I want to start out by saying this definitely is not my area of expertise. I was really into tax law while at university, but I’ve never worked in tax. I’ve definitely never worked on anything related to VAT. This is all an academic problem to me. Please read this post with that in mind, and don’t hesitate to tell me if I got something wrong.
Missing trader intra-community (MTIC) fraud, such as ‘carousel fraud’, cost the UK Exchequer an estimated £2-3 billion in 2005-06. Today, HMRC estimates the cost to be between £0.5-1 billion per year, and the total cost is likely somewhat higher once we take into account the associated enforcement and monitoring costs.
A registered trader that imports goods from outside the EU is required to pay VAT at the border. However, there are no fiscal frontiers between EU member states. As such, a system of ‘deferred payment’ applies whereby imports are received VAT-free and VAT is only levied at the time of the importer’s next periodic VAT return. Carousel fraud exploits this system of deferred payment, as illustrated below.
In short, Company B is able to receive goods VAT-free, charge VAT on the sale of the goods, and issue valid VAT invoices entitling Company C to claim input VAT. Company B, however, can disappear without ever having paid VAT to HMRC. Company D is able to export the goods and claim a VAT refund despite the fact that Company B never paid VAT.
There have been a number of proposals for how VAT should treat intra-EU trade to address this problem. A first set of proposals seeks to maintain the current design of VAT but strengthen enforcement: tighter checks on firms seeking to register for VAT; requiring guarantees in the case of suspicious registration; adopting joint and several liability rules, which would allow traders to be held responsible if they should have been aware of the fraud; and establishing better information sharing between national tax authorities, which would allow the country of import to be made aware more quickly of exports to it that have not shown up in its own VAT system.
A second set of proposals alters the current design of VAT but maintains zero-rating of intra-EU exports: the use of reverse charging, which places VAT liability on the buyer rather than the seller; requiring traders to deposit the amount of VAT charged to customers into an account and making refunds contingent on the tax authority being able to verify the tax had been paid; and requiring a third party to guarantee VAT payments.
A third set of proposals fundamentally alters the current design of VAT by doing away with VAT zero-rating for trade between EU member states. One solution is export rating, which involves taxing intra-EU exports at the rate of the country from which the goods were exported with credit then being available in the country of import and some clearinghouse mechanism to ensure revenue ultimately accrues to the country in which consumption takes place.
An alternative solution is uniform rating, in which intra-EU exports would continue to be zero-rated for the national VAT in the country of export but would then be subject to a uniform VAT rate determined by the EU. This could either be in the form of a separate tax administered by an EU-level tax authority applicable to all exports (CVAT), or as a uniform rate applied to transactions between all VAT registered traders in the EU, with member states free to determine the rate of VAT on sales by traders to final consumers (VIVAT).
Many tax commentators are of the view that the problem exposed by MTIC fraud is implicit in the zero-rating of exports in the absence of frontier adjustments. In other words, the problem cannot be ‘solved’ without changing this aspect of VAT. I do not wish to unduly quarrel with this conclusion. However, dramatic administrative improvements could nonetheless be made to the current system.
One possible response to MTIC fraud, even prior to considering the set of proposals above, is to improve investigation and auditing. Crawford, Keen and Smith correctly identify the problem with this approach:
“Ex post audit and investigation, while important, is unlikely to forestall considerable loss of revenue, because the essence of the fraud is that money is made quickly, in the time gap before the missing trader is required to remit the VAT it has supposedly charged on its sales. Once the money has disappeared into the complex web of transactions, tracing and recovering unjustified VAT refunds becomes time consuming and costly.”
The complex web of transactions makes it difficult for HMRC to follow the paper trail back to the missing trader, e.g. from Company D to Company B. When Company D files for a VAT refund, the VAT invoice HMRC sees is from Company C. HMRC then has to look at Company C’s accounts to realize that Company B has not reported or paid its VAT liability. This paper trail, while widely regarded as an advantage of VAT in terms of enforcement, is still rather slow—too slow, in fact, to address carousel fraud with a great deal of success.
Tax authorities could require VAT payments to be made earlier than in the present system so that when a refund is paid it can be checked against past VAT payments. While it is true that this would close the time gap between when VAT is collected and remitted, this eliminates the cash flow benefit firms enjoy by retaining VAT until the next periodic payment becomes due. Furthermore, HMRC would still need to follow a paper trail back from the exporter to confirm whether the previous firms in the VAT chain have paid VAT.
It is important to realize that the missing trader (Company B) not only never pays VAT, but also never reports VAT. The missing trader issues VAT invoices and (supposedly) charges VAT, but never tells HMRC. HMRC only finds out about the missing trader’s VAT liability after following the paper trail back; HMRC is not given a complete picture of the “complex web of transactions” until long after the fact. In short, the problem with “ex post audit and investigation” is that it is ex post. Better information could make the detection of carousel fraud faster and more efficient.
If HMRC were aware of every VAT invoice at the moment it was issued there would be no need to follow the audit trail backwards. When the next periodic payment becomes due, HMRC would know immediately that Company B has failed to remit VAT. This is because HMRC would already have a record of how much VAT Company B had collected.
And if this instantaneous reporting of VAT invoices were electronic, as it surely would have to be, HMRC could even ‘work forward’ to detect likely cases of carousel fraud before Company B ever fails to make payment. HMRC’s computer system could monitor the ‘improved audit trail’ for suspicious cases. By going through past known cases of carousel fraud, HMRC could identify certain trading behavior that is characteristic of carousel fraud and HMRC’s computer system could flag similar cases as they arise.
But the current VAT audit trail is neither instantaneous nor electronic. Currently, VAT invoices are ad hoc, bespoke instruments. Registered traders create their own invoices, whether paper or electronic, containing the necessary information. There is no centralized system. This might have made sense decades ago, but appears anachronistic to anyone who has ever played a massively multiplayer online role-playing game (MMORPG) with a living economy, such as Runescape or World of Warcraft.
In these games, players can trade the game’s virtual items with one another directly; however, the trade occurs within the confines of the game itself. For example, a Dragon Hatchet is a virtual item in the game Runescape, and only exists as such within the confines of that game. To trade a Dragon Hatchet with another player, both players would need to log in to Runescrape and enter into an in-game trade; the trade only occurs on, and through, Runescrape. What the two players do not do is trade a Dragon Hatchet in real life and notify Runescape of their trade.
Yet this latter approach is how VAT invoices work. While VAT invoices only have the meaning HMRC ascribes to them, they are not issued through an HMRC system. Registered traders create and transfer VAT invoices between themselves, and then notify HMRC to bring those invoices within the VAT system. As a result, the VAT audit trail is neither instantaneous nor electronic.
What we want, therefore, is a closed, self-contained HMRC system for issuing VAT invoices. VAT invoices should be a product of the system; anything that is not issued through HMRC’s system would not be a VAT invoice. Each registered trader would have an online VAT account. To issue a VAT invoice, a seller would log in to its VAT account, fill out an HMRC invoice form, and send it directly to the buyer’s VAT account. The buyer’s VAT account would receive the VAT invoice, and the buyer would be notified of such.
There is no reason why the system would have to be complicated or cumbersome. It could operate like any other online service. HMRC could even create an online app so that the system is accessible even by smartphone. And the VAT invoice form would tell traders what information is required. This would make it intuitive for businesses to use, and would create a convenient data set for HMRC to analyze.
It is commonly said that VAT is self-enforcing in ensuring that the VAT invoice is issued for the correct amount, as the interests of the buyer and seller are adverse. But the system outlined above goes a step further, as the seller’s interest in receiving a VAT invoice for the correct amount also ensures that the buyer has reported the VAT liability to HMRC. While the buyer is still not concerned with whether the seller actually remits the VAT charged to HMRC, this system achieves the objective of making the VAT audit trail both instantaneous (at the point of issuing a VAT invoice) and electronic.
HMRC currently receives snapshots of the VAT system through periodic reporting and payment, and is able to plug the holes in this information by following the audit trail. But the VAT audit trail, a radical innovation in tax enforcement decades ago, appears slow and clumsy by today’s standards.
By bringing VAT invoices into a closed system created and operated by HMRC, we provide HMRC with real time data on all VAT invoices free of holes. The system contains 100% of all VAT invoices, as the concept of a VAT invoice only has meaning within the system. The system of issuing VAT invoices becomes the system of reporting VAT invoices. This surely is an improvement.