Full Citation

Title: Setting the Level and Annual Adjustment of Military Pay

Citation Type: Miscellaneous

Publication Year: 2021

ISBN: 9781977405852

ISSN:

DOI:

NSFID:

PMCID:

PMID:

Abstract: A key objective of the military compensation system is to enable the military to attract and retain personnel in sufficient numbers and quality to meet its manpower requirements, and to do so efficiently. With respect to the quality of military enlisted recruits, the U.S. Department of Defense’s (DoD’s) stated requirement is that 90 percent of recruits each year and in each service must be at least high school graduates and that 60 percent must be above average in terms of aptitude. To meet these requirements, past commissions, including the work of the Ninth Quadrennial Review of Military Compensation (9th QRMC) in 2002, found that military pay should be set at around the 70th percentile of the earnings for similar civilians. The 9th QRMC found that pay must be higher than average civilian pay because of the unusual demands and arduous nature of military service. The specific benchmark of the 70th percentile was chosen based on research from the 1990s that used data from the late 1980s and mid-1990s. Setting military pay involves not just setting the level but also determining the annual military pay adjustment. By law, the annual increase in military basic pay is guided by changes in the Employment Cost Index (ECI), a measure of the growth in private-sector employment costs computed by the Bureau of Labor Statistics within the U.S. Department of Labor. The ECI is also used to guide changes in the pay of federal General Schedule civil service employees, and, since 1967, the annual pay adjustment for military pay has been tied to the adjustment for General Schedule employees. More recent research, including the work of the 11th QRMC in 2012, has found that military pay, as measured by regular military compensation (RMC), far exceeds the 70th percentile and has done so for quite some time.1 This finding leads to the question of whether the level of military pay is being set efficiently, and, in particular, whether RMC is too high relative to the earnings of similar civilians. Or, put differently, it raises the question of whether the 70th percentile is still relevant as the benchmark, and, if not, what the new benchmark for setting the level of military pay should be. Regarding the ECI, research from the early 1990s (Hosek et al., 1992) found that the ECI does not accurately measure the opportunity wages of active duty personnel or perform well in terms of tracking recruiting and retention outcomes. The use of the ECI led to a “pay paradox” in the 1990s, when growth in basic pay fell well short of changes in the ECI, even though the services were meeting their recruiting and retention objectives. The research recommended the use of an alternative index, the Defense Employment Cost Index (DECI), that better reflected the demographics of military personnel and so more accurately measured the opportunity wages of active duty personnel. Not surprisingly, the DECI performed better than the ECI in terms of tracking recruiting and retention outcomes. The 7th QRMC in 1992 chose not to recommend that the DECI replace the ECI but recognized its advantages and recommended that its continued development be supported by DoD. No such development has occurred since then. The 13th QRMC requested that the RAND National Defense Research Institute assess the continued relevance of the 70th percentile as a benchmark for setting the level of military pay and the advantages and disadvantages of the DECI. It also requested that RAND provide computer code for computing the RMC percentile and for computing the DECI, for DoD’s future use. This report summarizes our analysis and findings and includes, in appendixes, the relevant computer code. Our approach involved using military personnel and pay data provided by the Defense Manpower Data Center (DMDC), active duty personnel survey data provided by DoD’s Office of People Analytics, and Current Population Survey data from the Bureau of Labor Statistics. Our analysis builds on and extends previous studies that have computed RMC percentiles (Hosek, Asch, and Mattock, 2012; Hosek et al., 2018; Smith, Asch, and Mattock, 2020) and the Hosek, Asch, and Mattock (2012) study that computed the DECI and compared it with the ECI.

Url: https://www.rand.org/content/dam/rand/pubs/research_reports/RRA300/RRA368-1/RAND_RRA368-1.pdf

User Submitted?: No

Authors: Asch, Beth J; Mattock, Michael G; Smith, Troy D; Ward, Jason M

Publisher: RAND Corporation

Data Collections: IPUMS CPS

Topics: Labor Force and Occupational Structure

Countries: