ERI Blog

October marks the opening of ERI Salary Surveys’ data collection phase. Over the next six months, organizations throughout the U.S. and Canada will submit compensation data for use in ERI’s traditional industry-specific and job function salary surveys.   As our research team prepares incoming data for the 2013 reports (available in August), let’s examine some data trends from previous years.

Our analysis compares salaries between all 2011 and 2012 submissions for both nonprofit and for-profit reports. For each year, total averages were calculated by job title and included all available data (that is, industry and area breakouts were ignored) to maximize sample size (akin to a national report).

The preliminary dataset included 852 job titles common to reports from both years. The 2011 data included 24,264 incumbents from 448 unique organizations, while the 2012 data had 37,814 incumbents from 601 unique organizations. Titles with fewer than 10 observations in either year were dropped from the analysis, resulting in 196 titles for comparison. In an effort to show meaningful patterns, the job titles were grouped into job families. Subsequent posts will examine administrative and computer occupations.

Generally speaking, wage rates for 65% of the benchmark jobs increased from 2011 to 2012. Across the entire sample, the average change in salary was 3.6%. Please note, this is not a comparison of the same person in the same job, so these are not realized increases or decreases for an individual.

Of particular interest is executive pay, which included eight titles for comparison. As can be seen in the table below, organizations reported higher salaries in 2012 for six of the eight executive titles, ranging from 8% to 17%. Survey results for Chief Operating Officer and Top Legal Officer decreased in 2012.

Upon further examination, survey data for positions showing decreases had much greater disparity in sample size from one year to the next than those indicating increases. Matching individual salary changes across years is beyond the scope of the current analysis. Furthermore, while it may be possible to match organization submissions from year to year, it is not possible to ensure that the data corresponds to the same incumbent. (Also of note, constraining the analysis to the same organizations is not part of traditional survey methodology, and doing so would further limit the sample.) Nevertheless, such year-to-year comparisons do underscore the importance of understanding the survey methodology and sample sizes.