NATURE AND COMPUTATION OF LABOR TURNOVER The question of labor turnover is of comparatively recent origin. One of the first references to it was made only a few years ago by Professor John R. Commons, in an article on the "Wage Earners of Pittsburgh," wherein he tells of an establishment (a machine works) which "in a single year of continued prosperity, 1906 .. hired 21,000 men and women to keep up a force of 10,000." In the same article Commons quotes a Pittsburgh employer to the effect that "2,000 hirings in a year for 1,000 permanent positions was not an exaggerated index of labor's mobility in the Pittsburgh district." Labor turnover represents a definite economic loss to employer, employee, and society at large. The employer's loss is due to (1) the clerical cost incidental to discharge and replacement; (2) the cost of training newly hired employees; (3) the wastage, breakage, etc., attributable directly to the inexperience of new employees; (4) the maintenance of idle machinery and equipment while separations are being replaced; and (5) the decrease in general efficiency of the organization due to the presence of large proportions of green hands. Furthermore a large turnover involving, as it must, the presence in the organization of considerable numbers of inexperienced hands has a definite tendency to increase the frequency and severity of industrial accidents. A considerable extent of the unemployment, even in normal times, may be said to be due to the constant shifting of workers from position to position. Such shifting necessitates, of course, temporary periods of idleness frequently injurious to the interests of the worker. The fact that society at large cannot possibly be the gainer from a 1 Charities and the Commons, XXI (March 6, 1909), 1054. 3 2 For estimates of the cost of labor turnover see (a) Magnus W. Alexander, "Hiring and Firing, Its Economic Waste and How to Avoid It," American Industries (August, 1915), p. 18; and (b) W. A. Grieves, "The Handling of Men," a paper published by the Executives' Club, Detroit Chamber of Commerce. 3 Chaney and Hanna, "The Safety Movement in the Iron and Steel Industry," Bulletin 234, U.S. Bureau of Labor Statistics, p. 131. constant shifting of labor needs no demonstration. There is no doubt that the general efficiency of our national industrial organization is diminished appreciably by large volumes of labor turnover. Our entrance into the war, necessitating, as it did, the fullest possible development and most efficient application of our industrial powers, brought the labor-turnover question to the fore. To increase productivity it became necessary to reduce the turnover of labor. With this came the imperative necessity of defining the nature of labor turnover and of devising correct methods of measuring its extent. Labor turnover is caused by separations which necessitate the employment of more than one person per position per year. By position in this connection is meant an extent of working time equivalent to the hours worked by one steadily employed person per year. The excess of the number of persons employed over the number of available positions represents a correct measurement of the extent of labor turnover. Table I, taken from a recent study of the author and presented herewith, shows the actual application of this method of measuring labor turnover. The number of available positions is shown in the table in column 4, entitled "Equivalent Number of Full-Time Workers." The figures of this column were arrived at by dividing the total annual labor hours of each occupation by a number (of hours) equivalent to the full annual working time of one permanently employed person. The footnote to column 4 indicates in detail the method followed in arriving at the total annual working hours of one permanently employed person. The figures in column 5 were arrived at as a result of an occupational census which covered the year in question. A somewhat less satisfactory method of measuring labor turnover is found in comparing the number of available full-time positions, frequently referred to as the standard number of jobs, with the total of separations. This method was used in Table II, presented herewith. The number of full-time positions, or the standard number of jobs, shown in column 2 of Table II was arrived at in a manner similar to the one used in the study of labor turnover in the Cleveland cloak, suit, and skirt industry, as shown in Table I. The annual number of hours of labor of one full-time worker (the labor hour equivalent of one fulltime position) was also computed in the manner shown in Table I. The difference between the method used in Table I and the method applied in Table II lies in the determination of the number of persons employed during the year to fill the available TABLE I1 LABOR TURNOVER IN THE CLEVELAND CLOAK, SUIT, AND SKIRT INDUSTRY FOR THE YEAR ENDING APRIL 1, 1918 Taken from the Monthly Labor Review, U.S. Bureau of Labor Statistics, August, 1918, p. 6. *The equivalent number of full-time workers was arrived at by dividing the aggregate of hours actually worked by all employees by the number of hours worked by one employed all the year round, termed a full-time worker. The number of annual hours of a full-time worker (2,200) was arrived at by multiplying the prevailing weekly hours of labor (48) by the annual number of weeks (52) and deducting 56 hours for seven legal holidays, 48 hours, or one working week, for an annual vacation, 122 hours (about 5 per cent of total working time) for temporary disability, and 70 hours, or about nine working days, for the fuelless days of the past year. The allowance for temporary disability was based upon the actual record of one of the larger firms, which record showed a percentage of time lost by reason of temporary disability of slightly below 5. |