Insurance SIFIs - Metric ideas from the AAA

Spotted this, and it looked intriguing - American Academy of Actuaries proposing metrics "regarding size, interconnectedness and risk assumption" to assess whether insurance companies are systematically important - the AIG angle still seems to be driving this desire (resisted by almost every insurance lobbying group) to include a couple of insurance giants in the SIFI pot. I haven't looked for the paper itself, so google away if you are hot on this topic!

The second notable point here was that a "voting insurance expert" has beeen elected to the Financial Stability Oversight Council - not sure if that increases or lessens the likelihood of some insurers being captured as SIFIs.

Central Bank of Ireland - Consultation on Risk-based Impact Metrics

Having followed this subject from when the consultation first opened, I was fascinated to see what had materialised. The comment and response document and the individual responses are worth poring over if you are on that side of the pond. My take was as follows;

• Four impact categories confirmed (High/Medium High/Medium Low/Low)
• Surprisingly clear tone that supervision for 'Low' impact firms will be minimal, reactive, and that failure is not a problem. 'Medium Low' firms will however receive proactive supervision with some inspections.
• 'Vigorous Enforcement' will be used to discourage ineffective risk taking - evidence of this already in themed inspections and issued fines
• Best practice from Australia, Norway and Canada utilised (may look into the reasons behind this)
• Impact metrics are being road tested - will not be used for levy setting immediately
• "Primary consideration" for metric selection is measurement of the scale of the firm (as opposed to nature or complexity in Solvency II jargon)
• The PRISM system will be expected to identify business model and governance weaknesses
• Metrics for Life firms all seem reasonable, if pre-determined by common sense. 
• Sticking with the use of divisors to normalise the different metrics onto a common scale - I really like this idea, and advocate it for ERM risk profiling as well, so I'm glad they will "trial and error" it.

Of the comments, I looked specifically at what the IIF had to say, and whether their requirements were met by the outcome. I spotted;

• Alignment of levies with resource consumed was on their agenda
• They wanted weighted averages to be used for the divisors, and regular review of their appropriateness
• Premium income by business line and market share were on their list of possible metrics, but didn't make the cut (the former did, but only as APE).

All in all, the split of impact metrics and PRISM as two ways of determining allocation of resource seems like it has the potential to lead to internal dispute further down the line - hopefully I will get a chance to look in more detail at both aspects at a later date