#367 | Survivorship for Accumulation
The theory behind using survivorship for accumulation is pretty straightforward. Outside of any anomalous pricing at any particular insurer, the cost of insurance rates of a survivorship policy will be less than the cost of insurance rates of a single life policy on either of the individual insureds. Lower cost of insurance charges mean less drag on cash value accumulation. Less drag means better cash value growth. All else being equal, therefore, survivorship seems like it should be a natural fit for accumulation.
But there’s a countervailing force. For the purposes of setting 7702 limits, the life insurer has to use survivorship mortality rather than single life CSO mortality, which means that the minimum death benefit per dollar of premium is quite a bit higher on survivorship than single life. On a 55 year old male with a $100,000 7 pay premium, the minimum non-MEC death benefit is $1,105,058 on John Hancock Accumulation VUL 21. For two 55 year olds, the minimum non-MEC death benefit with the same premium is $1,718,344 on the survivorship version of the product (Accumulation SVUL 20).
Even with a substantially higher face amount, the mortality charges in the survivorship version are cheaper – much cheaper. You might even say that they’re infinitely cheaper, at least in the first year, when the COI for the single life product is $147 and it’s $0 in the survivorship policy. Over the first 10 years of the policies, both of which are funded at the maximum non-MEC premium using a level death benefit and CVAT, the single life policy has COIs that tally up to $7,558 whereas the survivorship is at just $226. The logic of using survivorship for accumulation works, at least the way John Hancock has priced the mortality for these two products.
However, survivorship mortality is merely deferred. Over the life of the policy, the CVAT corridor for the survivorship policy is wider than on the single life policy. As a result, the mortality charges in the survivorship policy eventually catch up to the single life policy and actually exceed it. You can see that playing out in the John Hancock policies. Take a look at the COI charges over time relative to the NAR. The dotted lines are NAR and the solid lines are the Cost of Insurance charges.
Despite the fact that the COIs ramp up, the survivorship policy has better accumulation performance by about 5bps at age 100 when illustrated at 6% gross (5.36% net) as I’ve done so far. The strange thing is that when you start changing the illustrated rate, the relationship between the two policies changes. Take a look at the age 100 IRRs for the two contracts at different illustrated rates.
What gives? Although they share a name and basic mechanics, these two policies are not quite identical. The VUL has a 15 year base charge period whereas the SVUL only has base charges for 10 years. Both policies pay a Policy Credit which, in classic John Hancock style, seems simple but actually has some complexities. The VUL starts the Policy Credit in year 21 whereas the SVUL starts it in year 16. The policy loads are also slightly different. These differences add up. By year 10, for example, the VUL handedly outperforms the SVUL because has lower – but longer – base charges. But the situation reverses by year 20 because the SVUL Policy Credit has kicked in. These sort of differences make generalizations about performance between survivorship and single life policies difficult.
In theory, Whole Life is an easier comparison. The only difference between, say, MassMutual’s 3% single life 12 Pay and the 3% survivorship 12 Pay should be the mortality table. Between those two products, we see the same sort of differences in initial death benefit as in the John Hancock VUL products. For a $100,000 10 Pay premium, the male policy has a $1.60 million death benefit, the female has $1.71M and the survivorship policy has $2.25M. At age 100, the male policy has an IRR of 3.9% and the female is at 4.04%, a substantial (and somewhat surprising) difference between the two that equates to $292,418 in additional incremental cash value.
That difference pales compared to the survivorship IRR of 4.30% and its additional cash value of $623,267 over the female-only policy. Total cash value in the survivorship is 45bps and nearly $1M more than in the male-only contract. It’s a huge pickup. A bit of that, undoubtedly, is related to mortality deferral. But some of it is also a design distinction between the two policies. In year 1, the cash value of the single life 12 Pay is about 32% of premiums paid. In the survivorship 12 Pay, it’s less than 1%. The difference is short-lived, to be sure, but even a short-lived difference can have long-term implications. All else being equal, I would expect a policy with lower early values to illustrate better performance over the long-term than a policy with higher early values and I think a bit of that is happening between these 12 Pay products.
There are other factors at play as well. New York Life also offers a compelling Survivorship Whole Life product that handedly out-illustrates both its traditional Custom Whole Life and Secure Wealth Plus. One thing that separates New York Life from Mass Mutual is that a significant portion of New York Life’s retail Whole Life sales are small-face, which has different mortality implications than survivorship, which is almost always for the ultra-affluent. A Preferred risk class for survivorship may actually materialize different mortality than Preferred for single life simply because the demographic buying survivorship is different. As a result, the survivorship may have a disproportionate advantage compared to single life.
In order to get a truer comparison, we need to look at survivorship and single life products with identical structures and sold to generally the same demographic. Penn Mutual’s Guaranteed Whole Life II and Survivorship Whole Life fit the bill. Both products have minimal cash value in the first year and Penn Mutual’s average face amounts for both products are high. With the same $100,000 premium as Mass Mutual and the same 3% guaranteed interest rate, the male-only policy has a $1.54M death benefit, female-only is at $1.63M and survivorship is at $2.11M. The survivorship quickly stretches its legs, beating the IRR on both of the single life policies by a significant margin. But over time, the advantage erodes. By age 100, the survivorship policy has an IRR of 4.35% compared to 4.23% for the male-only policy and 4.29% for the female-only policy.
Taking all of these things into consideration, I think it’s fair to say that, all else being equal, survivorship is a more efficient product for accumulation than single life, especially compared to male-only contracts. How much? Somewhere between 5 and 30bps over the long-run, depending on the cell and the assumed performance. To put that into context, the benefit is about the same as overfunding a policy using GPT rather than CVAT on either a single-life or survivorship policy.
In the same way as there are drawbacks for GPT that aren’t immediately obvious (such as administrative complexity), there are drawbacks to using survivorship – chiefly the fact that there is very little accumulation-oriented survivorship product inventory, particularly for Universal Life. Squeezing out an extra few basis points of IRR probably isn’t worth going with a product that the client wouldn’t have chosen otherwise. And then there’s the not-trivial matter of actually getting both spouses to agree to go through underwriting and running the risk that the underwriting doesn’t turn out how you thought it would, which might lead you back to single life anyway. Survivorship just puts a lot of new variables into the mix.
And for what? As I wrote in an article last year, the reality is that the drag on policy cash value growth due to Cost of Insurance charges shouldn’t even count as drag, at least not all of it*. Cost of Insurance Charges go to provide a death benefit that has value. Over the life of a policy across all policyholders, the aggregate benefit of mortality is worth the aggregate cost for mortality. The additional efficiency of a survivorship policy is really just a reflection of a deferred death benefit, not an actual increase in the overall value to the customer. It simply allows for a very slight shift towards accumulation and away from protection.
Nonetheless, I think there’s a place for survivorship accumulation products, particularly in Whole Life. It’s a reasonable option. It is actually more efficient. There are certain advanced sales situations where it makes sense, from a planning standpoint, to use an accumulation survivorship product even if everything else is equal. But for the vast majority of situations, accumulation-oriented single-life products are probably still a better fit.
*I’m assuming that the Cost of Insurance charges are accurately reflecting the pricing mortality experience of the life insurer, which is almost certainly not the case. Instead, life insurers tack a profit margin onto the rates. But, at the same time, they also make assumptions about mortality improvement that may not materialize. So for the purposes of the argument, I’m assuming that charges and benefits are matched.