If the system runs as designed, the net tax impact on each of the suppliers along the chain until the final consumer will be zero, since it can claim back whatever VAT/GST which it had to pay to obtain its supplies of the relevant goods and services. All these transactions will similarly be revenue neutral to the tax authority, since it refunds whatever output tax it collects from the suppliers when they make input tax claims.
A syndicate involved in MTF will generally defraud the tax authorities in one or both of the following ways: 1) by fraudulently making an input tax claim on purported supplies which it never paid for or received (and then absconding); or 2) by charging a purchaser output tax and absconding without handing over such tax to the tax authorities as required. Syndicates rely on the technical difficulties that the tax authorities face in tracing the supplies of goods and services down the supply chain and often make this more difficult by interposing many businesses along this chain. MTF is a serious problem for tax authorities particularly because it can involve defrauding the tax authorities not only of revenue which it should have received, but also of revenue which it should not have paid out in the first place, leading to significant revenue loss.
While it is clear that MTF is a criminal activity, it is often difficult for the tax authorities to go after the syndicates involved, given that they would have long absconded by the time investigations bring their fraud to light. The ones left holding the pieces are the other suppliers along the chain that may not have been involved in the fraud. Tax authorities often withhold or deny input tax claims of other suppliers along the chain once a fraud along the chain has been discovered.
In the Draft Goods and Services Tax (Amendment) Bill 2020 (the “Draft Bill”), Singapore proposes a new framework to deal with the problem of MTF. The approach is neatly summarised by a document released by the Singapore Ministry of Finance: “Annex: Proposed Changes to the Goods and Services Tax Act”, of which one point is of particular interest. The document states that the proposed legislative amendments will “allow the Comptroller of GST to deny a GST-registered business’ input GST claim, if the business knew or should have known that his purchase was part of or connected with a fraudulent arrangement. The burden of proving that the business knew or should have known of the fraudulent arrangement lies on the Comptroller, with the standard of proof being the balance of probabilities. This is similar to the approach taken in the United Kingdom (“UK”) and the European Union to safeguard tax revenue.”
Section 4 of the Draft Bill amends section 20 of the Goods and Services Tax Act (Cap 117A, 2005 Rev. Ed.) (“GSTA”) by inserting a new subsection (2A): “a taxable person is not entitled to credit for any input tax on any supply made to him which the taxable person knew or should have known to be a part of any arrangement to cause loss of public revenue (whether or not the loss was in fact caused).” The last part of the provision seems to adopt a different position from that under UK and EU law, since it does not require the tax authorities to show that loss to the public revenue was in fact caused.
However, the Draft Bill goes on to insert the following provision as well: “(2B) For the purpose of subsection (2A), an arrangement to cause loss of public revenue is an arrangement comprising 2 or more supplies (whether or not the supplies are in the same or different chains of supply), the effect of which is for one or more persons to evade or avoid paying any amount of tax, or to obtain any credit for or refund of tax which the person or persons would not otherwise be able to obtain.” (emphasis added) The phrase “the effect of which is” and not “the effect of which would be” may suggest that the effect must actually take place, in which case it is quite likely that there would be a loss of the public revenue anyway. Illustrations of these arrangement are also provided in a new proposed Ninth Schedule to the GSTA.
Also of interest is the new proposed subsections (2D)-(2F) which concern the test for whether a taxable person should have known that a supply made to the taxable person is part of an arrangement mentioned in subsection (2A) and may be summarised as follows. A taxpayer would be held to “should have known” if the circumstances connected with a supply made to it or by it, or both, carried a reasonable risk of the supply being part of such an arrangement, and it did not take every reasonable step to ascertain whether the supply was a part of such an arrangement. A taxpayer would also be held to “should have known” if such reasonable risk existed and either it 1) unreasonably concluded that the supply was not part of such arrangement; 2) was unable to so conclude; or 3) did not make any conclusion.
If a taxpayer did not take every reasonable step to ascertain whether the supply was part of such an arrangement, or took such steps, but failed to make any conclusion, then he would be held to “should have known” even if a reasonable person, having taken all the reasonable steps, would have concluded that the supply was not part of such an arrangement.
The statutory framework may be broken down as follows. The proposed legislation will apply an objective test to determine whether the circumstances of the supply are such that there was a reasonable risk of the supply being part of such an arrangement. If there is such a reasonable risk, a taxpayer must take all reasonable steps to determine whether the supply was part of such an arrangement and come to a reasonable conclusion that the supply was not part of such an arrangement. If it either does not take all reasonable steps, does not come to a reasonable conclusion, does not conclude that the supply was not part of such an arrangement, or does not come to any conclusion at all, then it will be taken to “should have known”. In such cases, it will still be taken to “should have known” even if a reasonable person, having taken all reasonable steps, would have concluded that the supply was not part of such an arrangement.
To simplify things even further, the effect of all the provisions read together is that the test for whether the taxpayer “should have known” is a wholly objective one with two main limbs. Once it is objectively ascertained that the circumstances of the supply are such that there was a reasonable risk of the supply being part of such an arrangement, the only way that the taxpayer can avoid being taken to “should have known” is if it shows that it took every reasonable step and came to an objectively reasonable conclusion. This position is indeed set out in subsection (2F) which clarifies the abovementioned positions. The proposed legislation then proceeds to set out in subsection (2G) (non-exhaustive) instances of the relevant circumstances of the supply.
One possible interpretation of “should have known” appears to be foreclosed by the proposed Singapore legislation. Consider the case of a taxpayer who did not take every reasonable step to ensure that the transaction it entered into was not connected to a fraudulent transaction, but in a case where a reasonable person would have concluded that it was not so connected. Certainly, it is possible to posit that a taxpayer cannot be said to “should have known” if a reasonable person would not have known after taking all reasonable steps, even if the taxpayer itself did not take those steps.
The proposed position in Singapore appears to exclude this possibility. As far as the proposed legislation is concerned, a failure on the part of the taxpayer to take all reasonable steps will affix the taxpayer with constructive knowledge, regardless of what a reasonable person would actually have concluded. Thus, a taxpayer would be particularly well-advised to ensure that it takes every reasonable step and scrutinises its transactions most carefully.