Frequency and Severity

There are not any statistics available, but some form of guessing frequency and severity for each risk is most likely the most popular approach to risk assessment.

Which is a problem, since that approach is fatally flawed.

There are at least three fatal flaws:

1. Guessing is a weak approach to assessing anything.
2. The Frequency/Severity idea only actually applies to a few rare situations.
3. Frequency/Severity pairs are not actually comparable.

But there is a simple fix for this.  That fix would be to pick two levels of frequency and then determine the loss that is likely at both levels of frequency.  Most useful would be to look at worse losses that might occur under “Normal Volatility” and also look at the losses for each risk that would be considered a “Realistic Disaster”.  Losses from different risks CAN be compared on each of those two levels.

For more information about the Frequency Severity approach and this alternate approach, see:

For ERM, a Better Solution to Guessing Frequency and Severity Pairs for Risks on the Willis Towers Watson Wire