Subprime: Algo First database extols enterprise risk management

Op risk failures underlying the subprime crisis are more significant than initially thought

Algorithmics’ latest newsletter says the impact of op risk failures in the US subprime crisis was much bigger than first thought. “The deeper I dig into [the subprime crisis], the more I realise that it is intricately linked to op risk issues,” says Penny Cagan, head of research and managing director at Algo OpVantage.

Algorithmics’ First database had set out the underlying operational risks contributing to the crisis in its June newsletter.

“Initially it appeared that poor operational risk practices, such as those associated with preparing and reviewing documentation and the due diligence process, errors in servicing loans and models that do not appropriately reflect risk, contributed to the credit problems,” she says. “However, the ensuing volatility could potentially lead to an increase in operational risk issues, such as processing errors and fraud, leading to an escalating feedback mechanism.”

The newsletter highlights the lessons to be learnt from the subprime mortgage crisis. What started as a credit-risk event has grown into a global market crisis that holds essential lessons for the future of op risk strategy. The interconnectivity of risk elements – with op risk failures at the core – is the main message from the Algo newsletter. Firms need to abandon the traditional approach of separating credit, market and op risks into separate silos, and to look at managing risks more holistically through an enterprise risk management (ERM) strategy to prevent this type of crossover event.

“Hopefully the subprime crisis will be seen as a call to pause and look upon one’s operation,” says Cagan. “The bad decisions are made in boom-time periods of exuberance, and at difficult times like now everybody needs to look at their risk practices. There is a need to step back and look at the risk analysis and the risk management culture.”

As sensible as the ERM strategy sounds in theory, in practice it is much more difficult to implement. And with many in the industry still undecided on exactly what constitutes ERM, widespread adoption of this strategy still seems distant.