VAT Fraud and Terrorist Funding - the Azizi Extradition Allegations Part 2

VAT FRAUD AND TERRORIST FUNDING – THE AZIZI EXTRADITION ALLEGATIONS PART II VAT FRAUD AND TERRORIST FUNDING – THE AZIZI EXTRADITION ALLEGATIONS PART II This paper considers the remaining seven (7) Missing Trader Intra-Community (MTIC) fraud schemes alleged (some conceded) to have been conducted by Samir Azizi, a 25 year old German/Afghan citizen, who was extradited from the United States to Germany on April 14, 2015.1 This is a two-part paper. Part I considered the first four (4) of eleven (11) MTIC schemes in the Complaint. It examined 26 of the 89 criminal counts, representing alleged VAT losses of €15,302,965 out of €61,104,368 in VAT revenues purportedly denied the German Treasury by Azizi. The MTIC fraud schemes considered in this Part II involve the seven (7) additional MTIC schemes and the alleged losses of €45,801,403 under 63 further criminal counts. Underlying both Parts of this assessment is a fundamental (but unanswered) question: “Who exactly is Samir Azizi? Is he the mastermind of a multi-million euro VAT fraud, a fundraising fraudster for terrorist organizations, or a youthful face-ofconvenience disguising the involvement of larger criminal organizations?” The terrorist connection is the most troubling aspect of this case. Judge Howard R. Lloyd in his Extradition Order states that “… there were also indicators that perpetrators were using the VAT procured through such fraud, not only for personal enrichment, but also to finance terrorism.”2 This terrorism reference is not further supported by third-party evidence rather it is supported only by the sworn statement of the German prosecutor that in turn (allegedly) rests on facts not presented in court.3 The prosecutor’s allegation is in the Formal Request for Extradition (FRE). That document states: There are numerous indications here that the cash flows are used to finance terrorism.4 There is a nuanced difference between Judge Lloyd’s statement that “the VAT procured through such fraud” (emphasis added) financed terrorism, and the FRE’s statement that “the cash flows are used to finance terrorism.” (emphasis added) Judge Lloyd appears to have been shown “something more.” What the FRE is saying is general. It observes that there are large sums of cash flowing through the Azizi fraud chains, and some of this volume assists terrorists. What 1 In the Matter of: the Extradition of Samir Azizi, Order Granting Motion for Certificate of Extraditability 5:14-xr-90282-PSG (March 20, 2015) District Court, N. D. CA, Doc. 60. 2 Extradition Order, Doc. 60, at 22, referencing the Formal Request for Extradition. 3 The author contact the Azizi defense team to make sure that there was nothing else available in public records on the terrorist elements of this extradition. Attorney Doug Rappaport assured the author that there was nothing had been overlooked. (Telephone conversation of July 17, 2015). 4 In the Matter of: the Extradition of Samir Azizi, Formal Request for Extradition, 5:14-xr-90282-PSG (official English translation) at 11 (hereinafter referenced as RFE). 1 is the nature of this assistance? It is not clear. Maybe funds are diverted to terrorists, or maybe terrorist money is laundered by the frauds. We are not told specifics in the FRE. Judge Lloyd goes further. He indicates that he has been persuaded that the cash flows include a very large amount of un-remitted (or inappropriately refunded) German VAT. Judge Lloyd slices off this VAT from the other cash flows and declares that it is the stolen VAT that is supporting terrorism. If we take the extradition evidence as a whole, the terrorist-funding allegation is not that Azizi was engaged in a simple skimming fraud where Azizi would (for example): (a) take a fraudulent refund of €100 from the German Treasury, (b) put €20 of the €100 in his pocket for himself, and then (c) send the remaining €80 off to a terrorist group. There is nothing like this in the extradition files. Instead, the facts and fiscal dynamics of these frauds sketch out a cash flow MTIC fraud that uses payment platforms5 to direct a predetermined “cut” of the “profits” to “workers” and to “investors.” The workers do not appear to be terrorists, but the investors may be a different story. The investors are the people who have put up tens of millions of euro to prime the pumps of this fraudulent enterprise, and these individuals could well have other interests. If there are payments made to terrorists through the MTIC chain they will be predetermined by the design of the payment platform. It is the person who controls the payment platform algorithm who is the person that could be directing cash to terrorists. This is not Azizi. The design of the payment platform is never something that Azizi is alleged to have been involved with. There is no suggestion that Azizi thought-up, arranged, or in any way controlled the payment platform. Its design is indeed something that Aziz is aware of, but it is not Azizi’s brainchild. Azizi is presented as someone who joins a pre-planned fraud chain with an established payment platform already in place, not as someone who masterminded MTIC schemes. PAYMENT PLATFORMS & CASH FLOW MTIC FRAUD A cash flow MTIC fraud utilizing a payment platform is illustrated below: Figure 1: Payment Platform 5 A payment platform is commonly a bank account at an established banking institution where a number of fraudsters hold sub-accounts. Funds are transferred among the sub-accounts as loans are established and payments are made among the fraudsters without ever being visible to the banking institution, or outside regulators. Marius-Christian Frunza, Fraud and Carbon Markets: The Carbon Connection (2013) at 49. Center for the Study of Democracy, University of Trento, Teeside University, Financing Of Organized Crime at 301-404 (discussing the early payment platform the First Curacao International Bank (FCIB) which existed from 2004 through 2007). 2 The key to understanding this variant of MTIC fraud is to carefully consider the fraud opportunities that are inherent in the cash flow mechanism. All payments for supplies are made through loans. The loans are paid off in a rapid succession as funds cascade through the system in a reverse direction from the flow of supplies. Loan obligations, payoffs, and all excess cash remain in the payment platform – a collective bank account physically located outside the jurisdiction of the fraud. To see the payment platform in operation read the above diagram in reverse order (going from G through to A & B). The natural flow of goods or services is moving in the other direction (from A & B through to G). In this example assume that G (a business in another EU Member State, not Germany) purchases goods from F (a business in Germany) for €1010. F is called the “distributor” in the fraud chain. F collects no VAT from G, because this is an intracommunity supply. The funds received by F are “moved” in the Payment Platform (a shared account in a legitimate bank where both G and F have sub-accounts). In other words G does not pay F by presenting cash, a check, or a credit card. Payment is made with an accounting entry that removes funds from G’s sub-account in the payment platform into F’s sub account in the same payment platform. Importantly, the funds are already in play before the transaction begins. This is not a case where a withdrawal is 3 made from some other institution, and an infusion of new funds show up in F’s account. Functionally this is an “internal direct deposit.” It is a direct deposit that does not change the overall balance of the master account (the payment platform) in the selected bank. The payment only changes which party among all the participants in the payment platform has claim to these funds. This transaction is invisible to the traditional banking system. Because F had purchased the goods from E for €1,000, F makes a €10 profit on the sale. F files for a VAT refund of €190 from the German Treasury (it has an input credit of €190 from its purchase from E and an output VAT amount of €0). F would never have paid cash for the goods from E. In a cash-flow platformbased MTIC chain F would have purchased from E on E’s grant of credit. An interest free loan from E for €1,190 would support the entire purchase. €1,190 is the full price of the goods (€1,000), plus the VAT due (€190) at the German rate of 19%. The VAT refund of €190 is deposited in the payment platform account. It is the only addition of new funds into the account in the entire series of transactions in the MTIC chain. F’s receipt of funds from G allows it to pay off the loan from E in full. Payment is again made through the payment platform where E is also a shared account holder. E is similarly burdened with a debt from D for the goods it purchased and then re-sold to F. The debt is for the goods it acquired (€997), plus the VAT due on this purchase (€997 x .19 = €189.43). E’s profit on this sale is €3 (€1,000 - €997 = €3). E files a German VAT return, remitting €0.57 (output VAT collected from F of €190, less input VAT paid to D of €189.43 = €0.57). By constructing a transaction chain where the buffers incur a small positive VAT obligation the fraudster hope to escape the attention of the authorities. E’s payment of the debt incurred from D allows D to similarly pay off its debt incurred when it purchased the same goods from C. C would have extended full credit to D at no interest for the entire cost of €1,182.86. This amount is the cost of the goods (€994) plus VAT of €188.86 (€994 x 0.19 = €188.86). D would report profit on the sale of €3 (€997 - €994 = €3), and it would remit €0.57 to the German Treasury (output VAT collected from E of €189.43, less input VAT paid to C of €188.86 = €0.57). D’s payment of the debt incurred from C allows C to similarly pay off its debt incurred when it purchased the same goods from the missing traders A and B. A and B would have both extended full credit to C at no interest for €589.05 (each), or €1,178.10 in total. This debt amount would cover the cost of the goods at €495 each, plus VAT of €94.05 each (€495 x 0.19 = €94.05). The aggregate figures for C would be for purchases of €990, and VAT of €188.10. C therefore, would report profit on the re-sale of €4 (€994 - €990 = €4), and it would remit €0.76 in VAT to the German Treasury (output VAT collected from D of €188.86, less input VAT paid to A and B of €188.10 = €0.76). 4 In this example the German Treasury receives aggregate VAT payments of €1.90, but it refunds VAT of €190. Each of the buffers (C, D, and E) file returns and pay VAT. The distributor (F) files for a significant refund. The missing traders A and B file no returns, and do not perform a reverse charge. The sub-accounts of the missing traders are where the bulk of the “profit” from this fraud resides. This closed system only has two revenue sources. The €1,010 paid by G, which is the “perpetual investment” in this MTIC chain by the fraudsters, and €190, which is the VAT refund paid by the German Treasury after receiving the return filed by the distributor (F). Inflows are €1,200. There are only two outflows. €1.90 in VAT paid by the buffers to the German Treasury, and the compensation due to the formal managers of the entities involved in the fraud chain. According to Azizi the compensation metric he was familiar with ran at 0.1% to 0.3% of sales.6 In this example this roughly approximates the “profit on sales” figure of €24 (the sum of €10, €3, €3, €4, €2 and €2 for F, E, D, C, B, and A respectively). Thus, the “net profit” after all expenses and investment in this example is €164.10. All of this money resides in the sub-accounts of the missing traders in this example.7 The speed of payments through this chain can be exceptionally fast. The money movement is a simple exercise in debiting and crediting sub-accounts contained within the master account of the payment platform. It can be fully automated. Once G’s €1,010 payment is made to F the funds cascade through the chain paying off one debt after another. If the “master mind” decides not to “roll over” the “net profits,” that is, not reinvest in the next turn of the carousel, payments will be made out to the external “investors” in the fraud (the people who put up the €1,010 that G used to pay F through the platform) to compensate them for the use of their funds (and the risk of loss in case of an enforcement action). Or, as Azizi indicates, the payment platform can be designed to both rollover a portion of the profits and distribute the rest according an algorithm. TERRORIST FUNDING Using only the evidence presented in the Azizi extradition case, it is difficult, if not impossible, to connect a specific MTIC fraud (from among the eleven schemes alleged to have been orchestrated by Aziz) to a specific instance of terrorist funding. The FRE is very thinly developed on this point. However, given the importance of the VAT fraud/ terrorist funding question, a guess at making a link is warranted. If there is terrorist funding in the Azizi extradition facts, where could we expect to see it? The strongest candidates are the iTrading and iCell schemes. Both of these presentations convey the impression that there is something 6 FRE at 55 (indicating that he received 0.1% for the buffers in iTrading, and between 0.1% and 0.3% if he participated as a missing trader). 7 Note: missing trader A and B have account balances of €589.05 each or €1,178.10 all together, which represents the full investment of €1,010, the separate transaction profit earned by A and B of €2 each, as well as the overall net profit from the fraud of €164.10. 5 more to tell. In both of these schemes the German prosecutor goes out of his way to present direct, confessional statements by Azizi that explain far more than is necessary to support the FRE. Why is the prosecutor doing this? In iTrading there is an extended discussion of the Deutsche Bank involvement in CO2 fraud that is truly eye opening, and in iCell there is a detailed discussion of payment platforms that makes the reader pause. In both instances the FRE conveys the impression that the prosecutor is holding back, that he is laying the groundwork for something that he may or may not need. Could it be a terrorist trump card; an argument that may not be needed, but if Azizi’s extradition was ever in question, then this trump could be played at a moment’s notice? In iTrading we are told a lot about Deutsche Bank’s pivotal role in CO2 fraud. We are told much more than is necessary. Azizi is not an employee or otherwise connected to Deutsche Bank. Why are we being presented these details? Similarly, in iCell we learn much more than we need to know about payment platforms and how they were used in this scheme to divide the “profits” among the “investors.” Azizi is not an investor in any of the alleged frauds; he is a worker and an arranger of companies. In fact, Azizi is rather short of funds when he is setting up his buffer companies. He is hardly the mastermind behind any of the alleged frauds. He is a worker. He has no access to “investor funds” of the magnitude needed to start a major MTIC fraud chain. What ultimately happens with the proceeds of the fraud is extraneous to the task at hand – the extradition of Azizi. But there is more. The FRE refuses to draw a line between the Deutsche Bank and the payment platforms. There is not much more to a payment platform than the bank where the master account resides, so one cannot help but ask: • Were the master accounts of any of the payment platforms placed with the Deutsche Bank? With a stable account balance reaching many tens of millions of euro one would think the location of these funds would be of interest to Deutsche Bank. • Did Deutsche Bank have any knowledge of the fraud itself? • If so, did the Deutsche Bank have any understanding, or control over where the payment platforms were sending the “profits” of the fraud; profits which came in large measure from trades that Deutsche Bank (Frankfurt) completed with Deutsche Bank (London)? • Did Deutsche Bank know or have reason to know that the profits of the fraud were being directed to terrorists (as the FRE and Judge Lloyd indicate was probably happening)? The following passage from Azizi’s confession, placed in the FRE by the German prosecutor, describe the operation of the payment platforms, and outline the distribution of funds. Azizi describes huge sums of money going to relatively obscure “investors.” Azizi treats iTrading and iCell as twin frauds. One deals in CO2 permits, the other in cell phones, but to the fraudsters the schemes are otherwise indistinguishable. Azizi explains: 6 Every company participating in the fraud received a so-called Payment Batch Report for each payment from the corresponding platform or from the bank where the platform was held. There it was apparent what happened with the money. These Batch Reports were hidden and should still to be found on the computer of the company iTrading/iCell. For example it could be seen, when the company iTrading paid €4,136,958.00 to the company Global Reach [Payment Platform], that from this amount four times 500,000 USD, and once 600,000 USD and an amount of 1,536,958.00 GBP were transferred from the platform. You could also see where the money was going. Therefore the fraud must be quickly understood, as e.g., these 1,536,958.00 GBP were paid to the company SVS Securities PLC, although this company was not a supplier of the companies iTrading or iCell. All the other amounts then went to the companies of the men pulling the strings. All amounts transferred in USD were sharing out of the loot and the amounts in GBP or EUR were amounts that flowed back into the fraud chain. (emphasis added)8 If there is a named terrorist, or terrorist organization being funding thorough this fraud it is not disclosed by Azizi. However, the German prosecutor’s case makes it easy to imagine that a terrorist with money might easily become an investor. One could imagine that one or more of the four 500,000 USD allotments, or the 600,000 USD amount discussed by Azizi could be money destined for terrorists. There is nothing to confirm or deny any of this. We are engaging in pure speculation. But it is speculation encouraged by the FRE’s assertion that there are “numerous indications” that the cash flow “is used to finance terrorism.” Azizi puts some names into the mix, but there is no follow-up in the extradition documentation: Companies that received the USD were obviously direct companies of the men pulling the strings and they are not disguised any more. The companies Unique GmbH, ACI GmbH, Rapid Link GmbH and FEG were demonstrably further fraud companies. The person responsible, inter alia, for the company Rapid Link was the “thin” Ali, he was a good friend of another man pulling the strings, Umesh Salvi.9 This paper will follow the seven remaining Azizi cases pointing out the important interplay with payment platforms, the apparent instances of double-counting in the assessments, and Azzi’s ability to rapidly substitute one company for another in a fraud chain to obfuscate the fraud. Azizi appears to be very good at managing to set up buffers and put them in place to make the fraud chains work. The seven remaining cases in the Azizi extradition are: (1) iCell GmbH & Co. KG – denial of €12,573,141 of input VAT deductions; using Global Reach Partners payment platform held at Lloyds TSB Bank in London; 8 FRE at 37 (note: Azizi appear to have oversimplified the details in his example, as it is unlikely that USD and GBP were trading at one-to-one at any time during the alleged fraud). 9 FRE at 37. 7 (2) AS Handels, GmbH – an assessment of €7,270,741, based on €3,335,383 in fraudulent output invoices, €2,430,601 of denied input VAT, and an attempted (refund) fraud of €1,504,757; (3) Nexo Chakfa e.K./ Nexo Chakfa GmbH – an assessment of €5,260,446, including €4,592,936 attributed to denial of input credits, and €667,510 assessed on output; using Global Reach Partners payment platform held at Lloyds TSB Bank in London; (4) Hamsterecke.de GmbH/ Hamster Mobile GmbH – an assessment of €6,302,224, including €1,224,650 of denied input VAT, and €5,077,574 attributed to fraudulent output VAT placed on sales invoices. (5) Amaan Enterprise GmbH/ Mobiltronics GmbH – an assessment of €10,308,708 attributed entirely to the denial of input VAT deductions; (6) BAK Enterprise, GmbH – an assessment of €3,565,554 based entirely on output VAT placed on sales invoices; (7) My iCell, GmbH – an assessment of €520,589, based entirely on the denial of input VAT deductions; using the Omnis Capital FX Ltd. payment platform in London, UK. (1) iCell GmbH & Co. KG €12,573,141 The €12,573,141 assessment against iCell GmbH is the largest aggregate assessment in the extradition Complaint. With 14 distinct counts (cases 27 through 41) iCell also leads in the number of discrete offenses. The iCell part of the Complaint looks at a group of companies that Samir Azizi controlled along with his cousins from the Soori family (notably Habib Soori). iCell dealt in cell phones. One of the major iCell fraud chains is illustrated below: Figure 2: iCell GmbH & Co. KG 8 Azizi controls five of the six companies engaged in this fraud. He controls both missing traders (Sabs Euro Trading GmbH and I.I. First Euro Trading GmbH), as well as all three buffers (Nexo Chakfa GmbH, iCell GmbH and Techstage GmbH).10 More than this, Azizi also controlled the operation of the chain. There are instances where Techstage could not make purchases from iCell, and Azizi rotated another company he controlled (A.S. Handels GmbH) into place to provide the needed supplies.11 The sales of cell phones along the iCell chain are largely artificial (“…these purchase invoices were not based on any actual supplies or services.”)12 The assessment in this segment is determined through denial of input credits on iCell’s purchases. No part of the assessment is concerned with invoices issued by iCell on its sales even though German law would attribute liability to iCell for fraudulent invoices on which it had placed the VAT amount.13 The largest number of purchases (44%) are from Nexo Chakfa (€5,522,089 in VAT deducted),14 as highlighted in the diagram above. The next two largest blocks are purchases covering an additional 9% from Hamsterecke.de GmbH/ Hamster Mobile GmbH (€1,186,662 in VAT deducted),15 and 30% from AS Handels GmbH (€3,765,857 in VAT deducted).16 There are “… further input tax amounts from several smaller invoices that are not to be accepted as neither the issuer of the invoices nor the recipient of the invoice are undertakings in the meaning of the VAT Act.”17 (2) AS Handels, GmbH €7,270,741 This entity was registered as a shelf company (AKMA Export Import GmbH on August 20 2003). It was bought by Azizi and re-named AS Cellectric GmbH on January 19, 2010, and then again on May 20, 2010 it was re-named AS Handels GmbH with Hamid Soori as the formal managing director (although the company was in fact run by Samir Azizi).18 In the first month of active trading, (when the firm was still named AS Cellectric) turnover exceeded €3 million. Purchases largely came from Sabs Euro Trading GmbH;19 sales were largely made to iCell, GmbH. Azizi controlled all three of these firms, so effectively Azizi was buying and selling to himself. In the first quarter AS Handels allegedly caused “VAT damage” of €3,335,383 though its sales (to iCell). It collected VAT without remitting the tax. In this quarter the VAT damage was the result of “… 10 FRE at 64-65. FRE at 64. 12 FRE at 65. 13 VAT DIRECTIVE, Art. 203; s.14c para.1 UStG; Finanzamt Osnabrück-Land v. Bernhard Langhorst Case C-141/96. 14 The Hagen tax investigation found fraudulent transactions. FRE at 11i 15 The Leipzig Tax Office and Dusseldorf Tax Investigation found fraudulent transactions. Id., at 12 16 The Dusseldorf Tax Investigation found fraudulent transactions. FRE at 210-11 17 FRE at 67. 18 FRE at 68-69. 19 FRE at 57 & 60. Sabs Euro Trading GmbH was determined to be a missing trader by the Düsseldorf Finance Court (Case no 5 V 3555/10 A (H(U)). The Munich Tax Investigation documented this 11 9 unlawful non-submission of VAT returns … the assessment of the damage incurred and resulting from VAT evasion based on sales invoices.”20 The iCell/ AS Handel relationship is more complicated than this. In the first quarter of 2010 AS Handles sold large volumes to iCell, but by April 2010 it began buying from iCell.21 In this case, the input VAT deductions on these purchases are denied. In addition to the missing trader Sabs Euro Trading and some unidentified other traders, AS Handles’ primary suppliers after the first quarter of 2010 were: (a) Adisony GmbH, which a Berlin Tax Investigation determined to be a buffer in a VAT fraud chain, and (b) Amaan Enterprise GmbH, which the Düsseldorf Finance Court found to be a nonexistent company created to facilitate VAT fraud. With respect to the Amaan supplies, AS Handles evidently applied for a VAT refund of €1,504,757, but it was not approved by the Tax Office. The Amaan transactions are classified as a fraud attempt, not a completed fraud. The diagram below presents the alleged fraudulent transactions entered into by AS Handles in the second, third and fourth quarters of 2010. Figure 3: AS Handles, GmbH (3) Nexo Chakfa e.K./ Nexo Chakfa GmbH 20 FRE at 71 (emphasis added). It is very likely that there is double-counting of VAT losses here. During the same tax period (the first quarter of 2010) where AS Handels’ VAT losses on sales invoices are €3,335,383 there are VAT losses determined for iCell based on purchase invoices of €6,675,700. 21 FRE at 64. 10 €5,260,446 There are 11 cases involving Nexo Chakfa e.K./ Nexo Chakfa GmbH (Nexo Chakfa), ten concern improper deduction of the input VAT credit (€4,592,936), and one involves the non-submission of a VAT return (€667,510). Nexo Chakfa goes through several name changes, but eventually ends up as a buffer controlled by Aziz set between other buffers also controlled by Azizi. Nexo Chakfa is in the cell phone trade. It did not have an office, did not record the IMEI number22 for each phone bought and sold, and the phones that were involved never physically arrived at the company’s registered office. The input credits are denied from purchases made from: • Sabs Euro Trading for €1,3306,432 • I.I. First Euro Trading for €2,076,855 • Hamsterecke.de for €18,715 • Microtech for €108,082 • Amaan Enterprise for €828,637 Figure 4: Nexo Chakfa e.K./ Nexo Chakfa GmbH The Nexo Chakfa cases also provide a limited glimpse at the operation of a payment platform (as described in Figure 1 above). There are huge amounts loaned among the entities in Figure 4 above, often without any immediate connection to actual deliveries of supplies. Azizi controls all of the entities. The following should be noted: 22 The International Mobile Station Equipment Identity (IMEI) is a number, usually unique, used to identify 3GPP and iDEN mobile phones, as well as some satellite phones. It is usually found printed inside the battery compartment of the phone, but can be displayed on-screen as well. 11 • • Both Sabs Euro Trading, GmbH and I.I. First Euro Trading are proven missing traders;23 • Between January 22, 2010 and February 28, 2010 Sabs Euro Trading o issued invoices for € 2.7 million, and o received payments of € 130,000.24 • Between January 7, 2010 and January 11, 2010 I. I. First Euro Trading o made deliveries on credit in excess of € 1.3 million but, o before those deliveries were paid for an additional set of invoices were issued (January 13, 2010) for €4 million, and eventually o by March 2010 I.I. First Euro Trading had extended credit between € 3 million and € 5.6 million.25 • Nexo Chakfa extended millions of euro in credit to iCell26 When payments are finally made (at the end of the chain) the funds are transferred to Sabs Euro Trading and I.I. First Euro Trading directly through an account shared with the other entities in Global Reach Partners, the master account for which was held at Lloyds TSB Bank in London.27 (4) Hamsterecke.de GmbH/ Hamster Mobile GmbH €6,302,224 There are 10 offences charged in the Hamsterecke.de GmbH/ Hamster Mobile GmbH (Hamster) counts, seven concern improper deduction of the input VAT credit resulting in a €1,224,650 assessment, and three involve the non-submission of VAT returns that result in a €5,077,574 liability. German authorities linked Azizi to Hamster through e-mails, chat protocols that Azizi engaged in with Selaiman Azizi and Helai Azizi. It was clear through these conversations that Samir Azizi controlled contracting, customers, prices and payments for Hamster. The formal managing directors of Hamster at various times were: Fred Wenzek, Shiraz Jilani, Richard Mintus, Andreas Feiherr von Rössing. Each of these formal managers have been (or are in the process of being) prosecuted. Figure 5: Hamsterecke.de GmbH/ Hamster Mobile GmbH . 23 München Tax Investigation found that Sabs Euro Trading had no real business activity, and the Frankfurt Tax Investigation found the same about I.I. First Euro Trading. FRE at 72-73. 24 FRE at 73. 25 FRE at 73. 26 FRE at 72. 27 FRE at 72. 12 Hamster bought substantially quantities of cell phones from Sabs Euro Trading. Its trading pattern substantially replicated that of Nexo Chakfa discussed in the immediately prior materials. In the Nexo Chakfa facts the German government traced sales: (a) from Sabs Euro through Nexo Chakfa and then on to iCell, as well as (b) from Hamster through Nexo Chakfa and then on to iCell. In this collection of transactions the German government is focusing on another substantial thread where sales go from Sabs Euro through Hamster and then on to iCell. Invoices from Sabs Euro Trading including a VAT amount of €1,082,349 are rejected outright.28 However, in the tax periods ending on September 11, 2010, October 11, 2010, and November 11, 2010 (when no return was filed) the liability was determined on the output VAT. The amount placed on invoices (primarily sales to iCell) of € 715,455, €3,575,311, and € 786,808 that should have been recorded on their respective returns become the basis of the assessment.29 Because the assessment in all the iCell offenses is based on a denial of the input VAT on purchase invoices, and because some of those invoices came from Hamster, it is highly likely that the same invoice (and VAT amount) is being counted twice.30 (5) Amaan Enterprise GmbH/ Mobiltronics GmbH €10,308,708 28 FRE at 60, 75 & 80. FRE at 80-81. 30 For example, when Hamster fails to submit the August (€715,455), September (€3,575,311), and October (€786,808) on September 11, October 11 and November 11 in 2010, iCell was also being assessed because the tax authority was denying iCell’s input credit on its August (€329,231), September (€740,942) and October 2010 (€289,758) returns due on September 9, October 11, and November returns. A note of caution is appropriate. The extradition documents are not sufficiently detailed to draw more than a strong suspicion of double counting. 29 13 Azizi is charged with sixteen tax fraud offenses though his de facto management of Amaan Enterprise, which wrongly claimed input credit from suppliers knowing that the invoices were fraudulent. The “supplies” were not based on actual supplies of goods or services. As with Hamster (above) e-mails and chat protocols show that Azizi established customer contacts, set prices, controlled payment transactions, and presented himself to third-parties as the company’s managing director.31 Figure 6: Amaan Enterprise GmbH/ Mobiltronics GmbH The Amaan facts are thinner than many of the other extradition fact patterns. Other than the fact that Azizi controls Amaan Enterprise, we are really only told that Amaan Enterprise is a buffer that engages in cell phone and CO2 trade. The three entities it purchases from have either been proven to be missing traders, or other buffers in a MTIC fraud chain. In each case Azizi knows that he is participating in a fraudulent scheme when he makes purchases through Amaan Enterprise. As a result the tax administration has denied the following input credits: • €3,284,187 is the input tax on purchases from Sabs Euro Trading, a known missing trader;32 • €4,582,109 is the input tax on purchases from Adisony GmbH, a company that the Berlin Tax Investigation determined was a buffer in a MTIC fraud chain;33 31 FRE at 83. RFE at 60,75, 80, & 85. The Munich Tax Investigation determined that Sabs Euro Trading was a missing trader. 33 RFE at 85, 32 14 • €2,872,769 is the input tax on purchases from Fashion Textile GmbH, a company that the Offenback Tax Office determined to be a missing trader, operating out of a fictitious address without any economic activity. (6) BAK Enterprise, GmbH €3,565,554 There are apparently strategic (litigation) choices made in fraud selections for this extradition request. BAK Enterprise is a good example of this. The tax investigation set out in the RFE makes it appear that in this case the government did not find more than: • the fact that Azizi was the de facto manager of BAK, and • the fact that BAK did not file VAT returns in the third and fourth quarter of 2011, as well as the first quarter of 2012. In reality, BAK seems to have been included in the RFE less for the fraud that it was involved in, than for the “eye-catching” identity-theft that Azizi performed when he set up BAK. It’s almost as if the prosecutor is saying that when Samir Azizi ran out of sisters he could use as formal managers of his fraud companies, he stole the identity of complete strangers. Azizi is being painted as a man who is not to be trusted. Azizi is connected to BAK through his own admissions. However he is also connected to BAK with a stolen identity card. This card was uncovered during German police search. The card bears Azizi’s picture, but has the identifying data and name of Ajitpal Sekhon. Ajitpal Sekhon it turns out is formally the managing director of BAK. The “real” Ajitpal Sekhon is a Canadian criminal serving 10 years for drug offences. He does not speak German, and has never been in Germany.34 Samir Aziz is masquerading as Ajitpal Sekhon (perhaps to lead any investigating authority off the trail). BAK Enterprise UG was founded by Ajitpal Sekhon/ Samir Azizi on April 6, 2009 to trade in and distribute water purification and filtration systems, refuse bin disposal systems and to provide consulting services in the environmental marketplace. On March 15, 2010 the company was re-named BAK Enterprise GmbH, and the registered office was moved from Hanau, Hirchstraße 7 to Darmstadt. The RFE notes: In contrast to the indicated object of the company, the company firstly operated in the field of the emissions allowances trade, subsequently in the field of cell phone trade. We are not told much about BAK in the early (CO2) years, but in the later years, BAK did not file returns, although it issued invoices for the onward re-sale of its acquired inventory. As a result, in the later years BAK became liable for the amount of VAT placed on the outgoing invoices.35 The RFE says, “… the assessment of the incurred damage caused by VAT evasion is based on invoices issued by the defendant himself and 34 RFE at 88. VAT DIRECTIVE, Art. 203; s.14c para.1 UStG; Finanzamt Osnabrück-Land v. Bernhard Langhorst Case C-141/96. 35 15 on the concealed VAT amounts.”36 The RFE isolates three quarterly returns that were not filed: • October 11, 2011 – €29,886 • January 11, 2012 – €2,162,830 • April 11, 2012 – €1,372,858 A check of the EU-ETS database for the involvement of BAK Enterprise GmbH in the emissions allowances trade shows a very different BAK Enterprise.37 BAK neither bought nor sold CO2 in the last two quarters of 2011 nor the first quarter of 2012. However, in 2009 through 2010 BAK Enterprise was a CO2 “distributor” (the final party in a fraud chain that exports the fraudulent supply out of the country). This appears to be a classic CO2 MTIC fraud chain. Listed in the EU-ETS under the name BAK Enterprises Unternehmergesellschaft (haftungsbeschränkt) (BAK Enterprises UG (haftungsbeschränkt)) BAK bought 2,979,000 CO2 permits (or 96% of all the permits it ever purchased between February 15, 2010 and May 19, 2010) from the German registry account of Astra Trading FZE (Astra Trading FZE (Dubai)). At the average rate of €13 euro for a CO2 permit in February 201038 these purchases in 64 units reflect a €38,727,000 purchase price with a German VAT due of €7,745,400. These permits were broken into 71 units, and sold almost immediately to the another party through two different trading accounts: • 34 units are sold to the Czech account of TES Praha, a.s. (TES Praha, a.s. Osobní vkladový), and • 35 units are sold to the Danish account of TES Praha, a.s. (TES Praha, a.s.). These transactions account for 96.84% of all CO2 sales ever made by BAK. In other words, BAK was a very simple “distributor” set between a Dubai company’s German trading account and a Czech firm accounts in the Czech Republic and Denmark. BAK bought almost exclusively from and sold to these companies. Figure 7: BAK Enterprises UG 36 RFE at 89. DIRECTIVE 2003/87/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL establishing a scheme for greenhouse gas emission allowance trading within the Community and amending COUNCIL DIRECTIVE 96/61/EC, O.J. (L 275) 32, Art. 4. 38 The BlueNext is cited for a €13 price per CO2 permit by the Climate Wonk, EU ETS Reduced Emissions Despite Over-allocation (February 22, 2010) available at: https://thecarboneconomist.wordpress.com/2010/02/22/eu-ets-reduced-emissions-despite-over-allocation/; see also Carbon Emissions Trading and Carbon Taxes, EU Environmental Policy (February 21, 2010) slide 7b, available at: http://www.tutor2u.net/economics/presentations/Europe-Carbon-EmissionsTrading/handout.pdf 37 16 As a distributor BAK Enterprise UG was entitled to apply for a refund of €7,745,400. There is no reference in the extradition documentation about this transaction, although it is clear from the EU-ETS that a refund should have been available. This refund is more than twice the size of the €3,565,554 that the extradition documentation references for the BAK frauds. The only conclusion to be drawn from this is that either the CO2 evidence was not sufficiently convincing, or that the prosecutor simply wanted to demonstrate that Aziz is a man who is not to be trusted, because he is not above using multiple false identities to carry out his fraud. The case of a convicted and still imprisoned Canadian drug dealer, Ajitpal Sekhon, made a memorable case in point. One would hope that the German tax authorities did not miss this very apparent fraudulent refund preserved in the public EUETS database for anyone to uncover. (7) My iCell, GmbH €520,589 The tax investigation around My iCell determined that this company was the successor to iCell. Both iCell and My iCell were controlled by Aziz, and it appears that this My iCell was used to disguise the continuing fraud. Like BAK Enterprise, the My iCell assessment is attributed to the denial of input VAT deductions. The amount of fraud My iCell is responsible for is relatively small (compared with the other schemes Azizi was involved with) and the damage to the German Treasury 17 was put at €520,589. Azizi allegedly knew that the invoices received by My iCell were not based on any actual supplies or services. Four VAT returns are impacted: • June 15, 2011 assessed for an amount of €38,340 • July 5, 2011 assessed for an amount of €231,948 • July 22, 2011 assessed for an amount of €228,598 • September 6, 2011 assessed for an amount of €21,703 According to the commercial register of the Neuss Local Court My iCell dealt in the distribution of telecommunication devices (cell phones), play stations, as well as consumer electronics, and in addition My iCell was in the import/ export business for petroleum products, recyclable materials, and fruits and vegetables. The official managing director was Marian Iancu,39 but Azizi was deemed to be in actual control based on emails and intercepted chat protocols. It was Azizi who found customers, set prices, made payment transactions, and presented himself as the real managing director.40 Figure 7 below shows My iCell replacing iCell in the MTIC fraud chain: Figure 7: My iCell, GmbH replacing iCell GmbH The bulk of all payments made by My iCell were made to Adisony. These transfers were made through the Omnis Capital FX Ltd payment platform in London, UK.41 It is believed that this payment platform is the successor to Global Reach Partner payment platform used by iCell. Payments made to My iCell also came through the Omnis Capital FX Ltd. payment platform. 39 RFE at 89-90. RFE at 90. 41 RFE at 90. 40 18 CONCLUSION This consideration of Samir Azizi’s extradition leads to unsatisfying conclusions, if we mistakenly assume that an extradition hearing is comparable to a trial on the merits. There is no requirement in an extradition proceeding for the judge to “weigh conflicting evidence and make factual determinations.”42 Extradition hearings are “designed only to trigger the start of criminal proceedings against an accused: guilt remains to be determined in the courts of the demanding country.”43 The court in an extradition hearing must still base its finding of probable cause on competent evidence,44 but the committing court’s function is merely to determine whether there is “any evidence warranting the finding that there was a reasonable ground to believe the accused guilty.”45 As a result, what we have in the Azizi extradition is the German government’s view of Samir Azizi’s activities with respect to a number of MTIC fraud chains. There will be further factual development, and further litigation. What makes the Azizi extradition unique is that the FRE is not just government allegations; it is also full of admissions. Some of these admissions are expressly referenced by the German prosecutor not just as a statement of Azizi’s interpretation of events, but as a statement of fact that the German government has confirmed as true though further investigation. There are three sections of the FRE where this occurs: a three page section dealing with WOC GmbH, where the operation of payment platforms is explained;46 a four page section dealing with Wega Mobile GmbH, where the payment platform’s allocation of profits among the investors is detailed;47 a six page section dealing with iTrading GmbH & Co. KG, which details the involvement of the Deutsche Bank (Frankfurt and London) in the fraud chain.48 But there is more. From the Complaint we know there is much, much more, close to 3,000 pages more in the Azizi confessions that has not been disclosed to the court, nor to the US Attorney and the defense team, and which has certainly not been made public. The Complaint expressly references this material: Within the scope of the investigation Samir Azizi has already made a statement with regard to the charges, initially through his defense lawyer (p. 1240 ff, 1467 ff). In the regard, Samir Azizi expressly concedes that in his capacity as the de facto managing director of WOC GmbH he is responsible for the tax liabilities incurred. In the further course of the investigation, Samir Azizi announced through his defense lawyer a comprehensive statement regarding all charges (annex, p. 1641ff) and provided the investigator in charge, the witness Orths, with extensive 42 Quinn v. Robinson, 783 F.2d 776, 815 (9th Cir. 1986). Valencia v. Limbs, 655 F.2d 195, 198 (9th Cir. 1981). 44 Collins v. Loisel, 259 U.S. 309, 316 (1922). 45 Fernandez v. Phillips, 268 U.S. 311, 312 (1925). 46 RFE at 17-19. 47 RFE at 34-37. 48 RFE at 51-56. 43 19 information on the facts of the case in several telephone conversations (annex, p. 2795 ff). In agreement with Samir Azizi the witness Orths drew up a protocol based on the information provided (annex, p.1645). In the course of the telephone conversations with the witness Orths, Samir Azizi essentially conceded to the charges also with regard to the other companies under investigation (annex, p. 2796 ff, with regard to Wega Mobile GmbH see annex, p. 2801 ff, with regard to the companies iCell and iTrading see annex, p. 2804 ff, Nexo Chakfa and AS Handel see annex, p.2808 ff).49 Within these 3,000 pages is most likely the extended discussion of the terrorist connection that appears to be referenced by the German prosecutor in the RFE and is echoed by Judge Howard R. Lloyd in his Extradition Order. There are most likely more references to the payment platforms, and to the involvement of the Deutsche Bank. None of this detail needed to be brought forward, although it is clear that it is ready. There is a second shoe to fall, but it is not ready to fall yet. One final note, when the Complaint indicates that Azizi caused “total damage: 61.104.368,00 Euro,” it does not mean that Azizi received (cash in hand) €61,104,368 from the German Treasury. In some cases, Azizi, … gave tax authorities incorrect or incomplete statements about tax relevant events (77 counts) and failed to inform tax authorities about taxrelevant facts, acting contrary to duty (12 counts). Thus, tax authorities could not determine the VAT or determined VAT assessments that were too low. In this way, Azizi was able to obtain unlawful tax refunds. In some cases, the payment of refunds depended on approval by the appropriate tax office and were not finally approved though the application [attempt] for refund was made.50 The details of the Complaint make it clear that €14,380,174 was only an “attempted” fraud, because the “appropriate tax office” did not “finally approve” the refund requested. In addition, €8,316,947 was assessed in circumstances where returns were never filed. This is also not a refund situation. Thus, the total VAT inappropriately refunded is closer to €38,407,247 (and it could be much less, if we knew the accounting details). Figure 8: Summary of Total Damages WOC GmbH Ferrograph GmbH Failure to Submit 549,076 199,424 49 False Information 689,585 Attempt TOTAL In the Matter of: the Extradition of Samir Azizi, Complaint for Provisional Arrest with a View Towards Extradition (18 U.S.C. § 3184) 5:14-xr-90282-PSG (March 31, 2014) District Court, N. D. CA, Doc. 1 at 11. 50 Id., at 3. 20 Wega Mobile GmbH iTrading GmbH iCell GmbH AS Handel GmbH Nexo Chafka GmbH Hamsterrecke.de GmbH Amaan Enterprise GmbH BAK Enterprise GmbH My iCell GmbH 3,335,383 667,510 1,819,497 10,909,022 4,821,107 3,935,358 4,011,144 5,395,976 6,304,969 1,136,361 7,752,034 581,792 906,248 4,003,739 3,565,554 8,316,947 21 520,589 38,407,247 14,380,174 61,104,368