2009/03/06

Systemic risks

.
The economic science has the term "systemic risk" as the key variable for insurance company-decision making.

Insurances are in the business of relieving their customers of risks - which is obviously risky for them, but not too bad as long as they're big enough and their risks not too systemic.
Systemic risks could turn into damages all at once or in a very short period. An insurance company specializing on hurricane damages in Florida would have a lot of systemic risks, while an insurance company that operates in dozens of markets and has only a small share of risks from Florida storm insurances has much less systemic risk.
Diversification is the smart counter to systemic risks, size (too big to fail) the dumb one. All companies have to consider this and have to look for a diversified portfolio - even though that's in conflict with efficiency and competence requirements at times.

I never bought into the idea of reinsurance companies who insure insurance companies - they didn't seem to be strong enough to do their job in a severe crisis.
I believed I didn't understand the principle, but the crisis last year suggests that I was rather right (and I was certainly only one in millions who had scratched their head about this).
The ultimate reinsurer is apparently the state - which is probably not up to the job either.


Maybe it would be a good idea to apply the systemic risk factor to security policy as well.
We know old rules of thumb that you better have strong allies and as few enemies as possible. Maybe systemic risk is a factor that can be used for a refined look at this.

It's one more argument to NOT alienate entire groups and build up a host of risks that might later turn against us with a common motivation.

Example:
It wouldn't be as bad to piss off all Arabs with a different reason for each country as to piss them off for one common reason.

The quantity of enemies/risks is probably not much of a problem as long as they don't turn on you all at once.

This could also be applied to military forces:
It's related to the unity of command principle applied in some forces. You're not as much in trouble if you disrupt the connection among superior OPFOR (non-systemic risk) as if you face a unified OPFOR (systemic risk).
Or seen the other way around; an army has too much systemic risk in its portfolio if it uses only one mode of communication - it will suffer badly if this one mode fails.

It's interesting how a risk-control element from economic science can explain the need for offensive counter-communications electronic warfare in warfare.

I cannot remember 'diversity' as one of the many near-holy grails in military forces - although redundancy is quite popular at least in the technical realm

Sven Ortmann

No comments:

Post a Comment