eek in January and February to 2.9 hours by Tovember. During the final 3 months of the year, vertime was the highest for the season since 1956. For 1961 as a whole, the workweek averaged 9.8 hours, not significantly different from the 9.7 hours of 1960. These averages, of course, bscure the sharp contrast in the situation at the espective yearends.

Hourly earnings for production workers in manufacturing averaged $2.32 in 1961, 6 cents or bout 2 percent higher than in 1960. Aside rom 1954, this was the smallest annual percentage ncrease since World War II.

Weekly earnings in manufacturing averaged a ecord $92.34 during 1961, 3 percent or $2.62 igher than in 1960. In the later year, earnings veraged above $100 per week for the first time n the durable goods sector, and averaged $83.01 n the nondurable goods sector. All major manuacturing divisions reported gains over the year, anging from more than $5 in petroleum refining -o $1 in furniture and apparel. In percentage erms, increases ranged from more than 4 percent n the petroleum, paper, tobacco, primary metals, and ordnance industries to about 11⁄21⁄2 percent in urniture, transportation equipment, and printing ind publishing.

Gains in average weekly earnings prevailed in he nonmanufacturing industries as well in 1961. The larger increases, of roughly $4 to $5, were egistered in contract construction, power and communications utilities, and transportation. Inreases in trade and service industries were generally around the $2 mark. Comparatively mall gains ($1 or less) were recorded in the elatively low-paying laundries and cleaning and lyeing plants group, as well as in coal mining. Earnings levels in major industries ranged from 3117.37 in contract construction to $49.28 in aundries and cleaning and dyeing plants.

Several price developments during 1961 affected earnings as well as purchasing power. Consumer prices were only 1.1 percent higher on the average han in 1960, continuing their relative stability of recent years. Thus, most of the increase in earnings represented a gain in purchasing power. On the other hand, because of the relative sta

See "Deferred Wage Increases and Escalator Clauses," Monthly Labor Review, December 1961, pp. 1319-1323.

bility of prices, they had little direct effect on wages under the escalation provisions of collective bargaining agreements.*

In 1961, the average production worker in manufacturing with three dependents had a record take-home pay (gross pay less withheld Federal income tax and social security deductions) of $82.18 per week, $2.07 and 2.6 percent higher than in 1960 (table 2).

In industries other than manufacturing, the level of spendable earnings was $102.98 in contract construction, $66.85 in trade, and $94.35 in mining.

Long-Term Trends in Industry Employment

Since 1947, employment in nonfarm establishments has increased from 43.9 million to 54.1 million in 1961 (table 3). Most of the 10.2 million expansion in jobs has been in response to the dramatic rise in demand for services, both private and public, and in the distribution, as distinct from the production, of goods. Most of these service-producing industries have grown

Except for cutrelatively small

persistently from year to year. backs in transportation and declines in trade employment, recessions have only slightly dampened growth in the service sector. (See chart 4.)

In the goods sector, employment has fluctuated widely without showing a persistent trend; employment in 1961 was only moderately higher than in 1947, and somewhat below the postwar high in 1953. The lack of growth in this area, together with the tremendous expansion in services, has produced a marked shift in the industrial composition of employment over the postwar period; the service industries accounted for 64 percent of all nonfarm payroll employment in 1961, compared with 58 percent in 1947. Highlights of trends in particular industry divisions are summarized in the remainder of this section.

Government employment expanded by 3.4 million or 60 percent from 1947, to 8.8 million in 1961. State and local units accounted for nearly all of the advance, largely because of the growing requirements of school systems.

Employment in the service industries division increased by 50 percent, from 5.1 million in 1947 to 7.5 million in 1961. Growth has been registered every year, although by smaller amounts in recession years.

Finance, insurance, and real estate establishments have added 1 million employees since 1947, bringing the total to 2.7 million in 1961. While

accounting for only 5 percent of nonfarm pay employment in 1961, this small complex of servi type industries has accounted for a tenth of gain in total payroll employment since 19

Jobs in trade increased from 9.0 million in 194 to 11.4 million in 1961, a gain of more than percent. However, the rate of growth in th industry has slowed in recent years compare. with earlier in the postwar period.

Although manufacturing industries employe 16.3 million workers in 1961, or about 700,000 more than in 1947, it is difficult to discern definite trend over the postwar period. Althoug it experienced a major recession in 1949, mar facturing expanded from 1947 to a postwar hig in 1953 under the stimulus of the Korean conflict After a decline during the 1954 recession, manfacturing employment nearly regained its 19 level in both 1956 and 1957. However, it dey clined sharply again during the 1958 downtur and its subsequent recovery into 1960 left th annual average about 400,000 short of the 1956-5) level. Under the impact of the recent recession. manufacturing jobs showed a further loss c 500,000 in 1961.

The lack of a definite pattern in manufacturing as a whole reflects both the cyclical volatility and the diverse secular trends of its individual ind tries. Jobs have tended to expand over the lor run in electrical equipment, paper, chemicals, and printing and publishing. In 1961, these industries accounted for nearly 3.8 million, or one-fourth e

manufacturing employment. On the other nd, there have been longrun declines in the tile and lumber industries, and in recent years, food processing and transportation equipment well.

Changes in manufacturing processes over the stwar years have had a dramatic impact on the nposition of factory jobs. Production work

those directly involved in manufacturing ocesses have tended generally to decline in mber. On the other hand, nonproduction rkers, those performing services and other pporting functions have increased as a proporn of total employment in every major manuturing industry. Their number has increased m 2.6 million in 1947 to 4.2 million in 1961, and eir proportion from 16 percent to 26 percent of manufacturing jobs.

Construction employment expanded-with inruptions-from about 2 million following World ar II to 3 million in 1956 and 1957. It has ce then shown no growth and recently has own some tendency to decline. Construction ployment was unusually slow to recover from e recent recession, and in 1961, averaged 100,000 wer than in 1960.

Jobs in the transportation and public utilities dustries fluctuated within a comparatively rrow range over the postwar years to 1957. recent downtrend in these industries brought e 1961 level 300,000 below the comparatively gh levels of 1956-57. Gains in truck and air ansportation have tended to offset declines in ilroad and communications employment. There has been a pronounced decline in mining ployment since the end of World War II, flecting both a slackening in the demand for al and dramatic changes in mining operations.

1961, mining employment totaled about 0,000 compared with close to 1 million in 1947 id 900,000 during the Korean conflict. The wntrend has been primarily in coal mining; troleum extraction, included in the mining dustry, has generally grown in the postwar riod.

The seasonally adjusted rates are based on adjustment factors derived n data through the end of 1961 and published in the February 1962 issue Employment and Earnings. The seasonal adjustment procedure provides annual review of the historical data and revision of the seasonal factors previous years where the later data indicate new patterns. The method described in "New Seasonal Adjustment Factors for Labor Force Comlents," Monthly Labor Review, August 1960, pp. 822-827.

As 1961 began, the rate of unemployment in the civilian labor force was 6.7 percent (seasonally adjusted) after rising steadily for over half a year." (See chart 5.) It continued to move up slowly during the first part of the year to a high of 7.0 percent in May. The rate remained near 7.0 percent until the last 2 months in 1961, when it fell sharply. At the end of the year, the unemployment rate was at 6.0 percent, its lowest level in 15 months. Although the overall rate of joblessness was between 6.7 and 7.0 percent throughout the first 10 months of 1961 and improved less rapidly than other indicators, there was some improvement during this period among men 25 years of age and over, who constitute over half of the labor force. Their unemployment rate reached a high for the year of 5.7 percent in May,

Beginning in January 1960, data include Alaska and Hawaii.

ent rate in durable goods manufacturing at 8.4 rcent was, however, below the 10.5 percent ached in 1958, but the unemployment rates in ade (7.2 percent) and finance and service (4.6 rcent) were higher than in 1958.

Wage and salary workers from goods-producing dustries (including agriculture) and from transrtation accounted for about 49 percent of total employment in 1961, compared with 56 percent 1958. On the other hand, the proportion of employed wage and salary workers from serviceoducing industries was higher in 1961 (36 versus percent) and new workers rose from 9 to 12 rcent of total unemployment. Self-employed d unpaid family workers accounted for 3 percent unemployment in both years.

The higher unemployment rate in trade comred with 1958 was reflected in an increase from O to 4.7 percent in the rate for sales workers, the ly white-collar group whose rate was appreciably gher in 1961. Among blue-collar workers, aftsmen and operatives had rates well below the

1958 level, although substantially over 1960. The unemployment rate for nonfarm laborersat 14.5 percent, the highest for any occupation group-was about the same as in 1958.

Although white-collar workers' unemployment rates were generally no higher, they accounted for a somewhat higher proportion of unemployment in 1961 than in 1958-21 versus 18 percent-with most of the increase in clerical and sales occupations. Workers in service occupations also accounted for a slightly larger proportion of unemployment in 1961-about 13% percent-compared with 1958. Blue-collar workers, at 50 percent of total unemployment in 1961, were down several percentage points from 1958, with operatives accounting for the larger part of the decrease.

On the whole, the differences in the composition of unemployment in 1961 and 1958 can be attributed both to the changing composition of the labor force and to the smaller impact of the recent recession on men in the central age groups and