inventory-taking during the November pay period. The ice cream industry reported a seasonal decline in employment of 8.6 per cent, and the men's clothing and the millinery and lace goods industries reported decreases of 9.5 per cent each in number of workers. The women's clothing industry reported 8.3 per cent fewer employees; machine tools, 7.7 per cent; cement, 7.4 per cent; carpets and rugs, 6.3 per cent; stoves, 6.7 per cent; glass, 5.7 per cent; and shirts and collars, 5.4 per cent.

Comparing the indexes of employment and earnings in November, 1931, with the index numbers of November, 1930, for each of the 54 industries, decreased employment and pay-roll totals are shown in each industry. The chewing and smoking tobacco industry showed practically no change in employment over the year interval, while decreases of 2.5 per cent or less were reported in the confectionery, cotton goods, hosiery and knit goods, men's clothing, and automobile-tire industries. The outstanding decrease in both employment and pay-roll totals was shown in the agricultural-implement industry, in which the level of employment in November, 1931, had declined 52 per cent over the year interval, and pay-roll totals were 56.6 per cent below the level of the corresponding month of 1930. Employment in the machine-tool industry had decreased 38.2 per cent over the year interval, and the fertilizer and the piano and organ industries reported a falling off of slightly less than 37 per cent in number of workers between November, 1930, and November, 1931. The automobile industry reported a loss of 26.5 per cent in employment over the 12-month period; the iron and steel industry reported 18.6 per cent fewer employees in November, 1931, than in November, 1930, and the decline in employment from November, 1930, to November, 1931, in the foundry and machine-shop products industry was 24.5 per cent.

In the following table is shown the number of identical establishments reporting in both October and November, 1931, in the 54 manufacturing industries on which the bureau's indexes of employment and pay-roll totals are based, together with the total number of employees on the pay rolls of these establishments during the pay period ending nearest November 15, and the amount of their weekly earnings in November, the per cents of change over the month and the year interval, and the index numbers of employment and payroll totals in November, 1931.

The monthly per cents of change in employment and earnings for each of the 54 separate industries are computed by direct comparison of the total number of employees and the amount of weekly earnings in identical establishments for the two months considered. The per cents of change over the year interval in the separate industries, the group indexes, and the general indexes are computed from the index numbers of employment and pay-roll totals. The per cents of change over the month interval in the several groups and in the total of the 54 manufacturing industries are computed from the index numbers of these groups, which are obtained by weighting the index numbers of the several industries in the groups by the number of employees or wages paid in the industries.

TABLE 1.-COMPARISON OF EMPLOYMENT AND PAY-ROLL TOTALS IN IDENTICAL MANUFACTURING ESTABLISHMENTS IN OCTOBER AND NOVEMBER, 1931, PER CENT OF CHANGE OVER A YEAR INTERVAL, AND INDEX NUMBERS OF EMPLOYMENT AND PAY-ROLL TOTALS, NOVEMBER, 1931

TABLE 1.-COMPARISON OF EMPLOYMENT AND PAY-ROLL TOTALS IN IDENTICAL MANUFACTURING ESTABLISHMENTS IN OCTOBER AND NOVEMBER, 1931, PER CENT OF CHANGE OVER A YEAR INTERVAL, AND INDEX NUMBERS OF EMPLOYMENT AND PAY-ROLL TOTALS, NOVEMBER, 1931-Continued

The trend of employment and earnings in 31 additional manufacturing industries, surveyed but not yet included in the bureau's weighted indexes of employment and pay-roll totals, is shown in Table 2 following. The combined total of these industries shows a decrease of 1.4 per cent in number of employees from October to November and a loss of 4.7 per cent in pay-roll totals. The per cents of change for the combined total of these industries are unweighted and represent only the changes in the total number of establishments reported. These 31 industries have been added to the bureau's employment survey at various times since February, 1929. Information for the index base year (1926), however, is not available, and therefore these industries can not be combined with the 54 manufacturing industries upon which the bureau's indexes of employment and earnings are based.

Nine of the thirty-one industries in this group reported gains in number of employees in November, as compared with October, and six industries showed increased weekly pay-roll totals. The beetsugar and cottonseed products industries, which reported the outstanding increases in employment last month, again reported the largest gains in employment from October to November, due to the seasonal expansion at this time of year in these industries. The iron and steel forgings industry reported an increase of 6.6 per cent in employment over the month interval, and the typewriters and supplies industry reported a gain of 3.0 per cent. The greatest de

crease in employment was reported in the fur-felt hat industry in which employment declined 10.6 per cent from October to November. Seasonal decreases in number of workers in the beverage and the marble and other stone products industries of 8.3 per cent and 7.9 per cent, respectively, were reported in November, as compared with October, and decreases ranging from 6.7 per cent to 5.4 per cent were reported in radio, men's furnishing goods, jewelry, and aluminum manufactures. The decreases in employment in the remaining 15 industries were 3.7 per cent or less. A comparison of employment and pay-roll totals over the year period is available for 9 of these 31 industries. Decreases in both employment and earnings are shown in this yearly comparison in each of these 9 industries, the rayon industry reporting the smallest loss in employment from November, 1930, to November, 1931, 1.3 per cent, while the greatest decrease in employment over the year interval, 41.7 per cent, was shown in the radio industry.

TABLE 2.-COMPARISON OF EMPLOYMENT AND PAY-ROLL TOTALS IN NOVEMBER, 1931, WITH OCTOBER, 1931, AND NOVEMBER, 1930, IN SPECIFIED MANUFACTURING INDUSTRIES, FOR WHICH DATA FOR THE INDEX BASE YEAR (1926) ARE NOT AVAILABLE

The total number of employees and amount of earnings in the 31 industries shown in the foregoing table have been combined with the totals of the 54 manufacturing industries shown in Table 1 in presenting the total of all manufacturing industries in the summary table.

In the following table is presented a recapitulation of the data by geographic divisions for the combined 85 manufacturing industries. Employment and pay-roll totals in the Mountain geographic division showed an increase from October to November due to the seasonal expansion in the beet-sugar industry. The remaining geographic divisions reported decreases in both employment and earnings over the month interval, the South Atlantic division reporting the smallest decline in employment, 0.9 per cent, and the West North Central division reporting the greatest falling off in number of workers, 4.7 The level of employment and earnings in each of these nine divisions in November, 1931, as compared with November, 1930, shows a considerable decrease over the year interval, the decrease in earnings in each instance being more pronounced than the decline in employment.

TABLE 3.-TREND OF EMPLOYMENT AND PAY-ROLL TOTALS IN MANUFACTURING ESTABLISHMENTS BY GEOGRAPHIC DIVISIONS, ESTABLISHMENTS REPORTING FOR OCTOBER AND NOVEMBER, 1931, NUMBER OF EMPLOYEES AND WEEKLY EARNINGS IN NOVEMBER, 1931, AND PER CENTS OF CHANGE OVER THE MONTH AND YEAR INTERVAL

1 Weighted per cent of change for the combined 54 manufacturing industries, repeated from Table 1, manufacturing industries.

Per Capita Earnings in Manufacturing Industries

ACTUAL per capita weekly earnings in November, 1931, for each of the 85 manufacturing industries surveyed by the Bureau of Labor Statistics, together with per cents of change in November, 1931, as compared with October, 1931, and November, 1930, are shown in Table 4.

Per capita earnings in November, 1931, for the combined 54 manufacturing industries of the United States, upon which the bureau's indexes of employment and pay rolls are based, were 2.3 per cent lower than for October, 1931, and 12.7 per cent less than for November, 1930.

The actual average per capita weekly earnings in November, 1931, for the 54 manufacturing industries were $20.68; the average per capita earnings for all of the 85 manufacturing industries surveyed were $20.75.