Going rogue
25 June 2013
Late nights, no vacation time and cancelled trades aplenty鈥攋ust three of the warning signs that a trader is up to no good. SLT discovers the reasoning behind the risky practice
Image: Shutterstock
In 1993, Allfirst Bank in Maryland hired foreign currency trader John Rusnak to bring profits from proprietary foreign exchange trading.
Prior to his arrival, foreign exchange trading at the bank was limited to assisting bank customers in hedging against currency risk鈥攂ut treasurer of the bank Dan Cronin envisioned a future in which Allfirst reaped the benefits of a foreign exchange trading centre.
Ultimately, his decision would prove fatal. Rusnak ploughed on under his core鈥攁nd wrong鈥攖rading belief was that the yen was going to rise against the dollar, and used spot transactions, forward transactions and options to incur losses of $691 million. Using techniques such as entering bogus options in the banking system and manipulating the bank鈥檚 VaR calculations, Rusnak completely disengaged from the back office, hiding losses to buy himself some time and see if he could win back the money he had lost.
But time ran out. On 17 January 17 2003, Rusnak was sentenced to almost eight years in prison. AIB Group sold Allfirst to M&T Bank of Buffalo, New York, five months later, with more than 1100 Allfirst employees losing their jobs in the sale.
鈥淚n the late 1990s, the investment banking industry grew massively and very rapidly,鈥� comments Shaun Mathieson, a consultant for global banking and markets at software solutions provider SAS.
鈥淢arkets were going up, and at the time, it was easier to make a lot of money. Any occurrence of errors and people actively trying to defraud the organisations didn鈥檛 cause much pain back then, relatively speaking, in fact, it was almost accepted as a cost of doing business.鈥�
鈥淏ut when the markets started to crash, if people had been stepping outside the boundaries of their role, the losses became very big and the problems that were endemic across the industry started to show up. The consequence being that the back office, the middle office, and the risk functions put in place to control the banks hadn鈥檛 evolved at the same rate as the front office, meaning the control structures around it were simply not as mature.鈥�
Ten years on, and the media laps up scandals of the fallen traders. Tom Hayes is the latest in the public spotlight; appearing in court charged with rigging the inter-bank Libor rate.
An ex-trader at Citigroup and UBS, he is accused of conspiring with RBS, J.P. Morgan Chase, HSBC and others to manipulate the key rate. Adding to the pile are Chen Jiulin, former managing director of China Aviation Oil in Singapore; J茅r么me Kerviel, who is pending appeal of his conviction in Soci茅t茅 G茅n茅rale鈥檚 鈧�4.9 billion trading loss in 2008; and Kweku Adoboli鈥檚 $2 billion unauthorised trading loss at UBS鈥檚 investment bank.
The public sentiment around these individuals never strays too far from condemnation of the entire banking system鈥攁s well as disbelief as to how practices can go on for so long unnoticed.
Aside from the bank鈥檚 internal risk controls, there are a few firms seeking to capitalise on the scandals. Ernst & Young developed new software in January alongside FBI investigators that scans emails for the most common words and phrases used by rogue traders and fraudsters.
According to the firm鈥檚 research, these include 鈥榗over up鈥�, 鈥榳rite off鈥�, 鈥榝ailed investment鈥�, 鈥榦ff the books鈥�, 鈥榥obody will find out鈥�, and 鈥榞rey area鈥�.
鈥淓mails, sent in their thousands, between employees, officials and external parties form the major part of what is mostly positive daily interaction in companies,鈥� said Rashmi Joshi, director for fraud investigation and disputes services at Ernst & Young, at the time. 鈥淒espite being the prime means of all conversations, such unstructured data plays almost no role in the compliance efforts of firms.鈥�
鈥淲hile most organisations only focus on the numbers when investigating discrepancies, what we are seeing are ways of analysing words鈥攅mails, SMS or instant messaging鈥攖o identify and isolate wrongdoing.鈥�
Software provider SAS, meanwhile, is trying to catch the data-led market, with a system that automatically ingests lots of data from across the investment bank. This is a combination of trading data, existing controls that the bank has in place, market data and behavioural data鈥攕uch as when people are logging into their work stations.
鈥淔lagging alerts鈥攆or example, long-dated settlement instructions on a particular trade to push back the settlement date鈥攁 trade cancellation, or an off-market trade, are all marked on the report,鈥� says Mathieson. 鈥淟ooking at the graph below, we can see that these are the three riskiest traders. With this individual here we can see how much they have been trading鈥攓uite interesting, because this individual has a very high-risk score but hasn鈥檛 been trading at very high amounts. We have three zones: green, yellow and red. Anyone in red or yellow, a case starts in case management and a workflow will have started, and that trader will be under investigation from a specialist team.鈥�
Some indications that an individual is cause for concern are an increasing volume of cancelled trades; both market and dummy counterparty.
鈥淭here are a number of legitimate reasons why people could use dummy counterparties in the front office, but as can be the case, often there are people who are trying to defraud the bank and cover up positions,鈥� comments Mathieson.
鈥淭hey will book their fake trades against dummy counterparties because there is less of a process associated with it. Nobody in the back office is going to pick up the phone to anyone else in another bank and ask if it recognises this other trade, which means that the trader in question has a longer period to play with 鈥� it is the same as using long dated settlement instructions.鈥�
Just emotion
As well as specific market activities, a case for suspicion can be formed from the most routine goings-on.
Take attendance, for instance, whereby an alert from HR identifies a staff member who has gone 12 months without taking any annual leave, and login times are monitored. 鈥淭hese can be suspicious because if a person is covering something up, they have to be there full-time,鈥� says Mathieson.
鈥淥nce they step outside the office for a week then they can鈥檛 manipulate their metrics any more, their VAR will spike and they will be caught immediately. What is often the case as well is that they work much longer hours than everyone else, because they have to do their day job too.鈥�
Long hours; no holiday; and the constant fear of being found out鈥攁 rogue trader鈥檚 life does not inspire much envy. While there are some aspiring heros in line for a big bonus, there can be more empathetic characters amongst the bunch, says Mathieson.
鈥淪ome rogue traders in the past have had that gambling buzz, and think they will take a punt on the market that isn鈥檛 risk-managed, that isn鈥檛 on one of their client accounts in the belief that they will both make a lot of money and be a hero.鈥�
鈥淭hen again, I believe that the more common cases arise out of the highly-pressured and intensely competitive environment. Somebody will make a mistake and lose, for example, 拢300,000; be very scared about declaring that and decide to not tell anyone in an attempt to cover it up and try and trade their way out of it.
Prior to his arrival, foreign exchange trading at the bank was limited to assisting bank customers in hedging against currency risk鈥攂ut treasurer of the bank Dan Cronin envisioned a future in which Allfirst reaped the benefits of a foreign exchange trading centre.
Ultimately, his decision would prove fatal. Rusnak ploughed on under his core鈥攁nd wrong鈥攖rading belief was that the yen was going to rise against the dollar, and used spot transactions, forward transactions and options to incur losses of $691 million. Using techniques such as entering bogus options in the banking system and manipulating the bank鈥檚 VaR calculations, Rusnak completely disengaged from the back office, hiding losses to buy himself some time and see if he could win back the money he had lost.
But time ran out. On 17 January 17 2003, Rusnak was sentenced to almost eight years in prison. AIB Group sold Allfirst to M&T Bank of Buffalo, New York, five months later, with more than 1100 Allfirst employees losing their jobs in the sale.
鈥淚n the late 1990s, the investment banking industry grew massively and very rapidly,鈥� comments Shaun Mathieson, a consultant for global banking and markets at software solutions provider SAS.
鈥淢arkets were going up, and at the time, it was easier to make a lot of money. Any occurrence of errors and people actively trying to defraud the organisations didn鈥檛 cause much pain back then, relatively speaking, in fact, it was almost accepted as a cost of doing business.鈥�
鈥淏ut when the markets started to crash, if people had been stepping outside the boundaries of their role, the losses became very big and the problems that were endemic across the industry started to show up. The consequence being that the back office, the middle office, and the risk functions put in place to control the banks hadn鈥檛 evolved at the same rate as the front office, meaning the control structures around it were simply not as mature.鈥�
Ten years on, and the media laps up scandals of the fallen traders. Tom Hayes is the latest in the public spotlight; appearing in court charged with rigging the inter-bank Libor rate.
An ex-trader at Citigroup and UBS, he is accused of conspiring with RBS, J.P. Morgan Chase, HSBC and others to manipulate the key rate. Adding to the pile are Chen Jiulin, former managing director of China Aviation Oil in Singapore; J茅r么me Kerviel, who is pending appeal of his conviction in Soci茅t茅 G茅n茅rale鈥檚 鈧�4.9 billion trading loss in 2008; and Kweku Adoboli鈥檚 $2 billion unauthorised trading loss at UBS鈥檚 investment bank.
The public sentiment around these individuals never strays too far from condemnation of the entire banking system鈥攁s well as disbelief as to how practices can go on for so long unnoticed.
Aside from the bank鈥檚 internal risk controls, there are a few firms seeking to capitalise on the scandals. Ernst & Young developed new software in January alongside FBI investigators that scans emails for the most common words and phrases used by rogue traders and fraudsters.
According to the firm鈥檚 research, these include 鈥榗over up鈥�, 鈥榳rite off鈥�, 鈥榝ailed investment鈥�, 鈥榦ff the books鈥�, 鈥榥obody will find out鈥�, and 鈥榞rey area鈥�.
鈥淓mails, sent in their thousands, between employees, officials and external parties form the major part of what is mostly positive daily interaction in companies,鈥� said Rashmi Joshi, director for fraud investigation and disputes services at Ernst & Young, at the time. 鈥淒espite being the prime means of all conversations, such unstructured data plays almost no role in the compliance efforts of firms.鈥�
鈥淲hile most organisations only focus on the numbers when investigating discrepancies, what we are seeing are ways of analysing words鈥攅mails, SMS or instant messaging鈥攖o identify and isolate wrongdoing.鈥�
Software provider SAS, meanwhile, is trying to catch the data-led market, with a system that automatically ingests lots of data from across the investment bank. This is a combination of trading data, existing controls that the bank has in place, market data and behavioural data鈥攕uch as when people are logging into their work stations.
鈥淔lagging alerts鈥攆or example, long-dated settlement instructions on a particular trade to push back the settlement date鈥攁 trade cancellation, or an off-market trade, are all marked on the report,鈥� says Mathieson. 鈥淟ooking at the graph below, we can see that these are the three riskiest traders. With this individual here we can see how much they have been trading鈥攓uite interesting, because this individual has a very high-risk score but hasn鈥檛 been trading at very high amounts. We have three zones: green, yellow and red. Anyone in red or yellow, a case starts in case management and a workflow will have started, and that trader will be under investigation from a specialist team.鈥�
Some indications that an individual is cause for concern are an increasing volume of cancelled trades; both market and dummy counterparty.
鈥淭here are a number of legitimate reasons why people could use dummy counterparties in the front office, but as can be the case, often there are people who are trying to defraud the bank and cover up positions,鈥� comments Mathieson.
鈥淭hey will book their fake trades against dummy counterparties because there is less of a process associated with it. Nobody in the back office is going to pick up the phone to anyone else in another bank and ask if it recognises this other trade, which means that the trader in question has a longer period to play with 鈥� it is the same as using long dated settlement instructions.鈥�
Just emotion
As well as specific market activities, a case for suspicion can be formed from the most routine goings-on.
Take attendance, for instance, whereby an alert from HR identifies a staff member who has gone 12 months without taking any annual leave, and login times are monitored. 鈥淭hese can be suspicious because if a person is covering something up, they have to be there full-time,鈥� says Mathieson.
鈥淥nce they step outside the office for a week then they can鈥檛 manipulate their metrics any more, their VAR will spike and they will be caught immediately. What is often the case as well is that they work much longer hours than everyone else, because they have to do their day job too.鈥�
Long hours; no holiday; and the constant fear of being found out鈥攁 rogue trader鈥檚 life does not inspire much envy. While there are some aspiring heros in line for a big bonus, there can be more empathetic characters amongst the bunch, says Mathieson.
鈥淪ome rogue traders in the past have had that gambling buzz, and think they will take a punt on the market that isn鈥檛 risk-managed, that isn鈥檛 on one of their client accounts in the belief that they will both make a lot of money and be a hero.鈥�
鈥淭hen again, I believe that the more common cases arise out of the highly-pressured and intensely competitive environment. Somebody will make a mistake and lose, for example, 拢300,000; be very scared about declaring that and decide to not tell anyone in an attempt to cover it up and try and trade their way out of it.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to 网爆吃瓜 Finance Times
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to 网爆吃瓜 Finance Times
