Saturday, 20 October 2012

How Risky are Your People?

We have seen with the global melt down how excessive risk taking led to individuals, portfolios, organisations and entire economies being over exposed to risks that have turned sour. Many of the European banks regularly under go 'stress tests' to bottom out the resulting losses  and it is going to be a long road back to solvency for many.

This over exposure to risk can also happen within a business and need not be limited to financial investments. It could manifest itself in putting large resources into certain projects, agreeing stage payments or credit terms with large clients, targeting certain markets or taking on certain competitors.

Pretty much every area of your business takes positions both internally and externally where they absorb some risk. It is worth asking, how well have any of these decision makers been considered for their propensity to take acceptbale levels of risk. In the 'real politik' of stability and contractual arrangements, you would not remove these managers if you realised that they were excessive risk takers, but you may put additional checks and balances in place to oversee the risk aspect of their decisions.

This then begs the question, how could you identify managers that may be poor at recognising and exposing themselves to risk. The answer is simple, you assess for risk. You could use a variety of risk rating tools, perhaps some role play scenarios or some observational based study. None of these on their own would be definitive as biological factors, over confidence bias, payment structures, organisational culture would all feed into risk taking behaviour. For example some studies have found that testosterone is positively associated with risk taking.

Such assessments would however give you a benchmark or indication of the risk takers in your business, relative to their peers. One such tool is the Iowa Gambling Task, introduced by Antoine Bechara, António Damásio, Daniel Tranel and Steven Anderson. It is designed to assess risk preferences by simulating real-life decision making using uncertainty, rewards, and penalties.

Initially, it was used to compare individuals with prefrontal lesions with healthy individuals. Participants play a game where they are given four decks of cards and an amount of money (e.g., €1000). They are then instructed to select cards one at a time and try to lose the least amount of money and win the most.

Some decks are "bad decks" and others are  "good decks", because some will lead to losses over the long run, and others will lead to gains. For example, turning a card can give a (relatively large) reward (€100 in decks A and B and €50 in decks C and D) or a penalty (large in decks A and B and small in decks C and D).

Playing from decks A and B leads to an overall loss, while playing C and D leads to an overall gain.  After encountering a few losses, normal participants begin to avoid the bad decks with large losses. They recognise that the occasional large gain (of €100 in deck A) will more than likely be wiped out by a large loss from the same deck.

You could tweak this concept to make it more relevant to the type of decisions your managers make and if you have the data available it may be interesting to see if their previous risk history correlated with their score.

You could measure how quickly managers recognised and moved away from the 'bad deck' with its large but long term unsustainable rewards. You could tweak the rules to apparently encourage short term risk taking (by offering bonuses for reaching a win target quickly) but also state long term viability cannot be compromised. This would then allow you to see which managers are primarily focused on short term and fail to take sufficient notice of long term implications.

When designing a job specification, if a key requirement was the ability to read risk, to appreciate long term goals rather than large unsustainable quick wins, then perhaps this type of test could form part of the recruitment process.

If a role involved minimal supervision and significant autonomy, then it may inform decisions you make on introducing additional oversight or exposure measurement. For the European banks, it’s too late for any of this.

There are several free and open source assessments similar to Iowa Gambling Task available. A version of the Iowa Gambling task is available as part of the PEBL Project .There is a customisable version that works with Google Spreadsheets or your own spreadsheet. There is also a version for android smart phones and tablets. As with any tool, you need to ensure you are fully licensed, based on your own circumstances before use.

No comments:

Post a Comment