AG43 - should it change?

[JMO Fan]

I have heard several actuaries complain that AG43 needs to improve. For those who are familiar with it, what do you think is wrong? What would you change?

[MathGeek92]

it's wrong because the methodology applies "real world" techniques to value an option and set reserves and capital according to this "real world" answer.

Any risk mitigation technique that is then used, forces the realization that the reserves are under held because once you start putting in your mitigation technique, it forces up reserves. The natural end point is the Market Price of the option.

Next, any mitigation you do use, regardless of how conversative your utilization assumption might be, is deemed ineffective by reducing the amount of credit you get for hedging.

It encourages risk taking and discourages any mitigation resulting in masking the true nature of the risk.

The reserve should be set equal to the PVB - PFfees of the rider benefit only under a risk neutral valuation. Capital then can be set equal to a mutiple of it (a C2 type risk due to pricing assumptions being wrong).

Just my opinion

[JMO]

Not only says that it needs improvement, and why, it also explains how to accomplish the improvement.

[JMO Fan]

Statutory accounting requires greater conservatism than GAAP ("going concern"). Is it sufficient for AG43's purpose to simply make the present value basis more conservative, or is there a need for adding a margin for the risk? How should conservatism be reflected in the statutory reserve under AG43?

[MathGeek92]

Originally Posted by JMO Fan

Statutory accounting requires greater conservatism than GAAP ("going concern"). Is it sufficient for AG43's purpose to simply make the present value basis more conservative, or is there a need for adding a margin for the risk? How should conservatism be reflected in the statutory reserve under AG43?

If the underlined above is the goal, then AG43 fails miserably as the Stat reserve is almost universally guaranteed to be less than the MV of the option, thus I would say that STAT is <<< GAAP

1. You get to use ALL future profit of the entire VA block to offset losses.
2. Any hedges are haircut, thus reducing the effectiveness of hedging, thus encouraging more risk taking because reserves and capital will be less if not hedged versus hedging, a totally ridiculous outcome, IMO. I should get capital RELIEF from hedging my risk versus INCREASING my capital - caps for emphasis only... not really yelling it.

If the Goal is to make it more conservative than GAAP, then one could simply make up a "drop" amount and force reserves to be the MV of the option AFTER the drop. For example, at issue my option value is "zero" because I calibrate my PVB=K% * PV Rider fees. If I revalue (keeping my K% locked) assuming the market goes down say 15%. I would be forced to hold reserves = X. (a positive amount, thus creating conservatism over the MV of the option)

HOWEVER, If I'm hedging my risk, and I have a detailed hedge program in place offlaying the market risk of the option, then my reserves should be < X, not > X. They may not be zero, but it should help to hedge, not hurt.