How a broker spent $520m in a drunken stupor and moved the global oil price

by Rowena Mason, Telegraph.co.uk

It’s probably not uncommon for City traders to wonder how they burnt so much cash during a drunken night on the town.

But Steve Perkins was left with a bigger black hole in his memory than most when his employer rang one morning to ask what he’d done with $520m of the oil trading firm’s money. …

It was 7.45am on June 30 last year when the senior, longstanding broker for PVM Oil Futures was contacted by an admin clerk querying why he’d bought 7m barrels of crude in the middle of the night.  ….

By 10am it emerged that Mr Perkins had single-handedly moved the global price of oil to an eight-month high during a “drunken blackout”. Prices leapt by more than $1.50 a barrel in under half an hour at around 2am – the kind of sharp swing caused by events of geo-political significance. Ten times the usual volume of futures contracts changed hands in just one hour. …..

By the time PVM realised the trades were not authorised and swiftly began to unwind the positions, losses of exactly $9,763,252 had stacked up.  Telegraph article

John Bates comments:

One powerful way to prevent this kind of accident or fraud is through the use of stringent pre-trade risk controls. The benefits of being able to pro-actively monitor trades include catching “fat fingered” errors, preventing trading limits from being breached, and even warning brokers and regulators of potential fraud – all of which cost brokers, traders and regulators money. PVM is a good example of this.

Ultra-low-latency pre-trade risk management can be achieved by brokers without compromising speed of access.  One solution is a low latency “risk firewall” utilizing complex event processing as its core, which can be benchmarked in the low microseconds.  Errors can be caught in real-time, before they can reach the exchange. Heaving that drunken trader right overboard, and his trades into the bin.  Bates blog

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.