a) above, increase of “worldwide debt”, even by injection of short-term interest-free debt to Company B or its overseas subsidiary, would imply a higher “worldwide gearing debt
amount”. The lax definition of “worldwide debt” is in sharp contrast with the determined (though flawed) attempt of the government to avoid such short-term interest-free loans under the concept of “cost free debt capital”, as discussed in Pt 3.2 above. Again, the New Zealand rules may offer a possible option to tackle the issue. In calculating a similar “worldwide group debt percentage”, any “temporary increase in the amount of total debt ... must be excluded from the calculations if the increase ... has a purpose or effect of defeating the intent and application” of the percentage.[122]
(c) Genuine interest-free loans.
As discussed above, “worldwide debt” includes interest-free loans. This implies that the gearing ratio calculated under the “worldwide gearing debt amount” test includes such loans. This contrasts with the definition of “adjusted average debt” which, in general, only includes “debt capital ... that gives rise to debt deductions ...”.[123] The two terms are compared with each other to determine if a company has excessive debt under the regime. It is highly doubtful if the two are comparable at all in this respect. Australian multinational investors would not complain on this point, as the incomparability is to their advantage. One may suspect that this is another way the government, either intentionally or unintentionally, treats Australian multinational investors more favourably than foreign inward investors.
(d) The 120% uplift.
It is not clear why the “worldwide gearing” test is set at 120% of the worldwide gearing ratio of the Australian multinational investors. As discussed in (a) above, this allows a deduction more than the “fair share” of interest expenses attributable to the Australian assets. The Ralph Report explained the rationale for the 120% uplift as to accommodate “businesses with a high level of gearing in Australia that need equity to expand offshore”.[124] The statement seems to imply that the investors would be allowed to have a higher gearing in Australia than overseas. If that is the case, the policy would contradict with the fundamental objective of the thin capitalisation regime, namely, to avoid excessive allocation of loans to Australian operations. Nevertheless, one may suspect that the policy follows the convenient example from New Zealand’s thin capitalisation rules, which allows an uplift of 110%.[125]
If “adjusted average debt” exceeds “safe harbour debt amount” and “worldwide gearing debt amount” (if applicable), a company may still avoid disallowance of its debt deduction if the former is less than the “arm’s length debt amount”.[126] The policy objective of the “arm’s length debt” test is to recognise that “some funding arrangements may be commercially viable notwithstanding that they exceed the prescribed limits. It also makes the rules more consistent with Australia’s DTAs”.[127] The intended scope of application of the test is quite limited, as the government states that the “test will be of most use in those industries where it is common practice to operate with higher debt to equity ratios”.[128] The statement implies that the test would unlikely be applicable to companies outside such specific industries. The legislation, in this instance, correctly reflects such intention. In particular, the test is based on a notional arm’s length debt from an independent commercial lending institution that the Australian operation would reasonably be expected to obtain. The notional amount is determined under the assumptions that, among other things, the entity has no foreign operation or investment, and
no credit support from group members.[129] The government recently indicated that it intends to further tighten the test by the additional assumptions that “the entity’s only activities ... were the Australian business” and “the entity’s only assets and liabilities ... were those [attributable to the Australian business]”.[130] In other words, the test ignores the credit rating and financial strength of any associate of the company, and simply looks at the company’s Australian operation in isolation. This approach is a very strict application of the arm’s length principle.[131] One can comfortably predict that a company with any credit support from its associates in obtaining a loan would most likely fail the test.
It is not surprising that the rules of the test have been described as “carefully framed in such a way they will be difficult, if not impossible, to satisfy in many cases, even where all the debt of an entity effectively comes from third parties”.[132] The following points make a taxpayer even more reluctant to rely on the test:
(a) The “arm’s length debt amount” determined by the taxpayer is subject to the Commissioner’s agreement.[133] A taxpayer may prefer the certainty offered by the other mechanical tests, unless it can obtain an advance ruling from the Commissioner. However, a ruling may be required every year, as there will be inevitable changes in the numerous factors covered by the test.
The test has to be performed every year, as the “maximum allowable debt” amount has to be determined every year.[134] This requirement is confirmed by the government’s comment that the “arm’s length debt analysis must ... be conducted in respect of each income year”.[135] The implication is that the same loan that passes the test this year may not pass the test in future years. This seems unfair especially if the loan comes from a third party who finds no problem to continue to provide the same loan.[136]
Some suggested that a letter from an independent commercial lending institution stating the maximum amount of loan it would be prepared to lend may satisfy the test.[137] However, this is highly doubtful as the ATO states clearly that such letter “could not of itself satisfy the arm’s length debt test”.[138] The ATO argued that the letter “would not satisfy the legislative requirements of Div 820. For instance, the bank may not have made adequate regard to the factual assumptions required by the legislation ...”.[139] It appears there is no shortcut for the test. The ATO tries to assist taxpayers to apply the test by suggesting a non-mandatory “6-step methodology”.[140] Basically, the methodology organises the factual assumptions and factors listed in the law into a step-by-step format. The basic problems of the test stay intact. It is doubtful if the methodology would be of substantial help to taxpayers.
The new thin capitalisation regime is more comprehensive than the old version in its scope and application. However, the new regime may not be effective to achieve the policy objective of preventing interest deduction due to excessive allocation of debts to Australian operations by multinational investors. One of the problems arises from the change of the basis of the statutory gearing ratio limit from “debt to equity” to “debt to asset”. The change can have the unintended effect of allowing a greater amount of safe harbour debt. The adoption of “assets” in the basis also creates mismatch problems between accounting and tax concepts, which can distort the safe harbour debt amount. Recognition and valuation of “assets” and “liabilities” under accounting rules are problematic and subject to manipulations.
Tests of the “gearing ratio” are problematic. Definitions of “adjusted average debt”, together with the way “debt deduction” is defined, may inadvertently catch genuine interest-free loans that should be treated as quasi-equity. Legitimate commercial decisions to waive interest payments on an existing loan may attract penal effect from the regime. Definition of and the policy design around “cost free debt capital” create absurd results. Definition of the term fails to focus on the “temporality” criterion of interest free loan, and may inadvertently catch genuine long-term interest-free loans. Instead of being a reduction to “assets”, the item is wrongly added to “adjusted average debt”, creating a penal effect on foreign multinational investors but an even more undeserved rewarding outcome for Australian multinational investors. The inclusion of the item in the denominator of the disallowance ratio contradicts with the design principle and policy objective of the regime.
The “worldwide gearing” test is designed in such a way that Australian assets of Australian multinational investors may be fully funded by loans. The “120% uplift” in the test allows such investors to have a higher gearing ratio in Australia than overseas. These results are against the basic policy intention of the regime. The failure of the test to exclude interest-free loans creates loopholes for manipulation and renders the term “worldwide gearing debt amount” conceptually incomparable with “adjusted average debt”.
It is disappointing to see a new piece of tax legislation with these conceptual flaws and design problems. It is reasonable to expect that quality control on the legislation should have been performed more vigorously and thoroughly. The issue may have been, at least partly, due to the sheer volume of new tax legislation that was pushed through in a short period of time.[141] One may prefer a better balance between quality and quantity.
[*] Lecturer of Taxation Law, the University of Sydney. The original version of the paper was prepared during the study for Master of Taxation in the University.
[1] Income Tax Assessment Act 1936 (Cth), Div 16F (“ITAA36”).
[2] Review of Business Taxation, A Tax System Redesigned (1999) 659 (“Ralph Report”).
[3] Ibid (recommendation 22.1).
[4] Ibid 665 (recommendation 22.5).
[5] Explanatory Memorandum to the New Business Tax System (Thin Capitalisation) Bill 2001, para 1.6 (“EM 2001”).
[6] ITAA36, s 159GZS.
[7] ITAA36, s 159GZA.
[8] ITAA36, s 159GZG.
[9] For example, pursuant to s 820-95 of the Income Tax Assessment Act 1997 (Cth) (“ITAA97”), safe harbour debt amount for outward investor (general) is defined in general as, disregarding any amounts attributable to the entity’s overseas permanent establishment, ¾ of the entity’s assets less non-debt liabilities, subject to various adjustments for “associate entity debt”, “associate entity equity”, “controlled foreign entity debt”, “controlled foreign entity equity” and “associate entity excess amount”. The adjustments are designed to avoid double counting debt and equity holdings between associate entities, and to exclude debt loaned to and equity contributed to the entity’s controlled foreign entities. In order to focus on the basic issues of “debt to equity” vs “debt to asset”, these adjustments are in general ignored in the discussion. In respect of debts among associates, the analysis will generally focus on the entity that ultimately receives the debt and does not on-lend to another associate. Similar qualification applies to reference to “equity”.
[10] EM 2001, above n 5, para 2.55.
[11] In order to focus on the issue of “equity” vs “assets” bases, ignore for the moment the alternative tests of “arm’s length debt amount” and “worldwide gearing debt amount”, which will be discussed in detail below.
[12] The analysis will be more complicated if Company A on-lends the loan to its associate, but the overall result will be the same. Assume Company A on-lends the $99 loan to its associate Company B, who does not on-lend the funds to another associate. The maximum allowable debt for Company A becomes $0.75, as the $99 loan to Company B is deducted from its $100 assets as “associate entity debt”. However, the issues of “asset” basis is simply pushed downwards to Company B, who will have the same result under the thin capitalisation regime as Company A under the original scenario.
[13] EM 2001, above n 5, para 1.6.
[14] Ralph Report, above n 2, (recommendation 22.3).
[15] Ibid [Emphasis added].
[16] EM 2001, above n 5, para 1.16 [Emphasis added].
[17] Ibid para 1.18.
[18] Ibid para 1.22 [Emphasis added].
[19] Ibid para 2.5 states that “under the new regime the maximum debt to equity ratio allowed is 3:1, compared to 2:1 under the current provisions”. The statement is repeated in para 3.9 under the Chapter for outward investing entities (non-ADI).
[20] Ibid para 2.55 [Emphasis added].
[21] New Zealand introduced its thin capitalisation regime in 1995. The rules are contained in Pt FG of the Income Tax Act 1994 (“ITA94”). In particular, ss FG 3 and FG 4 effectively stipulate that the maximum allowable debt is 75% of assets.
[22] Ralph Report, above n 2, 114. The drafting standard requires tax legislation to “disclose the policy intention and design purpose underlying the rules”.
[23] ITAA97, Div 974.
[24] K Trotman and M Gibbins, Financial Accounting: an Integrated Approach (2003) 374-375.
[25] ITAA97, s 974-75(1).
[26] ITAA97, s 974-130(4). In particular, lease rental expenses are specifically excluded from the definition of “debt deduction”, provided “the lease is not a debt interest”: ITAA97, s 820-40(3)(d). Under the definition of “adjusted average debt”, in principle only debts that give rise to “debt deduction” are included: ITAA97, ss 820-85(3) and 820-185(3). This is subject to the important qualification of “cost free debt capital”, which is discussed in detail in Pt 3.2 below.
[27] ITAA97, ss 820-95 and 820-195.
[28] ITAA97, s 1(1).
[29] Ralph Report, above n 2, 111. The principle requires that business tax law should “avoid differentially taxing ... entity financing alternatives”.
[30] Explanatory Memorandum to the Taxation Laws Amendment Bill (No 5) 2003, para 1.90 (“EM 2003”).
[31] Ibid para 1.91.
[32] Ibid para 1.92.
[33] Ibid para 1.93.
[34] Ibid para 1.91.
[35] ITAA97, s 820-915.
[36] ITAA97, s 820-905.
[37] ITA94, s FG 4(8). The seemingly harsh effect of the words “or effect” is dealt with by Tax Information Bulletin Vol 7, No 11, March 1996, which states at 20 that temporary changes in asset levels “that occur in the ordinary course of a taxpayer’s business would be unlikely to have an attendant avoidance purpose or effect and would therefore not be caught”.
[38] ITAA97, s 974-70(1)(b).
[39] ITAA97, ss 820-95, 820-195 and 820-205.
[40] ITAA97, s 820-680(1).
[41] EM 2001, above n 5, para 1.72 [Emphasis added].
[42] TR 2002/20, para 4.
[43] The ATO recognises this issue and allows these taxpayers to continue adopting the legal meaning “for the income year commencing on or after 1 July 2001 and the following income year”: ibid para 10.
[44] EM 2003, above n 30, para 1.131.
[45] Ralph Report, above n 2, 114. The standard requires tax rules to be “clear, certain and consistent”.
[46] Under Taxation Laws Amendment Bill (No 5) 2003 (“Bill 2003”), a new s 820-680(1A) will stipulate that “the entity has an asset or liability ... if, and only if, according to the accounting standards, the asset or liability can or must be recognised at that time”.
[47] Trotman, above n 24, 414.
[48] Ibid 372 [Emphasis added].
[49] For actual examples of the different policies, see ibid.
[50] Ibid 414.
[51] Ibid 493-4.
[52] Ibid 414.
[53] Ibid.
[54] The situation can be compared with the recent introduction of a new amortisation regime for intangible assets for companies subject to corporation tax in the UK. The regime, effective from 2002, basically adopts the accounting definition of “intangible assets” in the tax law. But that decision was the result of a lengthy consultation process since 1998, with the possible drawbacks of adopting accounting rules clearly and explicitly examined: see for example “Reform of the taxation of intellectual property: a Technical Note by the Inland Revenue”, issued by the Inland Revenue in March 1999, paras 113 and 114.
[55] ITAA97, s 820-680(1).
[56] Trotman, above n 24, 368. The discussion in the paragraph on revaluation of non-current assets under accounting standards is based on the same source.
[57] EM 2001, above n 5, para 8.51.
[58] ITAA97, s 820-680(2).
[59] Bill 2003, Sch 1 Pt 4 item 14, new s 820-680(2A).
[60] The government simply states in para 1.47 of EM 2003 that the amendments clarify that a company can use such revaluation for thin capitalisation purposes. Literal reading of para 8.42 of the original EM 2001 would suggest that the true intention of the government was to require the external independent valuation only if a company wants to use a revaluation figure for thin capitalisation purposes that is different from the value reflected in the financial statements. If that is the case, the original legislation fails to reflect correctly the policy intention of the government, and violates the drafting standard “Policy transparency” in the Ralph Report, above n 2, 114.
[61] AASB 1041, “Revaluation of Non-current Assets”, para 7.1(b) [Emphasis added].
[62] Bill 2003, Sch 1 Pt 4 item 14, new s 820-680(2B).
[63] EM 2003, above n 30, para 1.50. The proposed new rule will be in s 820-680(2C).
[64] ITAA97, s 820-690.
[65] ITAA36, s 159GZG.
[66] ITAA97, s 820-680(1).
[67] Ralph Report, above n 2, 104.
[68] Internal Revenue Code, s 163(j)(2)(C)(i).
[69] Review of Business Taxation, An International Perspective (1998) 103.
[70] ITAA97, ss 820-95, 820-195 and 820-205.
[71] ITAA97, s 995-1(1).
[72] ITAA97, s 820-680(1)(b).
[73] EM 2001, above n 5, para 2.52.
[74] Ibid para 2.53 [Emphasis added].
[75] Trotman, above n 24, 17 [Emphasis added].
[76] TR 2002/20, para 5. The ATO acknowledged that its position contradicts with that of the government in EM 2001: see para 30.
[77] TR 2002/20, para 29.
[78] EM 2003, above n 30, para 1.131.
[79] Similar to the confusion surrounding the meaning of “assets”, some taxpayers may have already restructured their financial affairs based on the government’s interpretation. The ATO allows these taxpayers to continue adopting such position for the income year commencing on or after 1 July 2001 and the following income year. See TR 2002/20, para 10. In other words, these taxpayers will have to restructure again within the time limit to conform to the new meaning of “liabilities” under the regime.
[80] ITAA97, ss 820-85(3) and 820-185(3) [Emphasis added].
[81] ITAA97, s 995-1(1).
[82] ITAA97, Subdiv 974-B. For instance, “debt interest” includes a financing arrangement under which, among other things, “it is substantially more likely than not that the value provided ... will be at least equal to the value received ...”: ITAA97, s 974-20(1)(d). Under the valuation rules in s 974-35, an interest free loan will satisfy the requirement provided its term is not more than 10 years.
[83] ITAA97, s 820-40(1)(a)(iii).
[84] ITAA97, s 8-1(2)(a).
[85] The disallowed interest is $19 x (190 - 150)/190, ie $4: ITAA97, s 820-220. Therefore, the deductible interest is $19 - 4, ie $15, which is equivalent to the interest attributable to the amount of safe harbour debt amount of $150.
[86] The disallowed interest is $10 x (190 - 150)/190, ie $2.1, resulting in a deductible interest of $10 - 2.1, ie $7.9.
[87] Supplementary Explanatory Memorandum to the New Business Tax System (Thin Capitalisation) Bill 2001, para 1.4 (“Supplementary EM 2001”).
[88] ITAA36, s 159GZF(1)(a).
[89] IT 2479, para 10. This is to be contrast with the situation under which the loan is interest-free only if some conditions are satisfied. Such “contingent” interest-free loan would be counted as “debt”: para 11.
[90] Ibid.
[91] ITAA97, ss 820-85(3) and 820-185(3).
[92] ITAA97, s 820-946. The existing provision has a technical error, which inadvertently catches some debt interest that is issued shortly before the valuation day, even though its term is not less than 180 days. The error will be removed by Bill 2003, Sch 1 Pt 7 item 31.
[93] Explanatory Memorandum to the Taxation Laws Amendment Bill (No 4) 2002, para 1.37 (“EM 2002”) [Emphasis added].
[94] Disallowed interest is $9 x (90 - 75)/90, ie $1.5: ITAA97, s 820-220. Deductible interest is $9 - 1.5, ie $7.5, equivalent to the interest attributable to the amount of “safe harbour debt amount”.
[95] Disallowed interest is $9 x (190 - 150)/190, ie $2.9. Deductible interest is therefore $9 - 2.9, ie $7.1.
[96] The actual allowable interest is still different from the base case because in the disallowance formula in s 820-220, “cost free debt capital” is added to the denominator of the disallowance ratio. This is arguably also wrong, and will be discussed below. If this issue is fixed, the approach will produce the same deductible interest as the base case.
[97] ITAA97, s 820-95.
[98] The amount is $(90/10 x 1.2)/(90/10 x 1.2 + 1) x 80, ie $73.2: ITAA97, s 820-110(1).
[99] Disallowed interest is $9 x (90 - 73.2)/90, ie $1.7: ITAA97, s 820-115. Deductible interest is therefore $9 - 1.7, ie $7.3.
[100] As “worldwide debt” is defined in ITAA97, s 995-1(1) to include also interest-free loans, the “worldwide gearing debt amount” is $(190/10 x 1.2)/(190/10 x 1.2 + 1) x 180, ie $172.
[101] Disallowed interest is $9 x (190 - 172)/190, ie $0.9. Deductible interest is therefore $9 - 0.9, ie $8.1.
[102] The thin capitalisation regime is also biased against inward investors in other areas. For instance, “worldwide gearing” test, which basically allows a taxpayer to have a gearing ratio for its Australian operations up to 120% of its worldwide gearing ratio, is available to Australian multinational investors, but not foreign investors. Furthermore, an Australian multinational investor would not be subject to the regime if Australian assets of the investor and its associates represents at least 90% of their total assets: ITAA97, s 820-37. However, foreign investors do not enjoy similar treatment, and are subject to the regime even if their assets in Australia represent not more than 10% of their total assets. The bias seems to contradict with the government’s intention to promote Australia as “a location for internationally focused companies”: see Treasury, Review of International Taxation Arrangements (2002) vii.
[103] Due to the same reason as explained in n 69 above, the exact amount of deductible interest is still affected by the faulty formula of disallowance ratio in s 820-115. If “cost free debt capital” is properly excluded from the denominator of the ratio, the approach will produce the same deductible interest as the base case.
[104] ITAA97, ss 820-115 and 820-220.
[105] EM 2002, above n 93, para 1.51.
[106] The issue occupies paras 1.28 to 1.53, out of a total of pars up to 1.73 in the Chapter for thin capitalisation in EM 2002, above n 93.
[107] Supplementary EM 2001, above n 87, para 1.8.
[108] ITA94, s FG 4(8). For a brief discussion of the provision, see Pt 2.2(c) above.
[109] ITAA97, s 820-946(1)(c)(ii).
[110] EM 2002, above n 93, para 1.42 states the government’s policy intention of mentioning s 820-37, namely, the definition of “cost free debt capital” should cover debt interest issued by a borrower that, among other things, is not excluded from the thin capitalisation regime due to the threshold test in s 820-37.
[111] Ralph Report, above n 2, 114.
[112] EM 2003, above n 30, 1.71. A similar technical error exists in the definition of “associate entity debt” in ITAA97, s 820-910(2)(b)(ii). It will also be rectified by Bill 2003: EM 2003, above n 30, para 1.73.
[113] The term “Australian multinational investors” is used here for easy discussion purposes. Technically, it specifically refers to outward investing entities (non-ADI) (as defined in ITAA97, s 820-85(2)), that is not also an inward investment vehicle (general) or an inward investment vehicle (financial) (both terms defined in ITAA97, s 820-185(2)): ITAA97, s 820-90.
[114] Ralph Report, above n 2, 665. The government follows that rationale and states in para 3.61 of EM 2001, above n 5, that the worldwide gearing rule “recognises that Australian businesses may need to borrow funds in order to expand offshore”.
[115] ITAA97, s 820-90.
[116] ITAA97, s 820-110.
[117] ITAA97, s 995-1(1).
[118] The amount is $(290/10 x 1.2)/(290/10 x 1.2 + 1) x 80, ie $77.8.
[119] Disallowed interest is $9 x (90 - 77.8)/90, ie $1.2. Deductible interest is therefore $9 - 1.2, ie $7.8.
[120] See the example in Pt 3.2(b) with “cost free debt capital” of $100.
[121] EM 2001, above n 5, para 3.3.
[122] ITA94, s FG 5(7).
[123] ITAA97, ss 820-85(3) and 820-185(3). This is of course subject to the exception of “cost free debt capital”, which as discussed in Pt 3.2, is wrongly included in “adjusted average debt”.
[124] Ralph Report, above n 2, 665.
[125] ITA94, s FG 3(b). Interestingly, the New Zealand thin capitalisation regime applies only to foreign multinational investors: ITA94, s FG 1.
[126] ITAA97, ss 820-90 and 820-190.
[127] EM 2001, above n 5, para 2.6. This policy basically follows Recommendation 22.2(b) of the Ralph Report, above n 2, 660.
[128] EM 2001, above n 5, para 10.4.
[129] ITAA97, s 820-105. The “independent Australian operation” assumption is also stated in paragraph 10.3 of EM 2001, above n 5. Such assumption is again confirmed in TR 2003/1, para 17.
[130] Bill 2003, Sch 1, Pt 11, items 36 and 38. Paragraph 1.82 of EM 2003, above n 30, states that the amendment is to ensure that the policy intention of “independent Australian operation” assumption is achieved.
[131] Interestingly, the government acknowledged in para 10.25 of EM 2001, above n 5, that “prudent commercial lenders usually look at the consolidated financial position of the group ... and the resources on which it could draw within that group to fund interest charges and capital repayments”. But it went on and claimed that “when applying the arm’s length test to a single entity, it is necessary to limit the extent to which the wider group is taken into account”.
[132] E Chang and R Vann, “Thin Capitalisation Reform in Australia: Another Milestone or a Millstone?” (17 December 2001) Tax Notes International 1209. For an example of the difficulty in satisfying the test, see ibid 1212.
[133] ITAA97, ss 820-105(4) and 820-215(4).
[134] ITAA97, ss 820-90 and 820-190.
[135] EM 2001, above n 5, para 10.31.
[136] For a brief discussion of this issue, see Chang and Vann, above n 132, 1211. Similar concern was raised by Collins who argued that the requirement to apply the test annually “does not accord with truly arm’s length debtor/creditor relationships”. See P Collins, “Thin Capitalisation Tax Reforms: Ready for the Next Tidal Wave?” (2001) 30 Australian Tax Review 106.
[137] See, eg, R Deutsch et al, Australian Tax Handbook (2003) 1562.
[138] TR 2003/1, para 53.
[139] Ibid.
[140] TR 2003/1, para 55.
[141] To get a feel of the pace of introduction of new tax legislation in recent time, it is interesting to note that the Minister for Revenue recently commented, in a proud way, that she had “personally steered through the Parliament in the little under two years during which [she had] had portfolio responsibility for tax design ... 26 bills specifically amending the tax law with a further 4 bills pending”: Minister for Revenue and Assistant Treasurer, “Taxation Reform and the Exploration Industry – Laying the Foundations for Future Success”, speech to the APPEA Tax and Finance Conference on 29 August 2003.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/JlATax/2004/5.html