Pricing Question

[SlowGo]

Dear All

I ve managed to price a product for example, Fund a lump sum benefit at retirement that would be used to purchase a PPO plan, using monthly premiums till retirement with a death benefit rider

Now I am finding out the following

1. Premiums for lower ages are higher than those of advancing ages
2. Premiums for those with 5 years to retirement seems to be increasing with advancing age

With 1, I expect this because the future value at age 65 of a PPO premium is higher for a 20 yr old than a 40yr old for instance, so this is no surprise
2. For 2, I can't seem to get my head around this one, is it that the interest and mortality to retirement being higher for older ages is offset by the interest?

Oh, the PPO premium is level for all ages, this could be the problem?

Im not sure if I can justify charging a higher premium to a younger person than an older person

Please share your thoughts

[Hackeni]

SlowGo,

I am not sure that I understand your question exactly.

However, the level premium might cause implicit subsidy, i.e., higher premium charged to younger & healthier policyholders to offset the higher cost for older policyholders.

[SlowGo]

Sorry,

What I am getting at is that is it normal to expect a situation like 1 and 2 that I described above?

What is an appropriate way of designing this type of product?

There doesnt seem to be any incentive for the younger folks to buy into this insurance if their payment is decreasing as they get older (until 5yrs before retirement)

You are right about the cross subsidization

The problem with the product design for this client is that their premium is Flat for all ages, so designing a flat premium product for all ages will likely create cross subsidization

The only way I see around this is to try and split the premiums into age bands using age factors (this data is hard in South Africa, so I may use Milliman factors), and then applying the premium accordingly

I hope this is clearer
Thanks for responding

[Hackeni]

You might also consider this paper by Jeff Petertil, http://www.soa.org/files/pdf/aging_curves.pdf

[SlowGo]

Thanks