NEW YORK STATE DEPARTMENT OF FINANCIAL SERVICES In the Matter of BARCLAYS BANK PLC, BARCLAYS BANK PLC, NEW YORK BRANCH CONSENT ORDER UNDER NEW YORK BANKING LAW §§ 44 and 44-a The New York State Department of Financial Services (the “Department”), Barclays Bank PLC, and Barclays Bank PLC, New York Branch (collectively, “the Parties”) stipulate that: WHEREAS Barclays Bank PLC is a major international banking institution with more than 132,000 employees and total assets exceeding $2 trillion; WHEREAS Barclays Bank PLC has operated a foreign bank branch in New York State (“the New York Branch”), licensed, supervised and regulated by the Department since 1963; WHEREAS the New York Branch has more than 500 employees and total assets exceeding $36 billion; WHEREAS Barclays Bank PLC and the New York Branch (collectively, “Barclays” or the “Bank”) engaged in manipulative conduct and attempted to manipulate benchmark foreign exchange (“FX”) rates around the world, during at least 2008 through 2012, to benefit Barclays’ own trading positions; WHEREAS in some instances, Barclays conspired with other banks in order to coordinate trading, attempt to manipulate exchange rates, or coordinate bid/ask spreads charged; WHEREAS the coordination and conspiracy to manipulate were intended to benefit Barclays trading results by maximizing profits or minimizing losses to the detriment of counterparties, thereby harming consumers of financial products and services; WHEREAS from at least 2008 to 2014, Barclays engaged in misleading sales practices, including by deceiving clients concerning the application of “mark-ups” to FX trades; WHEREAS the misconduct discussed in this Order involved providing false and misleading information to customers and markets; WHEREAS Barclays PLC has agreed to plead guilty to a violation of Section 1 of the Sherman Act (15 U.S.C. § 1); WHEREAS Barclays is entering into a resolution with the U.S. Commodity Futures Trading Commission (“CFTC”) relating to violations of the Commodities Exchange Act (7 U.S.C. §§ 9, 13) and CFTC Regulation 180.2, 17 C.F.R. § 180.2; and WHEREAS Barclays is entering into a resolution with the Board of Governors of the Federal Reserve System relating to violations of the Federal Deposit Insurance Act (12 U.S.C. § 1818); WHEREAS Barclays is entering into a resolution with the U.K. Financial Conduct Authority relating to principles of reasonable care to organize and control its affairs responsibly and effectively with adequate risk management systems, which is FCA Principle for Businesses 3; NOW THEREFORE, to resolve this matter without further proceedings pursuant to the Superintendent’s authority under Section 44 and 44-a of the Banking Law, the Department and Barclays agree to the following: FACTUAL BACKGROUND The FX Market and Fix Manipulation 1. The FX market involves the buying, selling and exchanging of currencies, and is one of the largest, most liquid and most actively traded markets in the world, with an average daily turnover of $5.3 trillion in April 2013. 2 2. In a typical FX spot transaction, currencies are traded against one another in pairs. As the world’s reserve currency, the U.S. Dollar is the most actively traded currency in the FX market. The overwhelming majority of all FX trading is conducted in G10 currency pairs, while the most popular currency pairs are the Euro/U.S. Dollar (EUR/USD), the British Pound Sterling/U.S. Dollar (GBP/USD) and the U.S. Dollar/Japanese Yen (USD/JPY). 3. FX market participants include banks, investment firms, commercial companies, central banks, hedge funds and retail customers, which trade various types of instruments, including spot, forward, swap, futures and options contracts. The primary FX trading centers are located in New York City and London, although FX trading occurs around the world and continues 24 hours a day, excluding weekends. 4. A typical FX spot transaction involves the exchange of currencies at an agreed rate for settlement on a spot date—usually two business days from the trade date. In addition to trading directly in the market, clients can also submit “fix” orders to various large international banks, including Barclays, which then shoulder the risk of the trade and agree to deliver the requested currency to the client at the “fix” rate, which is determined at a subsequent time based on trading in the interdealer market. There are numerous such “fix” or benchmark rates, which are set at specific times each day. The most widely referenced benchmark rates include the 4:00 pm (London time) WM/Reuters fix (the “WM/R fix”) and the 1:15 pm (London time) European Central Bank fix (the “ECB fix”). 5. For G10 currency trading, the WM/R fix rates are based on actual bids and offers taken from electronic trading systems, which at the relevant times used a one-minute window around the time of the fix. The ECB fix takes into consideration price levels established through trading at the time of the fix, but the precise methodology is not disclosed. 3 6. For certain less liquid currencies, particularly in emerging markets, benchmark rates, such as the reference rate published by the Chicago Mercantile Exchange (“CME”) and Emerging Markets Traders Association (“EMTA”) for U.S. dollars and Russian rubles (the “CME/EMTA USD/RUB Reference Rate”) are set based on indicative quotes obtained by a survey of certain market participants. 7. Prior to a fix, clients place orders with banks (including Barclays) to buy or sell a specified amount of currency “at the fix rate”—i.e., the rate that would be determined at the upcoming fix. Traders with net orders to sell a certain currency at the fix rate make a profit if the average rate at which they buy the currency is lower than the fix rate at which they sell to their clients, while traders with net orders to buy a certain currency at the fix rate make a profit if the average rate at which they sell the currency is higher than the fix rate at which they buy from their clients. 8. By taking these orders to transact at a yet-to-be-determined rate, banks expose themselves to exchange rate movements, and FX traders may buy or sell currency in or around the fix or fix window in order to manage this exposure risk and often to obtain a currency position large enough to complete the order. Depending on the volume of trading during or around a fix window, transactions to buy or sell currency can affect the fix rate. For example, a bank that sells, or even simply offers to sell, a large amount of currency just before or during the fix may cause the fix rate to move lower. The resulting rate movement would benefit the trader, who makes a larger profit due to the increased spread between the fix rate at which the trader committed to buy the currency from the client and the higher rate at which the trader sold the currency to the market. However, this increased profit comes at the expense of the client, who ends up selling the currency to the market (via the trader) at a lower fix rate. 4 9. This dynamic creates a risk that a trader will deliberately engage in trading behavior designed to manipulate the fix rate in order to increase his or her profits by buying or selling large orders, or simply by making bids or offers for large amounts with the intent to move prices rather than engage in actual transactions. These profits could be increased even further if the trader coordinated his or her trading activity with FX traders at other banks. 10. By sharing information through multi-bank chats and colluding in their trading, these already influential traders dramatically amplify their ability to manipulate prices. The fixes use a narrow time window, during which a small number of traders at a few global banks control the majority of the volume, and have leverage to influence prices. 11. A manipulative trading strategy employed to affect market prices rather than engage in actual transactions is “crossing the spread” and placing an order that unnecessarily reaches “deep into the book” – that is, several levels beyond the market’s offered bid/ask spread. This intentionally shifts the bid/ask levels in the desired direction rather than merely executes orders at the best available price. The technique is intended to push prices in a direction before a current bid/offer can be renewed by a counterparty at the previously existing price, or close to it. 12. Knowing that the times around a fix are volatile, other market participants have an incentive to stay out of the way of the banks executing fix orders. Successful coordination among the fix-executing banks and prior instance of successful manipulation of prices around the time of the fix thus reinforces the lesson that other market participants should stay out of the way at that time. Only members of the multi-bank chat among large fixing banks have the shared information and the visibility as to the direction of aggregate fix orders, as well as the ability and incentive to cooperate in fix execution. As one Barclays trader stated on a multi-bank chat: “we 5 are 3 of the top 4 eur books on the planet . . . if we cant help each other with liquidity . . . who can?” Barclays’ Misconduct 13. Barclays employs traders who trade FX financial instruments in the inter-bank market and with non-bank counterparties as part of the Bank’s Global FX business. Among others, Barclays’ FX traders trade in the following product sectors: FX spot, FX forwards and non-deliverable forwards (“NDFs”) and FX options. 14. The misconduct described in this Order was not confined to a small group of individuals; it involved more than a dozen employees, who acted with the knowledge and oversight of some senior desk managers, and spanned geographically across numerous countries, including Barclays’ offices in New York and London, among others. Barclays Engaged in Manipulative Conduct Regarding, Attempted to Manipulate and Conspired to Manipulate Trading in Certain G10 Currency Pairs 15. From approximately 2008 through 2012, certain FX traders at Barclays communicated with FX traders at other banks to coordinate attempts to manipulate prices in certain FX currency pairs and certain FX benchmark rates, including the WM/R and ECB fixes. The majority of these communications took place in multi-bank online chat rooms. 16. Certain FX traders at Barclays routinely participated in these multi-bank chat rooms and often had multiple chat rooms open at the same time. In their attempts to manipulate these benchmarks in the chat rooms, Barclays FX traders exchanged information about the size and direction of their orders with FX traders at other banks, as well as coordinated trading, and discussed the spread between bids and offers which the banks were showing to customers. The 6 exchange of information among the traders at multiple banks via the use of chat rooms facilitated this type of conduct. 17. FX traders, including at Barclays, used code words when communicating in multi-bank chat rooms, in order to conceal certain aspects of their communications and avoid detection. 18. One particular chat room, referred to by the participating traders and other traders as the “Cartel” included FX traders from Citigroup, JP Morgan, UBS, RBS and Barclays who specialized in trading the Euro. Membership in this exclusive chat room was available only to traders who were invited to the chat room. Two Barclays EUR/USD traders were members of this chat room: one from December 2007 to July 2011 and another from December 2011 to August 2012. 19. One Barclays FX trader, when he became the main Euro trader for Barclays in 2011, was desperate to be invited to join the Cartel because of the trading advantages from sharing information with the other main traders of the Euro. After extensive discussion of whether or not this trader “would add value” to the Cartel, he was invited to join for a “1 month trial,” but was advised “mess this up and sleep with one eye open at night.” This trader ultimately survived his “trial” and was permitted to remain in the Cartel chat room until it was disbanded at some point in 2012. 20. FX traders at Barclays employed various strategies in their efforts to manipulate the fixes. One method was known as “building ammo,” whereby a single trader would amass a large position in a currency and then unload the “ammo” just before or during the fix to try to move prices. 7 21. On January 6, 2012, one Barclays trader, who was also a Head of the FX Spot desk in London, attempted to manipulate the ECB fix by unloading EUR 500 million right at the fix time, stating in the Cartel chat room “i saved 500 for last second” and in another chat room “i had 500 to jam it.” 22. On September 13, 2012, Barclays had a net position from clients to sell EUR 119 million, and this same Barclays trader built approximately EUR 30 million in additional “ammo” from Fortis Bank SA/NV, and EUR 169.5 million from HSBC. This trader stood to make additional profit if the ECB fix dropped. He sold EUR 175.4 million in the nine minutes leading up to the fix. While the market was rising up until 5 seconds prior to the fix (with best offers hovering around 1.2913), this trader sold EUR 147 million in the last three seconds before the ECB fix, which ultimately was 1.2910. As a result, Barclays made a profit of $16,000 on these trades. 23. In another example, on December 6, 2011, Barclays had client orders to buy a net EUR 52 million at the ECB EUR/USD fix rate. In addition, a Barclays trader agreed to buy an additional EUR 143.5 million at the fix rate from other banks, leaving him with EUR 195.5 million to sell at the fix. This additional “ammo” from other banks offered the possibility of driving down the price of the EUR/USD for the setting of the fix, which would allow Barclays to pay less for the euros it had committed to purchase at the fix price, while potentially selling for more. In addition, through the cooperation of traders at other banks, additional client orders of EUR 120 million, pending at other banks, were also cleared ahead of the fix, so as not to counteract the upcoming Barclays effort to drive the price lower. A trader at another bank wrote to the Barclays trader: “u selling still? . . . get a late 20 so ive sold them” and stated that he was planning to “watch your magic haha.” 8 24. The trader sold the euros prior to the ECB fix with the intent to push down the price, and anticipating that the fix rate at which Barclays had committed to buy 195.5 million euros would be lower than the rate at which he would sell. In the 13 minutes leading up to the fix, Barclays sold EUR 196 million at an average price of (approximately) 1.33947. Immediately before the fix, the trader placed an order to sell EUR 100 million at 1.3394. The ECB fix ultimately was driven down to that exact level, and this low fix level, lower than the prices before and after the fix, resulted in a profit for Barclays of approximately $14,270. A few seconds after the fix, the price recovered to 1.3396, where it had been before the fix. 25. Dissipating the buy orders of traders at other banks before the fix, and concentrating Barclays’ sell orders near and at the fix, allowed Barclays to attempt to push the price at the fix lower than it would have been without this deliberate marshaling of orders to create the lowest possible price – a price below what Barclays itself had sold for, on average. Without the active cooperation and coordination among the traders at multiple banks, via the use of chat rooms, the Barclays trader would have had neither the information to indicate that pushing the price was feasible, since there were not large contrary orders pending, nor the tools to attempt to accomplish that forced, temporary push lower. 26. In another example of manipulative trading, intended to push prices down before the fix, rather than simply execute orders by selling at bid prices, on January 3, 2012, a Barclays trader repeatedly placed orders below the levels of the bids in the market. For this fix, Barclays had client orders that required it to sell EUR 400 million. Barclays also obtained additional “ammo” from other banks, raising its net position for sale to EUR 482 million. In the seconds prior to the fix, the Barclays trader submitted an order to sell EUR 50 million at 1.3013 when the market was trading at 1.3015/16, and then to sell EUR 100 million at 1.3012 when the market 9 was trading at 1.3014/15. A trader not focused on pushing prices lower would simply have submitted orders that matched the best offer, or more aggressively, obtained immediate execution by agreeing to sell at the bid price. Offers lower than the guaranteed execution at a higher available bid are strongly probative of an intent to move prices lower, rather than efficiently and fairly execute orders. On this day, Barclays obtained approximately $59,000 in profit from the fix trading. 27. An additional tactic for reducing the risks involved in seeking to manipulate market prices was for the traders at the various banks on a multi-bank chat to agree to stay out of each other’s way around the time of a fix, and avoid executing contrary orders while an effort to push prices was being deployed. Traders would also cooperate with price manipulation efforts by seeking to “clear the decks” of contrary orders early, in order not to dilute the deployment of the full “ammo” nearer to the fix, as part of an effort to move prices beyond the narrower range that would be maintained by a more routine, even execution of orders. 28. For example, in a June 28, 2011 chat with a trader from HSBC, a Barclays trader reported that another trader was building orders to execute at the fix contrary to HSBC’s orders but Barclays assisted HSBC by executing trades ahead of the fix to decrease that other trader’s orders: “He paid me for 186 . . . so shioud have giot rid of main buyer for u.” 29. In another discussion on a multi-bank chat, on December 1, 2011, with a trader from Citigroup, a Barclays trader indicated “If u bigger. He will step out of the way. . . We gonna help u.” 30. On February 15, 2012, a Barclays trader worked to clear the decks in advance of the ECB fix to assist his fellow Cartel member at UBS, stating “hopefully taking all the filth out for you [UBS Cartel member]” and “hopefully decks bit cleaner.” Despite having net sell orders 10