Call (+1) 650-212-1212

Blog

Operational Risk Scenario Analysis: If Modeling Processes and Systems Is Difficult, Try People

Sean Salleh 08 Oct 2013 Risk and uncertainty

Operational risk scenario analysis may sound like a mouthful, but here’s what it’s all about. Firstly, operational risk is the risk to a business of loss from processes, systems and people. While this includes external events, it typically excludes legal, strategic and reputational risks. Scenario analysis means considering the risk from events that have yet to happen. Put these two concepts together, and you get operational risk scenario analysis. It’s a hot topic for financial and insurance institutions in particular. But if they lose money because of operational risk, where exactly is the problem – in the inability to model the impact of people as an operational risk, or in the denial of quality input from other people that could help nullify the risk?

Diagram showing analysis of operational risk in IT systems Image source: wikimedia.org

Scary Loss Figures

The size of losses associated with operational risk can be considerable. Unauthorized trading activity (a rogue trader qualifies as a people risk) lost Barings Bank $1.4 billion in 1995 and left it bankrupt. The Enron aftermath led to JP Morgan paying out $2.2 billion for settlement. Trading gone out of control more recently cost French bank Société Générale €4.9 billion (about $6.6 billion). In theory, operational risk scenario analysis should be able to help here. It should let senior management assess the risk of the next rogue trader or other operational disaster, and decide upfront how best to react.

Important but Wrongly Modeled

While agreeing with its importance, others consider operational risk scenario analysis is too often misunderstood and wrongly applied – ‘arbitrary and inaccurate’ reads one analysis. Because organizations are often principally interested in modeling rare but high-impact events, they may lack the data that allows them to model such risks. As in other cases of limited or insufficient data, a facilitator with competence in expert elicitation may be able to usefully supplement available information.  Naturally, using more people to try to identify the true risk associated with other people (or systems or processes) in the first place introduces a further factor of risk. Understanding the nature and possible biases of expert elicitation is therefore a requirement for accurate findings in this kind of scenario analysis.

Factors for Your Analytica Model

If you build a model in Analytica for operational risk scenario analysis, you’ll need to model the right inputs and using appropriate probability distribution functions for those inputs. Directives like Basel III (international) and Solvency II (European) indicate the risks to be assessed for banks and their brethren. These range over internal and external fraud, employment practices and workplace safety, clients, products and business practices, damage to physical assets, business disruption and systems failure, and execution, delivery and process management.

Assessing Potential Damage

When you get to the stage of assessing possible losses, operational risk scenario analysis practitioners suggest an analysis of the frequency and the severity of the potential impacts. Probability distributions associated with frequency include the negative binomial and the Poisson distributions. Examples of those associated with severity are the log-normal and Weibull distributions. All of these (and many more) are immediately available in Analytica. Calculating Variance at Risk (VaR) is then a way of understanding what (in monetary terms) is at stake.

If you’d like to know how Analytica, the modeling software from Lumina, can help you with operational risk scenario analysis in finance, economics, agriculture, medicine and many other areas, then try a thirty day free evaluation of Analytica to see what it can do for you.

» Back

Sean Salleh

Sean Salleh is a data scientist with experience in guiding marketing strategy from building marketing mix models, forecasting models, scenario planning models, and algorithms. He is passionate about consumer technologies and resource management. He has master's degrees in Operations Research from University of California Irvine and Mathematics from Northeastern University.

Leave a Comment