John Hancock $22,500,000 Variable ILIT Breach

Errold F. Moody Jr. PhD, MSFP, MBA, LLB, BSCE, CFP

Life and Disability Insurance Analyst

September 2007

This is the overview of a Majestic Variable Estate Protection 98 Survivorship Life Insurance that represents a breach by the company and agent- but allowed by (non professional) trustees and accepted by the insured. The end result is a risk exposure of approximately 50% that the policy will not have sufficient sums to last a lifetime.

This was a contact initiated by the husband worth about $65,000,000 who had been sold a $22,500,000 variable life issuance policy inside a Irrevocable Life Insurance Trust. The reason for such a sale was that he "trusted" the agent. His two brothers had used the services of the agent (who became a family ‘friend') and he simply continued the process. (The total amount of insurance for these three brothers is over $100 million) But something seemed amiss in the presentation and sale and he wanted to know where he stood. That said, he was reluctant to involve his two older brothers who were the trustees and I had to do extra work to avoid any internal controversy. Unfortunately, the use of "trusted" family or friends as either the salesperson or trustee made this analysis more difficult. But it is not uncommon at all.

The facts are as follows: Husband age 46 Wife age 37 Agent is a CLU. He had been previously sold two other variable life policies within the ILIT for $5,000,000 with Pacific Life and $4,000,000 with Security Life. I was not provided all the background of these sales- though requested from the client- but remained suspect as to why these were not "combined" into the new sale. (Actually, I was wondering why they are done at all.) As with many such situations, it does not have to make sense. But while the family trustees wish to remain ignorant of the repercussions, a professional trustee cannot.

Survivorship life with husband as standard and wife as preferred. The estimated joint life expectancy at the time of issues was 43.7 years and based on 1980 Mortality rates. Policy issued 2006. This posed two questions. First, the policy was outdated- even John Hancock brokerage had no info on it (though I am sure it could have been received from the home office). Secondly, why would an agent use an out of date policy with a 1980 Mortality table versus the newer 2001 Mortality table (adopted in 2002). The newer tables showed longer lifetimes and generally LOWER premiums. This was not addressed as part of the report since I could not find definitive information.

The illustration showed a $253,000 initial lump sum and $58,000 for 16 years with no further premiums due to the internal growth of the funds asset (all figures rounded) . It was based on a 9% gross return and a 8.07% net. The illustration went for 63 years with no more than the total premiums as stated and a final insurance value of $35,000,000. The illustrations offered Internal Rates of return, maximum insurance charges and more. There was one illustration with a 0% gross return and another at 7.52% net- the first expiring after 24 years and the latter after 35 years. The John Hancock "Illustration Explanation' does indicate that "if the investment performance is less than illustrated, additional premiums may be required to keep the policy inforce". They also note, that "John Hancock recommends that you review additional information using various invesmtent returns to understand how actual performance may affect the policy values and premium payment schedule."

The intent may sound logical and adhere to basic fiduciary standards. Not so. Anytime any type of variable investment return is estimated at a flat rate, the whole process is not just suspect, it's pretty much useless. A flat rate of return analysis does not work, has not worked and universally will not work in the future.

As to the zero percent return- no reasonable person expects this and it is dismissed out of hand as simply not in the realm of real life. The 7.52% net is also misleading in that it also shows a flat rate of return and a 35 year period until default. That is not reflective of the risk in a variable policy. Consider for example starting the policy in 1999. Immediately thereafter the policy would have lost 44% of its initial lump sum value (2000-2002). The policy would have expired very quickly.

The JH statement that the "client review additional information" has no meaning whatsoever. What investment returns are they supposed to use? What does standard deviation mean in regards to policy values? What does a Monte Carlo indicate? There is no legal offset to liability in making a statement that the consumer (and agent) either does not understand (and is not offered as part of the illustration) or is understood by the company but completely and materially avoided or misrepresented as part of the analysis. That the agent cannot provide the commentary, never mind the analysis, countermands the attempt to limit liability. The point to this commentary comes from direct personal discussion with a brokerage manager for the San Francisco office of John Hancock regarding this policy and its illustrations. When I indicated that there existed a sizable probability of failure, he simply indicated that ‘John Hancock would indicate that the agent was the one that had the duty to inform the client of the ‘problems' not John Hancock'. Lovely thoughts, but useless in litigation.

Admittedly there are no licensing requirements for risk in a variable policy, nor the elements of risk, nor certainly any focus to the type of software necessary to provide the possibility of default (Monte Carlo) but if one is acting as a fiduciary, that is no excuse at all. You either know what you are doing and why or you will be liable.

Standard commentary on risk

Though there was no true commentary on risk from JH, the standard position on risk is that it goes down over time. Risk, in that case, is generally defined as standard deviation. It actually does statistically go down over time and leaves the consumer with the misconception that time will ameliorate whatever losses are sustained. Maybe. And therein lies the problem; how much time? Is any money being taken out during the loss period? Is there any other income available? Ad nauseam. In many cases, the losses sustained are simply too much (particularly while other income is being drawn put- certainly for insurance premiums under a vanishing premium philosophy), and the consumer will never recover.

The fallacy, beyond anything else, is the simple fact that time does not reduce the risk of loss, it INCREASES it. As shown in the formal paper, there is a significant probability that one would have only about 45% of the assets they expected at any given point in time. Sure, the consumer might have LOTS of money, but it is the risk of losing substantial amounts that is the key. This is not an insurance issue per se- it is the mandatory knowledge needed for a trustee simply to understand investment performance and risk. But it is also, obviously, necessary in order to address variable life insurance since the illustration shows that, under a ‘vanishing premium', there will always be adequate money to continually fund the policy. Not necessarily true. The test of the viability has to come from some other source/method than a flat rate of return. Even with the 8.07% net return that JH says is absolutely adequate to maintain the policy for life, a statistical analysis shows that that can be absolutely false.

Monte Carlo

The software used a random number generator that runs a trial 500 separate scenarios to determine what might happen with this variable survivor policy. I could not vary the investments to reflect the correlation of the current funds (in fact, I know of no software capable of doing this). I can only run 60% large cap and 40% bonds or a 100% large cap. But between those two, I can get a sense of what exposure one might have and then augment with added information and real life adjustments.

As one can see, the positive values can be enormous if the economic and investment gods are with you. But the termination of the premiums after 16 years forces the returns (8.07%) to stay consistent in order to maintain the policy for life. I repeat, past history simply doesn't allow that. One of the main problems is that there is so much time that is involved due to the young ages and the fact that premiums are stopped so "early". No matter, if I use the conservative 60/40 split, there is a 60% change of NOT having sufficient sums. If I go with the more aggressive stance of 100% large cap, the policy will NOT have sufficient sums 30% of the time.

First, I doubt that anyone would suggest that an investor maintain a 100% exposure in pure equities (large cap) for a lifetime. No matter, this risk for loss is NOT shown in the policy illustration and I know the analysis would be much worse if that was addressed. I also know by the report enclosed that the printout does not reflect such risk either. I have talked with the software firm's officers - they really don't know what I am talking about since they had some software investment programmers do the detail and they have never yet included such element.

So, whether one has an agent or investment adviser review the current allocation in a variable policy, the key is whether they state a relative degree of safety (non default) for the future. They cannot. Even making some adjustments in my mind for the way the program runs, it still seems that in the best of situations by using a large cap fund, one is still way over, say, a 5% acceptance of default rate (30%). And if I use a more conservative stance, the risk of default goes up even more (55% to 60% depending on the number of times I ran the scenario). Doesn't seem to be any way that I could vary the adjustments significantly save for simply increasing the premiums for years and years.

It is true, as I have repeatedly mentioned, that there may be lots of money that might be had if things went ‘right'- from $80,000,000 to over $300,000,000. Possible- but doubtful if one uses the insight from the likes of Benoit Mendlebrot, Nassem Taleb et al who define risk far more succinctly. Again, no matter, because I believe the policy will demand more funds in the future. It has more risk than a no lapse policy by far.

That said, will the default cause more payments than $60,000 annually in order to keep the current policy afloat? Unknown- the programs do not do that type of calculation at all. It would take a very sophisticated program to offer this and there are none in existence at this time. I do not know if one could actually be programmed to do such calculations.

Trustee requirement:

Not all of the report was shown here. But it should be evident that a variable policy cannot, under any circumstances, be utilized in an ILIT without the commensurate analysis of the risk of default at some time in the future. That the insurance companies do not provide the relevant risk analysis is no defense for a professional trustee. That the agent does not provide the analysis will also be scant offset- though to be fair, most trustees might escape some liability due to the supposed knowledge and expertise of the agent. That said, a professional trustee- actually any advisor- would require an independent advisor review each and every policy annually for adequacy. A professional trustee cannot release the responsibility to the various agents to come in and simply do an inforce illustration. Professional trustees are acknowledged to have the skills, knowledge and real life insight to request independence and an analysis that includes the basic formats of investment risk. A knowledge of Monte Carlo is mandatory. Those are the minimum standards as a fiduciary.

In short, the insured (or the trustees if you will) were sold a policy- actually all of the policies- by a woefully unknowledgeable and incompetent agent, CLU or otherwise. John Hancock violated its duty as well with a completely incoherent illustration that does not define the risks.

If the trustees wish to retain these policies, they should set aside a separate kitty for the loss exposure. The beneficiaries need to be made aware of this risk. The problem is, once again, that the insured simply does not have the internal wherewithal to confront his brothers to reconcile the problem.

In the alternative, the parties should attempt a 1035 exchange into a No Lapse policy. As such, no one will never have to attempt to decipher a policy illustration again. The beneficiaries will be provided the insurance as the trust required.

Postscript: The client sat down with the agent to discuss the issue of No Lapse- though without presenting my report. The agent said that if he is concerned about the allocation, he could have his own professional money managers review the allocation separately. He also noted that the No Lapse policy from JH had no internal growth for three years. He also stated that JH would charge a surrender charge for the 1035 exchange.

First of all, the money management firm also had a fiduciary duty to address the risk of investments over time. It is irrelevant to have someone comment on allocation without doing some type of a Monte Carlo analysis. They might shuffle the allocations around- fine. Maybe the correlations of the various investments will be better. No matter, the risk still remains and must be noted by them as well.

As to the No Lapse having no returns- what's the point? The agent's factual ‘statement' missed the point of using a No Lapse to begin with. You just want insurance- the internal buildup is not a concern. That said, if the agent was clueless to the risk of default, I can understand the comment. Useless, wrong, and misconstrues the entire concept of a variable policy- but I understand how a incompetent agent could say that.

I contacted JH directly- there were no surrender charges. Big difference in the premiums due for a No Lapse if an illustration is made with significantlly less funds to start with. Also wrong and a breach of duty for the agent not contacting the company directly.

All that said, it appears client will do nothing. I hope his wife is lucky over the next 50 years.