Changes in Occupational Ranking and Hourly Earnings, 1997-2005
by John E. Buckley
Bureau of Labor Statistics

Originally Posted: August 29, 2007

In June 2005, full-time workers in private industry and State and local governments averaged $19.70 per hour, according to findings from the ninth annual BLS National Compensation Survey (NCS). When the NCS series began in 1997, the comparable average was $15.77 per hour.¹

The National Compensation Survey is an establishment-based survey of a sample of 152 metropolitan and nonmetropolitan areas.² The sample represents the Nation's 326 metropolitan statistical areas (as defined by the Office of Management and Budget in 1994) and the remaining portions of the United States. Agricultural, private household, and Federal Government workers are not included in the survey. The 2005 NCS results mark the last observation before a break in the series. The upcoming 2006 national estimates will be based on the 2000 Standard Occupational Classification System, which replaced the 1990 Occupation Classification System (OCS), and the 2002 North American Industry Classification System, which replaced the 1987 Standard Industrial Classification (SIC) system. This article, using 2005 data, ranks 418 occupations by average hourly earnings and summarizes changes in wages for the same occupations over the 1997-2005 period.

Twenty top ranked occupations

Twelve of the 20 occupations with the highest hourly pay in 1997 also were in the top 20 in 2005, although there were some changes in ranking. (See table 1.) Four occupations retained the same rank:

• 1st: Airline pilots and navigators,



• 8th: Physics teachers,



• 11th: Medical science teachers, and



• 14th: Dentists.

Among the 12 occupations that remained in the top 20 over the period, physicians had the largest upward change in ranking, from 12th place in 1997 to 4th place in 2005. The median percent change in hourly pay for these 12 occupations was about 32 percent, ranging from a decrease of 6.1 for law teachers to an increase of 63.6 percent for physicians.

Average earnings of chief executives and general administrators, public administration, advanced only 1.9 percent from 1997 to 2005, resulting in the largest downward change--from 3rd to 10th place. Although managers and administrators, not elsewhere classified--the private industry counterparts of the public executives--were ranked 26th in 2005, their average hourly earnings increased about 24 percent. Chief executive pay has been a contentious topic in recent years and is currently being debated in Congress.³ The pay packages being questioned are often associated with executives in large national or multinational firms, where stocks or stock options are the main component of compensation. Data collected in the NCS include salaries for executives in small, medium, and large establishments, but exclude compensation in the form of stocks or stock options and nonproduction bonuses.

Table 2 presents a list of eight occupations that advanced to the top 20 from 1997 to 2005, as well as the eight occupations that dropped out of the top 20 over the period. Pharmacists had the most notable upward change in top 20 ranking, moving from 57th in 1997 to 18th in 2005. Engineering teachers also moved up sharply, from 22nd to 5th place, aided by a 78.7-percent increase in average hourly earnings. Social scientists, not elsewhere classified, had the highest percent increase in average earnings (118 percent). This and four other occupations with at least an 80-percent increase in average earnings were still below the earnings threshold of $44.49 per hour needed to advance to the top 20.

Two occupations with relative standard errors of less than 10 percent in both 1997 and 2005 dropped in rank over the period: chemistry teachers went from 17th in 1997 to 40th in 2005, and theology teachers went from 10th in 1997 to 29th in 2005. Other occupations had larger declines in rank, but their relative standard errors were considerably higher, making direct comparisons less meaningful.⁴

Ranking by hourly and annual earnings

In addition to publishing average hourly earnings estimates, the NCS also publishes annual salary estimates. Ranking occupations by hourly and annual earnings produces different results. (See table 3.) For example, in 2005, airline pilots and navigators, who placed first when ranked by hourly earnings, dropped to third place when ranked by annual earnings. The number of annual work hours is considerably lower for airline pilots and navigators than it is for other occupations ranked in the top 20.⁵ A difference in ranking would occur, for example, when two occupations have the same annual salaries (say $50,000) but one has a 40-hour weekly work schedule (2,080 annual hours) and the other has a 35-hour schedule (1,820 annual hours). On an annual basis they would have an equal rank, but, on an hourly basis, workers with the 35-hour schedule would rank higher.

Other positional changes among the top 20 occupations (when comparing hourly and annual wages) were substantially affected by college and university teachers. In terms of hourly earnings, 12 of the top 20 positions in 2005 were held by these kinds of teachers. But college and university teachers tend to work fewer annual hours than other workers, and, as a result, they placed lower in the rankings based on annual salaries. Agriculture and forestry teachers averaged considerably fewer annual work hours--about 1,346 in 2005--than others in the top 20 occupations (except for pilots and navigators). When comparing rankings based on average hourly and annual earnings, the ranking of agriculture and forestry teachers was 6th place on an hourly basis and 17th place on an annual basis.⁶

Twenty lowest ranked occupations

In 1997, the average hourly earnings of the 20 lowest ranked occupations ranged from $4.29 per hour for waiters and waitresses to $8.23 per hour for nursing aides, orderlies, and attendants. In 2005, the earnings of the 20 lowest ranked occupations ranged from $4.71 per hour for wait staff to $10.05 per hour for agricultural products inspectors and vehicle washers and equipment cleaners. The low earnings for waiters and waitresses may be misleading because the NCS is designed to measure employers' costs for wages and salaries, which precludes collecting tips as part of wages. As a result, total earnings for waiters and waitresses are understated. Rates for some other low ranked occupations--such as bartenders, baggage porters and bellhops, parking lot attendants, and taxicab drivers and chauffeurs--were similarly affected by the absence of information on tips.

Among the 20 lowest ranked occupations in 1997, 14 were in the bottom-20 category in 2005. Of the 6 occupations that moved out of the lowest 20, 5 remained near the bottom. The one exception was hand engraving and printing occupations, whose average hourly earnings more than doubled over the period (from $8.13 to $16.75), moving this job's ranking from 400th in 1997 to 240th in 2005.

Percent changes in hourly earnings, 1997-2005

Among the 418 occupations for which comparisons could be made, 227 had increases in hourly earnings of at least 25 percent over the 1997-2005 period. (See table 4.) Social scientists, not elsewhere classified, an occupational category that includes historians, anthropologists, and political scientists, had the highest percent increase in earnings--from $13.30 per hour in 1997 to $28.99 per hour in 2005, an increase of 118 percent. Two other occupations, hand engraving and printing and elevator installers and repairers, also had average hourly earnings that doubled during the period.

The 191 occupations with changes in average hourly earnings of less than 25 percent included 21 occupations whose reported average hourly earnings declined from 1997 to 2005. The occupations with declining average earnings from 1997 to 2005 included high-ranked jobs (law teachers' earnings dropped 6.1 percent), middle-ranked jobs (religious workers, not elsewhere classified, dropped 10.3 percent), and low-ranked jobs (garage and service station workers' earnings dropped 2.8 percent). The median decline for these 21 occupations was 5.9 percent.

John E. Buckley
Economist, Division of Compensation Data Analysis and Planning, Office of Compensation and Working Conditions, Bureau of Labor Statistics.
Telephone: (202) 691-6299; E-mail: Buckley.John@bls.gov

Notes

1 From 1997 to 2005, the wages and salary component of the Employment Cost Index (ECI) for civilian workers (that is, workers in private industry and State and local governments) increased approximately 29 percent. This differs from the change in NCS wage estimates from 1997 to 2005 because the ECI holds occupation and industry mix constant over time, whereas the NCS wage surveys provide a cross-sectional view of the economy at one point in time. For an estimate of real wage change over time, see Employment Cost Index Historical Listing, Constant-dollar 1975-2005 (December 2005 = 100), on the Internet at http://www.bls.gov/web/ecconst.pdf.

2 The 1997 survey was limited to 149 areas in the 48 contiguous States. Alaska and Hawaii were included in the 1998 and later surveys, and the sample increased to 154 areas from 1998 through 2003.

3 See "Current Controversies in Executive Compensation: Issues of Justice and Fairness," Knowledge@Wharton, May 2, 2007, on the Internet at http://knowledge.wharton.upenn.edu/article.cfm?articleid=1727.

4 The relative standard errors (RSEs) associated with each occupation's mean earnings provide an indicator of the reliability of the estimated mean. Basically, a low RSE indicates that the "true" mean is likely to be in a narrow range around the estimate. A high RSE indicates that the true mean is likely to be found in a broad range around the estimate, which indicates that the occupation's rank--the occupation's mean wage with respect to the mean wage of other occupations--is less meaningful. This can be illustrated by comparing two occupations with very close average earnings, but very different RSEs. The RSE for nuclear engineers, whose hourly wage was $39.93, was 5.7. Thus, at the 90-percent level, the confidence interval for this occupation was $36.19 to $43.67, a spread of $7.48. This indicates that approximately 90 percent of the time, the true population value would fall within that range. By contrast, the RSE for computer science teachers, whose hourly wage was $39.89, was 19.8, which means that at the 90-percent level, the confidence interval for this occupation was $26.90 to $52.88, a spread of $25.98.

5 Most airlines provide only flight pay and hours for pilots; hours worked in preparing for flights are not included, resulting in flight pay being overstated. Pay for other duties, such as training, which may be paid at a lower rate than flight pay, may also be excluded.

6 Due to the relatively high standard errors for agriculture and forestry teachers and some other occupations, users should exercise caution when making direct salary comparisons. This is especially applicable when the differences in average earnings are small.