from some source of a sum sufficient to make up the difference between their own savings and the amount required to purchase their annuities. The annual sum necessary would gradually increase for a few years, reaching its maximum about 30 years after the passage of the bill, but a few years after that the amount each year would fall off very rapidly until in about 50 years, when practically all now in the service would be dead, there would be no more need of appropriations. The plan would then be self-sustaining, and the condition of the civil service, so far as superannuation is concerned, nearly what it would be if a clean sweep of the service could be made and the plan inaugurated to-morrow with a complete list of new appointees. EXPERIENCE OF NEW ZEALAND. This difficulty was clearly perceived by the framers of the public superannuation act recently passed in New Zealand, as shown by the following press comment: Hitherto, no comprehensive scheme for the humane and equitable treatment during infirmity and old age of the servants of the State has been in operation in New Zealand, and critics of public affairs have complained of what they considered a glaring defect in the national life of the Dominion. But these critics had not fully realized the difficulties which presented themselves in the way of an application of the pension idea which would be at once financially sound and fair in its operation. As the Premier (the Rt. Hon. Sir Joseph G. Ward) pointed out in the course of a debate on the bill, the civil service in New Zealand is old compared with the age of the country, and it was because of that fact that there was a supreme difficulty on the part of the Government in putting on the statute books a superannuation act 40 years after some of the men had joined the service, the incidence of which was light in its burden upon members of the service.1 It is exactly the difficulty that confronts those who would devise a plan of retirement for the civil employees of the United States which shall be "at once financially sound and fair in its operation. The difficulty was met in New Zealand in the only way that can be devised without injustice to the whole body of employees, that is, by appropriation from the Government. While the New Zealand plan goes much farther than anything suggested in the plan under discussion in this report in the way of benefits (including benefits to widows and orphans), and has departed widely, in certain respects, from the savings bank idea, the necessity of keeping entirely separate the annuities paid on services rendered after the adoption of the plan and the annuities paid on services rendered prior to the adoption of the plan was clearly perceived, as shown by the following comments on their superannuation act: The pensions are liberal, and a scheme of this description, applying to present officers, many of whom can retire immediately on very fair pensions, must of 1 See The Insurance Record, London, Feb. 21, 1908. necessity entail a heavy liability on the part of any Government. The favorite method, however (vide New South Wales and Cape Colony civil-service schemes), has been to ignore this liability and to go on paying the pensions of the old men who retire, out of the contributions of the young men who join the scheme, until the funds are exhausted, and the outlay for pensions exceeds the income from contributions. The actuary, Mr. Morris Fox, has made the recklessness of this method quite apparent in his comprehensive reports, and the Government has agreed to start the scheme with an annual payment of £20,000, the subsidy to be increased by such further amounts as will be sufficient to pay the difference between the pensions falling due and the amount of pension the contributions would have actually purchased. (For example, if an old servant retires on £400 a year while his contributions would only have purchased £10, the fund pays the £10, and the Government finds the balance, £390 per annum.) The contributions of the younger members will therefore be accumulated at compound interest to help provide their pensions when they become payable, and will not be absorbed by meeting more immediate liabilities; the cost of providing current pensions being borne by the present taxpayers and not by posterity. If the scheme were commenced without contributions, the pensions falling due would be the measure of the Government's annual liability, and by meeting this liability (or rather the portion not paid for by the contributor) at once, Sir Joseph Ward has made a fair division of the annual outlay between present and future taxpayers. It is this simple, but ingenious, financial arrangement which differentiates the scheme from all others with which we are familiar, and the result is threefold; the solvency of the fund is secured, the present strain on the exchequer is the minimum compatible with soundness, and only a fair share of the liability is transferred to posterity. The Government is to be congratulated accordingly on having adopted this important actuarial recommendation.1 The necessity for keeping separate accrued and future liabilities was indeed well brought out in Mr. Fox's reports. There seems to have been no thought in his mind but that the whole of the liability for back services should be borne by the Government, a point which does not appear to have been questioned by the Government. As for future liabilities, it was impracticable, in view of the generous benefits provided under the scheme, for the employees to carry them alone and the help of the Government was expected with them also, but Mr. Fox insisted on the importance of reserving and accumulating the contributions to meet the contributors' portion of liability and not using them, in earlier years, to pay other claims which had not been provided for by contributions, namely, pensions to persons already in the service at the time of the establishment of the plan.2 COST DEPENDENT ON NUMBER OF EMPLOYEES INCLUDED. The cost of putting the proposed plan into operation depends, of course, entirely on the number of employees to whom the benefits of the plan are extended. If it were confined to the employees of the 1 See The Review (Sydney), Feb. 29, 1908. 2 See Civil Service Retirement in Great Britain and New Zealand (Senate Doc. 290, pp. 234-235). District of Columbia only it would cost very much less than if extended to the entire classified service, since the number of classified employees in the District is only 23,254 as compared with that of 170,228 included in the entire service. At the same time, as superannuation is very much greater in the District than it is outside the District, the need of a retirement plan is much more urgent in the District than it is outside. Census Bulletin 94 not only emphasizes the fact that superannuation among civil servants in the District is greatly in excess of superannuation among civil servants outside the District, but it also brings out in a general way that it is much greater among the civil servants in the District than it is among breadwinners throughout the country. The number of Government employees at least 65 years of age is 6,523. Of this number 1,852 are employed in the District of Columbia, and 4,671 elsewhere. Although less numerous in the District than elsewhere, employees of advanced age form a much larger proportion of the force in the District than they do of the force elsewhere. In the District practically one Government employee in 14 is at least 65 years of age, while elsewhere the corresponding figures are but about one in 34. Whether these figures represent any special tendency for Government employees to remain in service after persons in other walks of life would have retired, is, of course, the interesting question. Perhaps some light may be thrown upon it by comparing the age distribution of the Government employees in the District * * * 25 years of age and over with that of the breadwinners 25 years of age and over reported at the census of 1900. In the District of Columbia ** * ** 10 per cent of the male employees 25 years of age and over are at least 65. For the breadwinners in general the corresponding percentage is 6.3.1 The advisability of confining the proposed plan, in the beginning, to the District of Columbia can be urged not only on the grounds of less cost and greater gain where the need is greatest, but on the general principle that it is desirable to proceed slowly in the inauguration of new measures. The plan could gradually be extended by Congress to various classes of employees, as the wisdom of doing so was proved. In the proposed bill, section 15 reads as follows: SEC. 15. That the provisions of this act shall apply only to the classified civil service of the District of Columbia, which is hereby defined to include all officers and employees in the executive civil service of the United States in the District of Columbia, except persons appointed by the President and confirmed by the Senate, and unskilled laborers. No person serving in a position excepted from examination or registration as defined in the civil-service rules shall be included within the provisions of this act unless he has served in a competitive position for at least one year. Whenever any person becomes separated from the classified service by reason of appointment into the unclassified service, such separation shall not operate to take him out of the provisions of this act. The President shall have power, in his discretion, to exclude from the operation of this act any group of employees whose tenure of office is intermittent or of uncertain duration. 1 See Census Bulletin 94, pp. 12, 13. TWO CALCULATIONS OF MAXIMUM COST FOR ENTIRE CLASSIFIED SERVICE AGREE. Two calculations have been made of the cost of putting the plan into operation throughout the service. They virtually agree. The amounts of the two sums are different, but proportionately they are the same. THE FIRST CALCULATION. The first calculation was made under the direction of Mr. Benedict D. Flynn, assistant actuary of the Travelers Insurance Co., Hartford, Conn., for the Committee on Department Methods and was based on Table 67 of Census Bulletin 12, entitled "The Executive Civil Service of the United States," covering the classified employees as of June 30, 1903. The total number of employees included in that calculation was 103,030, and the maximum cost of paying them annuities for past services was found to amount to $66,985,778 in the course of 67 years, as shown by the following table: TABLE XXVII.-Showing maximum cost of annuities for back services for 103,030 employees. The second calculation, which was made under the direction of the author by the Bureau of the Census, was based on cards used in the compilation of Census Bulletin 94, entitled "Statistics of Employees of the Executive Civil Service of the United States," covering the classified employees as of June 30, 1907. The total number of employees included in this most recent inquiry was 170,228, and the |