H.—ll

8

Thus, if a policy-holder died of typhoid fever nine months after insurance, the claim could unhesitatingly be paid ; but if he died of inflammation of the lungs it would be necessary to provo that this was not merely an acute termination of a chronic consumption extending back beyond the date of insurance. 14. It is clear that by this method a much larger proportion of the premiums paid during the first two years would be reserved than of those paid subsequently. What should be done with the excess ? It is not required in order to meet the increased mortality of later years; the normal reserve (as fixed in section 5) would suffice for this. I therefore recommend that, in the case of each policy-holder, the excess in question, after deducting initial expenses of management, be accumulated at compound interest, and finally paid at his death in addition to the sum to which ho was normally entitled. This would give a slight extra advantage to the longest livers. It is important to observe, as the Secretary has pointed out to me, that many persons who would be declined if proposing for a whole-life policy in the Ordinary Branch would be allowed to take out a temporary insurance of a year or of two years, and could thus protect themselves, if they wished, during the probationary period. This would still further diminish the hardship of having such a condition, though it appears to me that the difficulty is already, to a considerable extent, met by my suggestion of paying in case of death by a disease the absolute commencement of which is proved to be later than the date of insurance. 15. I have hitherto assumed, in order to make the principle of the scheme more plain, that a division would take place at the close of each calendar year. Now, unless the number of lives at risk is very great—much greater than any number we can hope to secure—the death-rate will be sure to fluctuate enormously from year to year, though not as much as with an equal number of healthy lives, and the method of division referred to will give very uneven results. Especially will this be the case in the earliest portion of the history of the office. I would therefore propose that, as a matter of practice, divisions should take place triennially instead of annually. This would mean that at the end of each third calendar year the number of deaths during the three immediately preceding years would be ascertained (exclusive, of course, of deaths among policy-holders of less than two years' standing), and the divisible fund, determined in accordance with section 5, would be divided among the claimants in the manner explained in section 12. It would of course be necessary to make an interim payment immediately on proof of death. As the office grew older it would become safe to allow a tolerably large interim payment, say 60 or 70 per cent, of the average amount paid per "share during the preceding triennial periods; but at first, there being no previous experience to go upon, it would be necessary to fix the interim payment by guess, and I would suggest that it be fixed at one-fifth the amount of assurance which the corresponding premium would purchase in the Ordinary Branch. Thus, a person entering at thirty, and paying a premium of £2 4s. 3d. annually, would, in case of death during his fourth or fifth year of insurance, be entitled to an interim payment of £20. 16. I would throw open the Tontine Section to all males and females above the age of twenty resident in New Zealand. It is important to obtain as large a constituency as possible, in order to reduce the fluctuations to a minimum. 17. In case of failure to pay a renewal premium I should allow no cash surrender value. I consider cash surrender values objectionable in principle except at the end of a tontine period under the American plan ; but, of course, competition makes it necessary to allow them in the Ordinary Branch. The sum which had been reserved from the former premiums, after payment of share of past claims and expenses, should be regarded as a surrender value capable of (1) keeping alive the full share or shares for a certain limited period, and (2) of purchasing a certain fraction of a share free of further payments. Personally, I should prefer, in all cases after two full years' premiums had been paid, to keep alive the share for just six months, during which time, if ever, experience shows that revival will probably take place, and at the end of that time to apply the remainder of the surrender value to the purchase of a paid-up fraction of a share. What fraction of a share would be the equivalent of a given surrender value it would not for many years be possible to determine accurately, as, without knowing the rate of mortality to which the lives are subject, we cannot establish an accurate equivalence between single and annual premiums; but we should be erring greatly on the safe side if, for the present, we decided to convert the surrender value into an annual premium by ordinary mortality tables, and to apply this annual premium to the purchase of its appropriate fraction of a share. 18. It only remains to indicate the mode in which the cost of management would be met. Fortunately, this would be the simplest conceivable. Since there is no fixed sum assured, there can be no question of loading a pure premium, and there is no necessity for estimating in advance what the cost of management would be. The actual expense of each triennium would be deducted from the premium income accruing therein, and the residue would be treated as the pure premium to be operated on in accordance with section 5. If it is thought desirable to separate between the expenses of new and of old business the former could, as suggested in section 14, be deducted from the extra reserve made in the probationary years which it is proposed to accumulate at compound interest. In that case only the expense of old business would be deducted from the gross premium income of ordinary years. This would, on the whole, be fairer than the first-named plan, as under that there would be a tendency for the sums payable to claimants to fall during periods in which a large new business was transacted. But, as there is no medical examination, the cost of new business would be proportionately smaller than in the Ordinary Branch. F. W. Fbankland, 22nd May, 1883. Government Actuary.