GST Avoidance Not a Matter of Scale

Education Administration Limited v CIR (2010) 24 NZTC 24,238

Education Administration Limited (“EAL”) was registered under the Goods and Services Tax Act 1985 (“GSTA”). Between September 2003 and September 2005 it claimed GST input tax credits, which the Commissioner subsequently disallowed on the basis that they were the product of a tax avoidance arrangement. EAL challenged the Commissioner’s assessments but lost in the High Court. This case emphasizes the difficulty of distinguishing between tax avoidance and legitimate tax planning which falls within the scheme and purpose of the GSTA. It also suggests some limits on the way in which GST input tax credits may be accessed and applied in business.

Broadly, the facts were as follows. A Mr Grove managed two businesses known as Kip McGrath Education Centres (“KMEC”), which provided after-school private tuition to students and were owned by Mr Grove’s mother. He raised problems he was experiencing with the Centres’ accounting system with his mother’s accountant, Mr Grimmett. Together, they discussed developing a dedicated business management computer program that could be tailored to the specific needs of KMECs. Mr Grove also saw possibilities for such a product in both the international market and other New Zealand businesses. In early 2003, they decided to attempt to develop such a program.

To do so, they decided to establish two separate companies. Mr Grimmett incorporated Administration Systems Development Ltd (“ASDL”) on 26 August 2003 and his company was responsible for the development of the software. Mr Grove’s company, EAL, was incorporated on 1 September 2003, and was to be responsible for marketing and selling the software once it was developed. However, the software was the property of Mr Grove’s company even during the development stage.

ASDL charged EAL a fee for the development of the software at an agreed hourly rate of $160 plus GST. ASDL invoiced EAL for the full cost of the development work performed each month but required only 10% of each invoice to be paid immediately. Payment of the remaining 90% was deferred and was to be made from future sales of the software. It was agreed that EAL would be registered for GST on a monthly invoice basis and ASDL would be registered on a six-monthly payments basis. The effect of this was that EAL was able to claim input tax credits on the total amount invoiced by ASDL and use the resulting GST refund from the Inland Revenue Department (“IRD”) to pay ASDL the agreed 10% of the fee. EAL did not have to account for any output tax because it had not itself made any supplies yet.

In total, between September 2003 and September 2005, EAL paid Mr Grimmett’s company the relatively modest sum of $27,032.81, having received invoices totalling $342,414.00. During that same period, EAL filed 21 GST returns claiming GST inputs of $378,334.99 and GST refunds of $42,037.11, of which Inland Revenue paid $33,637.11. In April 2006, the IRD issued a notice of proposed adjustment disallowing the input tax credits and requiring EAL to repay all the GST refunds it had received on the grounds of tax avoidance.

In the High Court the Commissioner submitted that this was a case of two non-arms-length taxpayers with a common financial goal manipulating the GSTA in order to obtain an unsecured interest-free venture capital loan. It was submitted that the arrangement was deliberately structured to generate GST refunds in advance of any supplies being made in order to capitalize an asset-less company so that it could fund the development of a speculative software package risk-free.

The taxpayer submitted that the arrangements were permissible tax planning within the scheme and purpose of the GSTA. It argued that the creation of two separate companies, and the registration of those companies on either the payments basis or invoice basis, are both valid choices under the GSTA and that there were sound commercial reasons for two the companies to be established. Moreover, the GST benefit was temporary and modest and must have been within the contemplation of Parliament.

EAL submitted that the general anti avoidance section in the GSTA should only be engaged where the mismatched registrations led to gross distortions brought about by extreme timing differences and gross inflation of consideration, falling outside the scheme and purpose of the GSTA, neither of which were present in this case. The taxpayer stressed that a core value upon which the granting of the GST input tax credit is based is that the party claiming it should have been subjected to a real and genuine economic burden and that this was indeed the case for EAL which was subject to a real obligation to pay the balance of the ASDL invoices issued to it.

Both sides accepted that there were certain general principles that applied in the case:

a) The GSTA is predicated on the basis of an overall balance being achieved between the outputs and inputs of a registered person.

b) The wider the temporal gap between the eligibility of a taxpayer for an input tax credit and its liability for output tax, the less likely the arrangement conforms with the intent of the GSTA. A significant delay can indicate a crossing of the line into tax avoidance.

c) The GSTA permits accounting for GST on three different bases: the invoice basis, payments basis and hybrid basis. It therefore contemplates the possibility of there being transactions between parties using different accounting regimes and hence a degree of mismatch between the time at which input tax credit is claimed and the time at which output tax is paid in relation to the same supply.

d) However, while the GSTA permits a degree of mismatching, it seeks to limit the nature and degree of such mismatching. A gross mismatch in timing may accordingly be relevant in assessing the application of the anti avoidance section.

Justice French held that, assessed objectively, the arrangement in this case had tax avoidance as a more than merely incidental purpose or effect and had been structured in a way that could not have been contemplated by Parliament. She found that a number of factors, taken together, meant this case crossed the line into tax avoidance. Those factors were as follows:

a) Two separate companies were created when the different objectives could have been accommodated within a single company.

b) EAL was claiming GST refunds on amounts that might never become payable. Even on the best view of it, there was at least a two or three year gap between when the expenditure was incurred by EAL and when it was expected to be paid to ASDL.

c) The hourly rate of $160 plus GST charged by ASDL was held to be higher than a market rate. The effect of this was to artificially increase the amount of the invoices and hence the amount of the GST refund that could be claimed.

d) During the periods in question, EAL had no independent source of funds. Its sole source of funding was through the receipt of the GST refunds. The fact that EAL had no capital of its own was found not in itself to be sufficient to make the arrangement avoidance. However, when taken in combination with other factors, it assumed some significance.

There are several observations worth making about this case. First, the factual findings of the Court warrant examination. The conclusion that two companies need not have been used was, with respect, against the weight of evidence. Moreover there is nothing in tax law that says that taxpayers should use a particular structure (say a joint venture company) over others. In this case the finding about separate companies was entirely beside the points which were fundamentally at issue and was a red herring deployed by the Commissioner and swallowed by the Judge.

Similarly the hourly rate finding is highly debatable both as to correctness and relevance. In a case such as this the marginal relevance of a rate at $160 compared with say $120 was slight and need not have been part of the Court’s analysis at all. It is troubling that a Court would allow itself to be drawn into this sort of marginal analysis compared with those cases where there is undoubtedly a gross over (or under) valuation of goods and services. This will only encourage the Commissioner to second guess the ordinary run of charges for goods and services rather than save his poweder for cases of gross distortion.
 
The real issues in the case were the question of EAL’s commitment to pay the balance of the ASDL invoices it received, and the fact that it had no resources itself. On the first of these, the argument was lost essentially because the arrangement for payment of the balance of the ASDL invoices was too open ended. The evidence for the taxpayer was that commercial reality meant that both sides were committed to bringing the computer program to market as soon as possible and had been prevented from doing so by the long running dispute with the IRD. They argued there was an acknowledged debt from EAL to ASDL which amounted to a present but deferred obligation to pay. The Court did not accept that. It concluded that arrangements for the marketing and exploitation of the program (and payment of the balance of the invoices) were too remote. It is interesting to consider what might have been the situation if the balance of the invoices had been payable at a fixed time as the Commissioner argued.

As a passing comment it is pertinent to note that the Court seems to have been quite unsympathetic to the argment that the tax investigation had intruded on the plans to bring the computer programe to market. This is a timely reminder that taxpayers need to press on with their plans if progress is consistent with their case.
 
On the second issue, the Court really did not articulate the policy grounds which could have unseated the EAL argument. It accepted that the use of GST refunds as working capital can be legitimate and said that the fact that EAL had no capital of its own was not enough in itself to lead to a finding of avoidance. It then added into its analysis the irrelevant matters referred to above as if to make the argument over the absence of capital more significant. It need not have done so.

The underlying problem with EAL’s case was that its arguments over the legitimate use of tax credits could only go so far. Taxpayers are certainly permitted to use and apply GST refunds in their businesses and there is no basis for arguing that GST credits should somehow be separated from other revenues. That said, however, GST is normally a tax which follows the application by a taxpayer of its own resources. In other words input tax credits follow the disposition by a taxpayer of its resources to acquire goods and services, rather than the credits being the resource used to make the acquisition in the first place. There may be instances where the GST credit emerges before the application of capital, but these will normally be minor and temporary. In this case, however, instead of the GST input tax credit being incidental to a commercial acquisition by EAL of services from ASDL, obtaining the credit became an end in itself because EAL had no other resources.

There is tension between accepting on the one hand that GST credits may be used as working capital, and saying on the other that they cannot be the sole source of such capital over a sustained period of time, but the two propositions are not mutually exclusive. The Court does not deal with the case expressly in that way, but it is one basis on which the decision can be (and perhaps should have been) explained.

It is sometimes helpful to test an allegation of tax avoidance by asking what tax “norm” has been avoided by the alleged arrangement. In this case the norm would have been that GST input tax follows the genuine assumption of a financial cost for goods and services and the application towards that cost of business resources including but not exclusively GST input credits. Adopting a slightly different approach, that norm might safely be taken as within Parliamentary contemplation, and the EAL arrangement fell outside it.

Geoff Clews was leading Counsel for the taxpayer in the EAL case.