Should you let a trader mark their own books? Only if your CTRM helps detect fraud.
Good article last week from CTRM Center. The CFTC charged a trader with defrauding an affiliate of a large commercial bank by concealing losses. How? He inflated the value of his NYMEX Ethanol (Platts) Futures contracts to conceal losses he was incurring. Citi and its affiliate realized $40 million in losses before detecting it. Whoa, really?
In 2010, CTRM systems did have role-based access so you could bar traders from entering their own mark data. But, this got me thinking -- why not have CTRM systems flag marks that are out of bounds -- especially on exchange-traded products where data is published? Also, if your system is in the cloud, you can do analysis and know what the 'aggregate' mark should be.
It's time CTRM software used data and analytics to be proactive. Molecule is cloud-based and data-fed. Use cases like these are what all of us should be designing to make CTRM iteratively more powerful and secure.