*Formatting of these documents may vary from those filed with the Division of Insurance due to the conversion process to make them available through the web.

COMMONWEALTH OF MASSACHUSETTS

DIVISION OF INSURANCE

_______________________
Plan of Reorganization of    )
John Hancock Mutual Life )                                                                                       Docket No. F99-04
Insurance Company             )
_______________________)

Introduction

My name is James H. Hunt. I have reviewed certain documents in the public record in connection with the proposed Plan of Reorganization ("Plan") of John Hancock Mutual Life Insurance Company ("Hancock") and will testify in this statement to my opinion regarding allocation of value to Hancock’s policyholders eligible to receive consideration in the demutualization, and will comment on Hancock’s proposed Closed Block.

Qualifications

My qualifications and experience are as follows. From 1983 until my retirement in mid-1998 I was employed by SBLI as an actuary with responsibilities primarily for pricing of new products. In this connection, I compiled several mortality and lapse studies, and I became familiar with all actuarial aspects of SBLI’s in-force and new business. I never held a management position.

From 1965-1969, I served as Commissioner of Banking and Insurance in Vermont. From 1972 to 1976 I was a life insurance actuary in the New Hampshire Insurance Department. From 1976-1979 I was Director, State Rating Bureau, Massachusetts Division of Insurance. I have served on many committees of the National Association of Insurance Commissioners over the years.

The Consumer Federation of America ("CFA") is the nation’s largest consumer organization. It is sponsored by 260 organizations with a combined membership of 50 million. I serve as its uncompensated Life Actuary. For 15 years prior to becoming affiliated with CFA, I was director of the National Insurance Consumer Organization, which in 1995 became the CFA Insurance Group. I am a Fellow of the Society of Actuaries (1963) and a Member of the American Academy of Actuaries (1965). I operate a program in which I review cash value life insurance policies.

Scope of Testimony

The proposed Plan involves, among other things, a determination by Hancock of the amount and nature of the consideration payable to eligible policyholders, the development of a "fair and reasonable" formula for allocation of the pool of consideration among eligible policyholders, and the establishment of a Closed Block of assets for the purported purpose of protecting policyholders’ dividend expectations in the future and from which their policy benefits will be paid. My testimony will focus on whether or not the Plan is prejudicial to policyholders regarding the formula for allocating policyholder consideration which must be fair and reasonable, and the adequacy of funding of the closed block and its proposed operations.

In preparing this statement, I reviewed many documents in the public record of this proceeding, including but not limited to:

1. Policyholder Information Statement, Parts 1 and 2 and the exhibits thereto, in particular, the Closed Block Memorandum and [allocation of consideration] Memorandum;
2. Interim Report on the Allocation of Policyholder Consideration produced by the Division of Insurance’s advisor Tillinghast-Towers Perrin dated May 28, 1999;
3. Memorandum in Support of Tax Opinions of Debevoise & Plimpton, dated May 12, 1999;
4. Memorandum of Milliman & Robertson ("M & R ") dated May 4, 1994, submitted to the Massachusetts Division of Insurance ("DOI") in connection with the DOI’s review of the demutualization of State Mutual (now First Allmerica);
5. Memorandum of Godfrey Perrott of Milliman & Robertson dated January 26, 1995, prepared in conjunction with the State Mutual reorganization;
6. Memorandum of Law of Debevoise & Plimpton dated May 4, 1994, prepared in conjunction with the State Mutual reorganization; and
7. The Written Statement of Godfrey Perrott.

I also consulted with the 1987 Report of the Society of Actuaries Task Force on Mutual Life Insurance Company Conversions (TSA XXXIX, 295-391, hereafter "Report"), and the Exposure Draft Proposed Actuarial Standard of Practice on Allocation of Policyholder Consideration in Mutual Life Insurance Company Demutualizations (April, 1999, hereafter "Exposure Draft"). As an initial matter I note that Tillinghast-Towers Perrin ("Tillinghast") did not complete their report on their review of Hancock’s "Phase Two" testing of the allocation of value to eligible policyholders (discussed further below) in time for me to adequately incorporate the findings of that report into this statement. I must therefore reserve my right to supplement or amend this testimony at a later time as appropriate, as I must with regards to any other testimony.

In particular I note that, according to Tillinghast’s Interim Report on the Allocation of Policyholder Consideration dated May 28, 1999, the following issues were unresolved and/or expected to be addressed only in Tillinghast’s report on Phase Two calculations:

• Prospective investment return rates for Closed Block business (assumptions used to calculate Actuarial Contribution ("AC") should mirror Closed Block assumptions);
• Correction of the following: 1) documentation in the Global Assumptions Binder that is inconsistent with the AC models; 2) an error in the book yield for quarter 28 of the projection for certain lines; 3) minor inconsistencies between lines regarding treatment of investment expenses and defaults;
• the documented mortality assumption for Weekly Premium Ordinary ("WPO") policies (i.e., the industrial policies) is not consistent with the Closed Block funding assumptions;
• there are inequities in the approach used to allocate dividend option profits for some recently-issued Premium Notice Ordinary ("PNO") policies;
• treatment of certain types of profits on PNO policies had not been reviewed, but were to be in Phase Two;
• there are problems with calculating AC by policy (as opposed to by line) for the Monthly Debit Ordinary ("MDO"), WPO and PNO policies, which are individual products and make up the vast majority of Hancock’s policyholders;
• for PNO policies, testing to replicate the published AC factors covered the historic period only, with prospective testing to occur as part of the Phase Two review;
• in addition, there are outstanding issues with respect to calculating the consideration for Individual Nontraditional life policies, Individual Annuities, and GLTC (Group Long-Term Care).

The unresolved issues regarding allocation of consideration to the Premium Notice Ordinary ("PNO") policyholders, the largest group of individual policies and the largest group in the closed block, significantly impair my ability to adequately address any issues with respect to consideration to the PNO policyholders. This omission of data is significant because the PNO line, according to Phase One AC calculations, has an AC of $3,332.4 M, or 66% of total AC, while the total Phase One AC for Closed Block policies is $3,656.4 M, or 72% of the total AC. According to a recent draft document, Tillinghast’s Preliminary Conclusions on the Allocation of Value to Eligible Policyholders dated November 4, 1999, total AC for the PNO line was $2,535.4, or 60% of the total AC, and the AC for Closed Block policies total was $2,973.0, or 71%. In fact, Tillinghast did not prepare a detailed report on the Phase Two calculations prior to the deadline for filing pre-filed testimony, but rather presented their findings orally to the Division along with a slide presentation. Therefore critical data remains, as of the date of this statement, unavailable for review.

Allocation of Policyholder Consideration

John Hancock should allocate Variable Shares in its Demutualization by the "Historic Only" Method.

In the Written Statement Of Godfrey Perrott dated October 30, 1999, he notes at Page 11 that the Variable Component of the consideration to Eligible Policyholders will be based on "past and future contributions to surplus." This is of course the "Historic Plus" method, as contrasted with the "Historic Only" method, of determining each policyowner's fair share of approximately 80% of the Company's large surplus, which has been accumulated over more than a century. Mr. Perrott observes that the Historic Plus method has been used in all demutualizations since 1990, but that does not mean that it is the fairer method nor that it is faithful to the letter and spirit of the governing Massachusetts law. In fact, in a prior actuarial communication dated May 4, 1994, M & R observed that three out of four major precedents used Historic Only, with Equitable (an insolvent insurer) being the exception. I believe the Historic Only method is a much fairer way of allocating the Variable Shares. See memorandum of Debevoise & Plimpton and Ropes & Gray dated May 4, 1994, which makes the legal case for the Historic-only method unequivocally. See also M & R memorandum, May 4, 1994.

At page 13, Mr. Perrott quotes the 1987 Report of the Society of Actuaries Task Force on Mutual Life Insurance Company Conversions (TSA XXXIX, 295-391, hereafter "Report") as recommending (1) that the allocation of compensation among policyholders should "be based primarily on the relative contributions of policyholders to the surplus of the company," certainly a retrospective, or historic-only method, then adds without quotation (2) that such "contributions should include both past and anticipated future contributions to surplus." Item (2) of course is "Plus." It is not easy in reading the Report to find these thoughts coupled so closely, but early on in the Report the Task Force indicates that this is a "new measure (P. 297)." Nor is it easy in the rather long text to find the rationale for departing from the Historic-Only method, hitherto "the most appropriate and, possibly, only approach to be used in conversion determination value." (Page 318). In the next sentence, the Task Force adds, "This approach would also be consistent with those conversion statutes where aggregate membership value is based largely on statutory surplus." See also the recent exposure draft Actuarial Standard of Practice on Allocation of Policyholder Consideration in a Demutualization (stating some statutory schemes require Historic Only). The governing statute, Chapter 175 Section19E(3) says that consideration "shall be based upon the insurer's entire surplus . . . without taking into account the value of nonadmitted assets or insurance business in force." As noted in the May 4, 1994 M & R Memorandum, this requires the use of the "historic only" method of allocating policyholder consideration. I am aware that M & R later revised their opinion and stated that Massachusetts law allows the use of the "historic plus prospective" allocation method. In my opinion, the statute calls for the Historic Only method of allocation.

One looks for reasons for the Report's "new" method. Again at Page 318 we read, "In the view of the Task Force the amounts determined using this approach [Historic-Only] would only by chance represent a good measure for individual policies of the amounts that are required with future premiums to mature the policy (or group of policies) and to pay policyholder dividends." But this test appears to be one for determining the amount of the closed block, not for allocating the demutualization consideration fairly among policyholders based on their "relative contributions to the surplus of the company."

More telling from my perspective is the sentence that follows at page 318, "This approach [Historic-Only] has an additional serious deficiency as a conversion allocation methodology because it will produce negative contributions for most of a company's individual policyholders." I question whether "most" is accurate, and it surely is not accurate if weighted by the reserves behind each policy. Is the one year's duration policyholder with no (or minimal) reserves to have the same weight as the fifty-year-old policyholder with $10,000 in reserves? Is this not an actuarial judgment about a "political" risk; that a conversion might be more difficult if recent policyholders get fewer shares? The Task Force then says it does not believe Historic-Only is a "theoretically correct measure of policyholder contributions" and dismisses it as not "useful" without examining how it might be made theoretically more correct and useful.

Use of the historic plus prospective method more generally will tend to shift consideration away: 1) from lines of business where the amount of new business written has declined, towards lines of business where the amount of new business written has increased, and 2) from older policies with an established historic contribution to surplus to newer policies with no or little historic contribution to surplus. This was noted by Mr. Perrott of Milliman & Robertson with respect to the former State Mutual’s reorganization in 1995. See January 26, 1995 memorandum of M & R. In that case, Mr. Perrott estimated that the shift of consideration away from individual lines to group lines if the historic plus, as opposed to the historic only method were used, would be in the range of 5-11% of the total consideration.

In my view, the Historic-Plus method produces bizarre results. The fifty-years-duration policyholder with aggregate premiums and reserves far exceeding a new policyholder's may get no more variable shares than the newcomer, who in receiving fixed shares gets more than he bargained for and, one can argue, more than he or she deserves. The surplus of John Hancock was built up over 100 years; in what way does a brand new or recent policyholder whose actual, rather than prospectively calculated, actuarial contribution is negative deserve more than a fixed share? Or, if you wish, a much smaller variable share calculated by a Historic-Only method the actuaries did not take time to formulate? Also, I understand the $15 million policyholder represented by counsel received only 17 shares. Is not a variable policy entitled to variable shares based on future profits from cost of insurance charges and asset charges?

In my work reviewing cash value life insurance policies, I usually receive no feedback. The exchange is normally by mail, and surprisingly few customers take advantage of my invitation to call and discuss. In one case, however, involving recently demutualized Mutual of New York (MONY), I was advised of the number of shares that a young couple and her mother received. I had access to amounts of insurance and cash values on each from in-force illustrations. Mother had bought her $25,000 whole life policy in June of 1991 at age 59 and was in the 8th policy year last December when I did the work; her total cash (surrender ) value on the 1999 anniversary was illustrated at $4,804. The young couple's $100,000 whole life policies were bought in December 1995 at ages 33 and 32 (her), were just completing the 3rd policy year and had total cash (surrender) values of $2,080 and $1,480. There were no loans. Shares distributed were 12 for Mother, 13 for daughter and 15 for husband. Mother had had her policy more than twice as long as the couple's, and her cash value was greater than the total of the other two, yet she got less than half as many shares as her daughter and son-in-law. Surely Mother had contributed more to MONY's surplus (or drained it less); what common logic leads to a result like this? The Task Force concluded that Historic Plus was "easy to describe and understand," though they did not say by whom. Given an allocation formula like MONY's, and like that prescribed for John Hancock, one doubts very much that policyholders, generally without a clue about the workings of whole life policies, will (a) understand the "Plus" aspect of the formula or, if given examples like the one in this paragraph, (b) understand and agree with the results it produces. The significant difference in allocation that results from using the speculative historic plus method versus historic only is not disclosed anywhere in the materials, and certainly policyholders have no idea what the issue is. The "black box" approach to allocating 80% of Hancock’s policyholder consideration is not fair or reasonable to policyholders as required by statute.

The case for Historic-Plus would be stronger if a recent policyholder were giving up the chance to receive the policy values, including dividends, that were illustrated when he bought (or as the future experience provides). But the cardinal rule of a demutualization is found at page 296 of the Task Force Report, among others: "The assets initially allocated to this closed block would be an amount sufficient to pay the (then) current scale of dividends if the (then) current experience continued." Accordingly, the recent policyholder receives a fixed share and gives up nothing with respect to future dividend expectations according to the plan. (For the record, I am not arguing for a negative variable share to be offset against the fixed share.)

The uncertainty of historic assumptions for some lines can be exacerbated by projecting those uncertainties into the future, resulting in an inaccurate and therefore unfair allocation of variable consideration. For example, the Group Life and Other Health segment, which had a Phase One AC of $288.3M, or approximately 6% of the total AC, had no case data available prior to 1986, other than effective dates. Annual contributions to surplus for these policies were calculated by multiplying the estimated annual policy premium by the annual line of business profitability ratio for that line. How do we quantify how much this estimate may be off ? Millions of Hancock policyholders have small industrial policies (WPO line) whose policies were made paid-up in 1985. Hancock stopped writing this business in 1967, so there will be little if any prospective contribution to surplus; variable shares allocated will be very few in number. A large number of these policyholders are Massachusetts citizens. (See, for example, chart attached to the Letter of Melanie W. Bockley to Julie Bowler, Esq., dated June 30, 1999, showing that approximately 11% of Hancock’s policyholders as of the end of 1998 were in Massachusetts). The history of John Hancock dating back 100 years is the history of these small, probably working-class policyholders and the generations that preceded them. Because of inflation, the Actuarial Contribution as calculated for those still on the books will be small. But, I presume, the historic contribution of prior generations is not. It is highly questionable whether new Hancock policyholders should receive large number of shares based on prospective calculations when the small policyholders who made Hancock in the first place receive so few shares.

I will add that in reading the Policyholder Information Statement, I found no discussion of variable shares that could be understood in plain English. A statement such as that below, when extended with examples and rationale, could have been most helpful and could have influenced votes: "A large allocation of variable shares is being made to recent policyholders, despite their having made no contribution to John Hancock's surplus." In my mind, failure to make this or a similar disclosure is prejudicial to policyholders. The use of the Historic Plus method, which Hancock’s own actuarial consultant and law firms once opined was not permitted under Massachusetts law, is prejudicial to policyholders, including the Participants opposed to the Plan.

Closed Block

While I am not an expert on Closed Block matters, there are several comments I would like to make regarding the Closed Block, and do so based on my experience in the insurance industry generally.

• The public record in this matter is not fully developed enough to permit policyholders or the public generally to examine the Closed Block assumptions and funding.
• Tillinghast’s final review of Hancock’s Closed Block, if complete as of the date of this statement, has not been made publicly available in time for Tillinghast’s findings to be examined and discussed in pre-filed testimony. As of June 22, 1999, the date of Tillinghast’s Preliminary Conclusions on the Maintenance of Policyholders’ Reasonable Dividend Expectations, testing of the projected cash flow for the large PNO line as well as the MDO line had not been completed. See June 22, 1999 Preliminary Conclusions at page 15.
• Expense factors used to determine closed block funding were based on an undisclosed business plan projected forward for the life of the closed block.
• While Hancock must make up the difference, should after some time in operation it becomes clear that the Closed Block is not adequately funded, it is not clear what steps Hancock has taken or will take to ensure that the incentive of management to place interests of stockholders over that of the policyholders in the Closed Block does not prejudice Closed Block operations going forward. After all, the payment of dividends is still made at the directors’ discretion and there is a provision for terminating the Closed Block in the future.
• Hancock has not disclosed the difference between the contractual dividends (supposedly) protected by the Closed Block and the dividends policyholders received as members of the mutual, which included a distribution from the profits of the enterprise as recognized by Congress.
• the difference between the assets which fund the closed block, which are described at p. 25 of Perrott’s testimony as "a selected portion of John Hancock’s investment segments which support individual life insurance," and the entire company’s assets which would support policyholder dividends in the former mutual.
• Hancock has provided no information on performance of closed blocks set up in other demutualizations and established and opined-upon by their advisor M & R.

Finally, I am of the view that the definition of "membership interest" in the Plan is incorrect and misleading (including for the reasons articulated by David Schiff), further leading me to believe the Plan, as presented to policyholders, is unfair and prejudicial to their interests.

Date: November 8, 1999