Although past crises in the financial markets have always been put down to credit or market risk, scratch the surface and you find a plethora of op risk causes underneath. Remember Michael Milken and the events at Drexel Burnham Lambert that sparked the junk bond crisis in the late 1980s? Op risk. How about the 1987 stock market crash that was exacerbated by computer programmes running amok? Op risk. Even the emerging markets crisis of the late 1990s – which saw countries such as Thailand, South Korea, Russia and Brazil suffer currency runs, collapses in their government bonds and other economic problems – was down in substantial part to a range of op risks. These included inadequate policies and procedures when granting credit, lack of conflict-of-interest policies at all levels of banking organisations, and other factors that were basically op risk issues.
As the subprime crisis gathers momentum, op risk executives should take this opportunity to highlight the contribution this industry can make to the analysis of what has happened, and prevention of any repeat.
I invite readers of OpRisk&Compliance and anyone else in the industry to comment on ways op risk methodologies can be used to analyse what has happened in the subprime crisis, or how future events like this could be prevented with the implementation of op risk protocols. I'm keen for us as an industry to share thinking around this subject area, and to prove that op risk methodologies can contribute in fundamental ways to improving the safety and soundness of the banking system, as well as its overall profitability.
So how could the subprime mortgage situation have been detected earlier by the firms who originated the mortgages? How about by the firms that repackaged the mortgages into collateralised debt obligations and other instruments? What op risk precautions could investors have taken? And how about the rating agencies – what internal risks should they have been aware of around the rating process?
I'm keen to hear your thoughts. E-mail me and we'll publish as many as we can.
Editor [email protected]