COMPARISON OF TWO CALCULATIONS. At first glance the second calculation seems remarkably large when compared with the earlier one, but the difference in the amounts is readily accounted for, and the results are, in fact, when analyzed, astonishingly similar, considering the differences in the data employed in making the calculations. There are three reasons why the last result should be larger than the first: (1) The last calculation includes 67,198 more employees than does the first. (2) The last calculation is based on earlier ages of retirement, 60 and 65 having been adopted for two groups of employees, with 70 years for the third group, which naturally increases the cost above the first calculation, based on only one age of retirement, that of 70 years. (3) The last calculation is based on two mortality tables, the Combined or Actuaries' Table and the American Experience Table. The former, which was used in making the last part of the last calculation, shows a slightly longer expectation of life from age 70 than the latter, which was employed as the basis for the whole of the first calculation, and the use of the Combined or Actuaries' Table in connection with the American Experience Table has therefore raised the cost somewhat in the second calculation. GREATER NUMBER OF EMPLOYEES INCLUDED IN LAST CALCULATION. The greater number of people included in the last calculation is explained by two facts, first, the increase in the size of the service, owing both to the Government's enlarged activities and the growing tendency to bring into the classified service new groups of employees; and, second, the inclusion in the inquiry of persons not classified but reported as classified, and also of persons reported as unclassified but included as classified because investigation showed them to belong to the classified service. Not all the employees of the civil government of the United States were included in these computations. On July 1, 1907, according to the Official Register, the executive civil service, exclusive of the consular and diplomatic service, furnished employment to 286,902 persons. Of this number 101,028 were omitted from the statistics of Census Bulletin 94. In a few instances this was because the returns were received too late for tabulation, but more often it was because certain classes of employees are peculiar in some technical respect, such as the character of their appointment or the method of their compensation, and thus the inclusion of data concerning them would have tended to destroy the comparability of the figures for classes which are in technical respects alike. The most important of the classes omitted for the latter reason include 62,663 postmasters, 18,376 mechanics and laborers at navy yards and naval stations, 12,850 clerks in post offices not having free delivery, and 1,031 occasional employees of the Weather Bureau. Data for 4,584 employees of the Isthmian Canal Commission at work on the Isthmus were too incomplete to be included. From this statement of the most important classes omitted from Bulletin 94 it will be seen that these omissions did not prevent the use of the data collected for the bulletin in computing the cost of annuities for back services under the proposed plan, for the reason that it is not proposed to extend the plan to those classes. The first estimate was based on Table 67 of Census Bulletin 12, covering the executive civil service as of June 30, 1903, and embraced 103,030 employees. In preparing the present estimate the cards used in the preparation of Census Bulletin 94 showing the executive civil service as of June 30, 1907, were used, and covered 170,228 employees. Each of the 185,874 employees included in Bulletin 94 was represented in the Bureau of the Census by a tabulation card showing, in addition to other matter, the age, years of service, rate of compensation, and occupation of the employee. All the information was thus available for calculating the maximum cost of annuities for back services. In examining these cards to ascertain how many of these employees in the executive civil service were also included in the classified civil service, it was found that a considerable number of employees did not understand their technical status in the civil service and did not answer accurately the question as to whether they were classified or unclassified. Charwomen, for instance, frequently made the mistake of reporting themselves as classified. On the other hand, a considerable number of railway postal clerks reported themselves, erroneously, as unclassified. This seems, in the latter case, incredible, because the railway postal clerks were the first group of Government employees to be placed under civil service rules and their status has never been changed. The error did, however, occur, for the reason possibly that old clerks who came into the civil service under blanket rule realized only that they had never taken an examination, and supposed, therefore, that they had not been classified. From the cards it was impossible to tell the names of the individual represented by the card except on tracing the number of the card back to the original schedule made by the individual, but as it would have required months of time to do this, it was decided to include the cards of all employees who reported themselves as classified, and to make an individual separation of the cards of those who reported themselves as unclassified. These decisions were regarded as being in harmony with the general rule, pursued throughout the computations, that all doubtful cases of whatsoever sort were to be decided in favor of the alternative which would make the calculation of the cost of annuities for back services a maximum figure. In accordance with this rule, 170,228 employees were included in the computation, 67,198 more individuals than were taken account of in the earlier calculation. The classes excluded from the two calculations are as follows: Classes excluded from the first calculation. Classified employees in the District of Columbia (retirement age, 70)__ 22, 273 Classified employees elsewhere than in the District of Columbia (retire ment age, 70) ‒‒ 102, 464 124, 737 Less employees in the District of Columbia who had been in the service less than one year---. 1,781 Less employees elsewhere than in the District of Columbia who 19, 712 163 51 21, 707 103, 030 Number of persons included in the first estimate of cost of an- Not all of the employees of the civil government were included in the last calculation. Of the 286,902 persons shown in the Official Register, 170,228 were included, as follows: Classes excluded from the second calculation. According to the Official Register of July 1, 1907, the executive civil service, exclusive of the consular and diplomatic service, furnished employment to----- In preparing the estimate of cost of annuities for back services, the following groups of persons were omitted: 286, 902 Groups included in Official Register, but omitted from In preparing the estimate of cost of annuities for back services, the following groups of persons were omitted-Continued. Groups included in Official Register, and in Bulletin 94, but omitted from present computation-Cont'd. Firemen_. Cooks and bakers.--. Physicians and surgeons.. Farmers---- Domestics and waiters___ Stewards and quartermasters--- Heads of departments, bureaus, and independ ent offices__-- Other miscellaneous classes____ 502 441 371 254 236 201 159 115 2, 218 15, 646 116, 674 Number of persons included in the last estimate of cost of annui- EARLIER RETIREMENT AGES IN LAST CALCULATION. - 170, 228 The number of employees included in the calculation to be retired at the various ages of retirement follows: Railway postal clerks (retirement age 60) – Letter carriers and rural carriers (retirement age 65). All other employees (retirement age 70). Having the total number of employees shown in the first calculation (103,030), the total number of employees shown in the second calculation (170,228), and the aggregate annuities payable to the first group ($66,985,778) as a basis for the calculation, the aggregate annuities that should be shown by the last calculation may be determined, assuming, of course, that the first calculation is correct, that the age distribution has remained the same, and that salaries on the average have remained the same. Since no interest factor was used in calculating the cost of the annuities for back services, the expectation of life under the American Experience Table at the various retirement ages may be used as the value of an annuity of $1, and by multiplying the number of employees to be retired at the various ages by this value of an annuity of $1 at the corresponding retirement age, and then dividing the sum of these by the result of a similar computation covering the employees included in the first calculation, a result will be obtained which shows the ratio of the amount of total annuities payable for past services under the first calculation to the amount the last calculation should show. As the annuity payments under the last calculation are to be made quarterly, first payment in three months after the employee reaches the age of retirement, the annuity values (13.645, 10.845, and 8.415) used in the first part of the following calculation, covering the last estimate, are the curtate expectation of life for the respective retirement ages of 60, 65, and 70, according to the Combined or Actuaries' Table of Mortality, and correspond to annual payments when the first payment is to be made in one year, with 0.375 added to cover the additional cost of annuity payments made quarterly, first payment in three months, instead of at the end of the year. The first calculation was based throughout on the American Experience Table of Mortality, 1x+, first payment immediate, on the assumption that, on the average, the employees would retire at the age of 70 years. The complete expectation of life under this table at the age of 70 is approximately 8.25 years. The calculation may therefore be stated as follows: This difference might be reduced by carrying the decimals out farther, but since the difference is but slightly over seven one-hundredths of 1 per cent, further refinement would seem to be a waste of time. PROBABLE COST MUCH LESS THAN MAXIMUM. This total maximum sum of $130,581,273 is, however, only the possible total cost of putting the proposed plan into operation and applying it to the whole number of employees included in this inquiry; that is, to 170,228 persons. It is not the probable cost, which is greatly less. There are three reasons why the actual cost will be much less than the sum of $130,581,273 during the course of the next 78 years. (1) This sum allows for mortality, but not for resignations. (2) This sum allows for annuities based on present salaries, whereas the bill provides for annuities for back services based on the average salary received in the past (11 per cent of total compensation during entire period of service), which in the majority of cases is considerably less than the present salary. |