A Look At Cash Compensation for Active-Duty Military Personnel

A Look At Cash Compensation for Active-Duty Military Personnel Beth J. Asch James R. Hosek Craig W. Martin Prepared for the Office of the Secretary of Defense R National Defense Research Institute Approved for public release; distribution unlimited

REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188 Public reporting burder for this collection of information is estibated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burder to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 1. REPORT DATE (DD-MM-YYYY) 29-08-2002 2. REPORT TYPE 3. DATES COVERED (FROM - TO) xx-xx-2002 to xx-xx-2002 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER A Look At Cash Compensation for Active-Duty Military Personnel 5b. GRANT NUMBER Unclassified 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) Asch, Beth J ; Hosek, James R ; Martin, Craig W ; 7. PERFORMING ORGANIZATION NAME AND ADDRESS RAND National Defense Research Institute 1700 Main Street Santa Monica, CA90407-2138 9. SPONSORING/MONITORING AGENCY NAME AND ADDRESS Office of the Secretary of Defense, 12. DISTRIBUTION/AVAILABILITY STATEMENT APUBLIC RELEASE, 13. SUPPLEMENTARY NOTES 14. ABSTRACT See report 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT Same as Report (SAR) a. REPORT Unclassified b. ABSTRACT Unclassified c. THIS PAGE Unclassified 18. NUMBER OF PAGES 65 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 8. PERFORMING ORGANIZATION REPORT NUMBER 10. SPONSOR/MONITOR'S ACRONYM(S) 11. SPONSOR/MONITOR'S REPORT NUMBER(S) 19. NAME OF RESPONSIBLE PERSON http://www.rand.org/publications/mr/mr1492/mr149 rbezwada@dtic.mil 19b. TELEPHONE NUMBER International Area Code Area Code Telephone Number 767-9001 DSN 427-9001 Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39.18

The research described in this report was sponsored by the Office of the Secretary of Defense (OSD). The research was conducted in RAND s National Defense Research Institute, a federally funded research and development center supported by the OSD, the Joint Staff, the unified commands, and the defense agencies under Contract DASW01-01-C-0004. Library of Congress Cataloging-in-Publication Data Asch, Beth J., 1958 A look at cash compensation for active-duty military personnel / Beth J. Asch, James R. Hosek, Craig W. Martin. p. cm. MR-1492. Includes bibliographical references. ISBN 0-8330-3174-0 1. United States Armed Forces Pay, allowances, etc. I. Hosek, James R. II. Martin, Craig W. III.Title. UC74.A8335 2002 355.6'4 dc21 2002021395 RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND is a registered trademark. RAND s publications do not necessarily reflect the opinions or policies of its research sponsors. Copyright 2002 RAND All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from RAND. Published 2002 by RAND 1700 Main Street, P.O. Box 2138, Santa Monica, CA 90407-2138 1200 South Hayes Street, Arlington, VA 22202-5050 201 North Craig Street, Suite 202, Pittsburgh, PA 15213-1516 RAND URL: http://www.rand.org/ To order RAND documents or to obtain additional information, contact Distribution Services: Telephone: (310) 451-7002; Fax: (310) 451-6915; Email: order@rand.org

PREFACE This report presents an overview of military compensation for active-duty officer and enlisted personnel in 1999. It provides information on the receipt and amount of each type of cash pay, highlighting the patterns by year of service, occupational area, and branch of service. It also quantifies the range of variation in military pay and includes data on private-sector pay variation for comparison. The report should be of interest to policymakers and to researchers concerned with military compensation The research was undertaken for the ninth Quadrennial Review of Military Compensation, whose purpose is to investigate the adequacy of the military compensation system and recommend improvements as needed. The Office of Special Projects and Research, Office of the Under Secretary of Defense for Personnel and Readiness, sponsored the research. The research was conducted in the Forces and Resources Policy Center at RAND s National Defense Research Institute, a federally funded research and development center sponsored by the Office of the Secretary of Defense, the Joint Staff, the unified commands, and the defense agencies. iii

CONTENTS Preface... Figures... Tables... Summary... Acknowledgments... Acronyms... Chapter One INTRODUCTION... 1 Chapter Two PAY DEFINITION AND DATA... 3 Definition of Military Pay... 3 Data... 3 Chapter Three PAY LEVEL AND VARIANCE... 7 The Incidence of Pay Components... 7 Enlistment and Reenlistment Bonuses... 11 Average Pay by Year of Service... 17 Average Pay by Occupational Area... 20 Range of Variation in the Components of Pay... 27 Chapter Four CONCLUSION... 37 Appendix A. MONTHLY PAY COMPARISONS... 39 B. ADDITIONAL OFFICER PAY TABLES... 45 Bibliography... 49 iii vii ix xi xix xxi v

FIGURES 3.1. Percentage of Enlisted Personnel Receiving SRBs, Army, 1999... 14 3.2. Average SRB, Army, 1999... 14 3.3. Percentage of Enlisted Personnel Receiving SRBs, Air Force, 1999... 15 3.4. Average SRB, Air Force, 1999... 15 3.5. Percentage of Enlisted Personnel Receiving SRBs, Navy, 1999... 16 3.6. Average SRB, Navy, 1999... 16 3.7. Average Total Enlisted Pay by Service and Year of Service, 1999... 17 3.8. Average Total Officer Pay by Service and Year of Service, 1999... 18 3.9. Average Total Enlisted Pay by Year of Service, 1999... 19 3.10. Average Officer Pay by Year of Service (Commissioning Source Is Academy or ROTC), 1999... 20 3.11. Non-RMC Components of Enlisted Pay as a Fraction of Average total Pay, 1999... 21 3.12. Non-RMC Components of Officer Pay as a Fraction of Average Total Pay, 1999... 21 3.13. Average Enlisted Pay by Year of Service and Occupational Area, 1999... 22 3.14. Average Officer Pay by Year of Service and Occupational Area, Except Medical Officers, 1999... 23 3.15. Average Enlisted Pay by Year of Service and IT-Occupational Category, 1999... 24 3.16. Average Enlisted Pay by Year of Service and Air versus Nonair Occupational Areas, 1999... 25 3.17. Enlisted Pay Percentiles, 1999... 28 3.18. Officer Pay Percentiles, 1999... 28 3.19. Predicted Earnings Percentiles, Civilian Males with Some College, 1999... 29 3.20. Standard Deviation of Enlisted Pay by Year of Service, Air Force, 1999... 31 3.21. Standard Deviation of Enlisted Pay by Year of Service, Army, 1999... 32 3.22. Standard Deviation of Enlisted Pay by Year of Service, Marine Corps, 1999... 33 3.23. Standard Deviation of Enlisted Pay by Year of Service, Navy, 1999... 33 vii

viii A Look at Cash Compensation for Active-Duty Military Personnel 3.24. Standard Deviation of Officer Pay by Year of Service, Air Force, 1999... 34 3.25. Standard Deviation of Officer Pay by Year of Service, Army, 1999... 35 3.26. Standard Deviation of Officer Pay by Year of Service, Marine Corps, 1999... 35 3.27. Standard Deviation of Officer Pay by Year of Service, Navy, 1999... 36

TABLES S.1. Average Amounts of Cash Compensation, 1999 (Dollars)... xiv 3.1. Incidence and Average Amounts of Enlisted Pay, 1999... 8 3.2. Incidence and Average Amounts of Officer Pay, Commission Source Is ROTC or a Military Academy, 1999... 9 3.3. Incidence and Average Amount of Enlistment Bonus and SRB, 1999... 12 3.4. Incidence and Average Amount of Enlistment Bonuses by Year of Service, 1999... 13 3.5. Enlisted Year of Service Distribution by One-Digit DoD Occupational Code, FY 1999... 25 3.6. Officer Year of Service Distribution by One-Digit DoD Occupational Code, FY 1999... 26 A.1. Incidence and Average Amounts of Enlisted Pay for June 1999, for Members on Active Duty All 12 Months of 1999... 39 A.2. Incidence and Average Amounts of Officer Pay for June 1999, for Members on Active Duty All 12 Months of 1999 (Commission Source Is ROTC or a Military Academy)... 40 A.3. Incidence and Average Amounts of Enlisted Pay for June 1999... 42 A.4. Incidence and Average Amounts of Officer Pay for June 1999 (Commission Source Is ROTC or a Military Academy)... 43 B.1. Incidence and Average Amounts of Officer Pay, All Commissioned Officers, 1999... 45 B.2. Incidence and Average Amounts of Officer Pay, All Commissioned Officers and Warrant Officers, 1999... 46 ix

SUMMARY Every four years, the Quadrennial Review of Military Compensation (QRMC) examines the level and structure of military compensation to ensure that it continues to enable the armed services to meet its manpower requirements in a timely and costeffective manner. An area of interest to the ninth QRMC is the degree to which military pay varies among personnel and the extent to which special and incentive (S&I) pays contribute to total military pay. These pays are the key tool the services have to manage personnel flexibly and to vary pay in response to internal and external factors, such as the civilian economy, that affect their ability to attract and retain personnel. Because military compensation consists of a large array of pays and allowances, it is unclear the degree to which total cash compensation differs among military personnel. All military personnel receive basic pay. Basic pay is based on a pay table common to all personnel, regardless of occupation and branch of service. The services also make extensive use of the various S&I pays. In addition to providing the services with the flexibility to vary pay among personnel, these pays also enable them to recognize unusual duties and hazards and to provide individuals an incentive to enlist or reenlist in hard-to-fill skill areas. The common pay table, and the relative importance of basic pay in total cash compensation, would argue for substantial similarity in pay among military personnel. But the diversity and differential use of S&I pays by the services would argue for substantial pay differences among servicemembers and a substantial role for S&I pays. The purpose of this report is to investigate the role of the different components of military pay in total cash compensation, with attention paid to the role of S&I pays. We examine how large a portion of compensation consists of S&I pays and how much of the variance in compensation stems from S&I pays. We also consider how cash compensation varies among officer and enlisted personnel, by service, occupational group, and years of service. Since the competitiveness of military pay depends on how it compares to civilian pay opportunities, we also compare the variation in military cash compensation to the range in civilian earnings. This report focuses on describing military compensation and not on the important behavioral outcomes that result from variations in military pay. Our analysis is related to another study conducted by Kilburn, Louie, and Goldman (2001) that also examines the role of different pay components in determining the level and variance xi

xii A Look at Cash Compensation for Active-Duty Military Personnel of enlisted military compensation. Our analysis builds on this study by considering officers, not just enlisted personnel, by examining differences among occupational groups, and by using more data for a more recent year, 1999, as described below. DATA AND ANALYTICAL APPROACH Our analysis considers average values of military cash compensation and the average values of different components of cash compensation for active-duty officer and enlisted personnel in a single year, 1999. As described above, we examine differences in the average values by service branch, by occupational groups, and by years of service. We also examine the variance in compensation by considering both the standard deviation in pay and the difference between the 90th and 10th percentile of military compensation by service branch and years of service. The analysis relies on three sources of data. The first is the Joint Uniform Military Pay System (JUMPS) files for each of the 12 months of calendar year 1999, provided by the Defense Manpower Data Center (DMDC). These data indicate the incidence and actual dollar amounts of each pay component received by each servicemember in each month. The second source, also provided by DMDC, is known as the Proxy Personnel Tempo (PERSTEMPO) data file, a data set that provides monthly information on the characteristics of active-duty personnel, including occupation, service, and date of entry into the military. PERSTEMPO records for nine months of 1999 were matched to the pay records for 1999. For characteristics in the PERSTEMPO data that vary with time, such as years of service, we used the value of the characteristic as of September 1999 i.e., the end of FY 1999. Because the inflow and outflow of personnel during a year can affect the computation of annual pay, we restricted the analysis to members who served all 12 months of 1999. A final data restriction is that our analysis of officers was limited to officers whose commissioning source was either a military academy or ROTC. We exclude those who were direct appointments or who were commissioned through Officer Candidate School (OCS) or Officer Candidate Training (OCT) because these individuals tend to have high pay grades for their years of service (in the case of direct appointments) or have low officer pay grades for their years of service (in the case of OCS). Including these individuals might result in misleading figures about how the average value of officer pay varied by years of service. The third source is information provided by the Directorate of Compensation in the Office of the Secretary of Defense (OSD) on the average amount of BAH (basic allowance for housing), BAS (basic allowance for subsistence), and the federal tax advantage for 1999 by pay grade, year of service, and marital status. These averages were merged with the PERSTEMPO and JUMPS data. We used these averages rather than the actual values of these variables provided in the JUMPS data to place all personnel on an equal footing. Because members living in military housing receive zero benefits, using the actual dollar values would assign a zero value to their housing and, therefore, would bias our estimate of the housing benefit downward.

Summary xiii For the purposes of our analysis, we define total cash compensation to include BAS, BAH, S&I pays, bonuses, and the federal tax advantage that arises because some allowances are not subject to federal income tax. In short, cash compensation represents the pays that would appear in a member s monthly paycheck. While health benefits, future retirement benefits, and other forms of compensation are also important factors affecting manpower supply, they are not included in our definition of cash pay. The traditional definition of cash pay is Regular Military Compensation (RMC), which is the sum of basic pay, BAH, BAS, and the tax advantage. Our definition of total cash pay expands on the traditional definition because in addition to the components that constitute RMC, we also examine other miscellaneous allowances, S&I pays, and bonuses. Thus, we can analyze the role S&I pays play in total cash compensation. RESULTS The main result of our analysis is that considerable variation exists among the services in the incidence and average amounts of non-rmc pays and allowances. Nonetheless, these differences are overshadowed by the similarity in the average amount of RMC. The similar values of RMC at a given year of service and the fact that RMC accounts for 85 percent or more of average total cash compensation means that average cash compensation is quite similar across military personnel at a given year of service, regardless of branch of service or occupational group. Some exceptions can be found among certain officer communities, specifically doctors, aviators, and nuclear officers. Officers in these skill areas receive S&I pays that are quite large and that increase their pay well above the average pay in other skill areas. As Table S.1 shows, average annual cash compensation in 1999 for officers varied from $62,161 in the Marine Corps to $66,883 in the Air Force. For enlisted personnel, it varied from $29,355 in the Marine Corps to $33,744 in the Navy. By far, most of cash compensation is explained by the traditional measure of cash pay namely RMC. In the case of officers, RMC varied from $58,707 for the Marine Corps to $61,689 in the Army. Most of the variation across service branches in average annual cash pay and in RMC stems from differences in the years of service and pay grade distributions across the services. Because basic pay in the pay table varies by years of service and pay grade, differences in the distributions can result in variations in pay across services. Still, even when we conduct the analysis by year of service, we find remarkable similarity in average annual pay across the services at a given year of service. The dollar amounts of S&I pays, bonuses, and miscellaneous allowances and cost-ofliving allowance (COLA) shown in Table S.1 are unconditional averages i.e., they are computed across everyone in a particular branch of service. As Table S.1 shows, the average amounts of S&I pays and, therefore, the portion of average annual pay attributable to these pays are quite small. This result partly arises from the relatively low incidence of some pays and/or partly from the relative small dollar value of the pays among those who receive it.

xiv A Look at Cash Compensation for Active-Duty Military Personnel Table S.1 Average Amounts of Cash Compensation, 1999 (Dollars) Army Air Force Marine Corps Navy Officers RMC 61,689 61,599 58,707 59,761 S&I pays 927 2,810 1,889 3,134 Miscellaneous Allowances/COLAs 837 779 810 872 Bonuses 673 1,695 756 2,172 Annual Pay 64,125 66,883 62,161 65,940 Enlisted Personnel RMC 30,509 31,398 28,241 30,655 S&I pays 482 301 317 1,345 Miscellaneous Allowances/COLAs 832 1,015 785 967 Bonuses 372 381 11 777 Annual Pay 32,195 33,095 29,355 33,744 We examined the incidence of different pays and the average dollar amounts among those who receive them. The incidence of different S&I pays varies across the services and across the pay types. In most cases, even when the pay is relatively common throughout the enlisted or officer force, the average dollar amount among those who receive it tends to be small. For example, among enlisted personnel, between 12 and 26 percent of personnel received hostile fire pay in 1999, depending on service. Still, the average dollar amount of hostile fire pay among those who received it was only between $433 and $633. An exception to this finding is officer Aviation Career Incentive Pay. This pay is relatively pervasive, particularly in the Air Force, Marine Corps, and Navy (between 33 and 42 percent for those three services) and the average dollar figure was relatively high, between $5,456 and $6,155. In a few cases, the incidence of a pay is quite small, but the average dollar amount is high for those who receive the pay. For example, less than 1 percent of Army officers receive the medical officer retention bonus, but the average bonus is $36,260. Similarly, only 7.6 percent of Air Force officers receive the Aviation Officer Continuation Bonus, but the average dollar amount is $17,657. Not surprisingly, S&I pays are an important portion of total cash compensation for individuals in these specific skill areas. Among the enlisted force, one of the more important sources of pay is enlistment and reenlistment bonuses. The incidence of enlistment bonuses was the highest in the Army, and yet, in 1999, only 3 percent of Army enlisted members received enlistment bonuses. The incidence of reenlistment bonuses was the highest in the Navy, and about 15.5 percent received a reenlistment bonus in that service. These percentage figures include both initial payments of bonuses and anniversary payments. The incidence of bonuses varies by years of service, as would be expected, given that bonuses are targeted to personnel making enlistment or reenlistment decisions in early and midcareer. For example, between YOS 5 and 11, upward of 40 percent of Navy enlisted personnel receive a reenlistment bonus, far higher than the 15.5 percent figure computed across individuals at all YOS. Nonetheless, as mentioned ear-

Summary xv lier, RMC represents such a large fraction of average total cash compensation, that even when we consider the role of bonuses in total compensation by year of service, we still find that bonuses play a relatively small overall role during YOS 5 11. We also consider the size and determinants of the variation in military pay and how its variation compares to the variation in civilian earnings. We examine the range of cash compensation by considering the difference between the highest and lowest deciles (i.e., the 90th and 10th percentiles) of pay at each year of service. To examine the determinants of the variation, we compute the standard deviation of four increasingly inclusive measures of pay, RMC, RMC plus S&I pays, RMC plus S&I pays plus bonuses, and RMC plus S&I pays plus bonuses plus miscellaneous allowances and COLAs. The difference between the highest and lowest deciles varies somewhat by year of service. For enlisted personnel, the difference is about $8,000 at YOS 5, about $10,000 at YOS 10, about $12,000 at YOS 20, and about $11,000 at YOS 25. Thus, the largest difference is in midcareer, although it does not change much after YOS 10. In part, the range of variation at each YOS reflects differences in pay grade. Bonuses and S&I pays add to the variation. We compared the difference between the 90th and 10th percentile of military average cash pay by year of service with the difference between the 90th and 30th percentile of civilian average earnings, computed from the Current Population Survey by Professor John Warner at Clemson University. The computation was limited to males with some college education and was made for each year of experience to allow comparison with military pay. We used the 30th percentile as the lower bound because civilians who do not score well on the Armed Forces Qualification Test (AFQT) or who are not high school diploma graduates are unlikely to qualify for enlistment into the military but are likely to be found among those in the lowest percentiles of the civilian earnings distribution. The key conclusion of our military and civilian comparison is that the variation in civilian pay as measured by the difference between the 90th and 30th percentile is far larger than the range in enlisted earnings. For example, at 10 years of experience, the difference in civilian earnings is about $23,000, far higher than the $10,000 difference found between the 90th and 10th percentile of military pay at YOS 10. At 20 years of experience, the civilian difference was about $30,000, again far higher than the $12,000 figure we found for enlisted personnel. It is important to note that the civilian earnings figures are averaged across many civilian firms that differ in hiring requirements, occupational mixes, industry conditions, and location-specific conditions. Although the military s workforce is diverse, it is far more homogeneous in terms of these factors than the civilian economy at large is. Nonetheless, the range of variation among civilian males with some college is far larger than the range for enlisted personnel, more than we might expect because of the heterogeneity of the civilian labor market. To understand the extent to which the components of military pay contribute to variation in military cash pay, we examined the standard deviation of pay for the four, increasingly inclusive, measures of pay given above. Because some of the pay

xvi A Look at Cash Compensation for Active-Duty Military Personnel components, notably BAH, BAS, and the tax advantage, are averages from OSD, their contribution to the variance of total pay will be understated. The result is that the variance in RMC (which is the sum of basic pay, BAS, BAH, and the tax advantage) at a given year of service will reflect differences in pay grade and marital status of personnel at a given year of service. The findings regarding the determinants of variance differ across the services. For the Air Force, the standard deviation of enlisted pay is in the $3,000 to $4,000 range over most years of service, with about half due to bonuses during YOS 4 to 11, after which variation in RMC accounts for most of the difference. The variation stemming from RMC reflects the diversity of pay grades in those years of service for the Air Force. Notably, S&I pay components (excluding bonuses) account for little of the variation in Air Force pay. For the Army, the standard deviation of enlisted pay lies within a fairly narrow band between $3,000 and $4,000 during YOS 6 25 and even less in YOS 1 5. Unlike the Air Force, pay variation stemming from RMC is more prominent in the early career (YOS 1 8) because of the greater diversity of pay grades in the Army in those years of service. The variation due to S&I pays are also greater than in the Air Force. For the Marine Corps, the variation in enlisted pay rises steadily from about $1,500 at YOS 1 to $5,000 at YOS 24. The increase is due to variation in RMC and miscellaneous allowances and COLAs. The Marine Corps makes little use of bonuses. The Navy has the greatest variation in enlisted pay among the services. Between YOS 1 and 4, the standard deviation of pay rises from $2,000 to $5,000, with most of the increase arising from bonuses. Over the YOS 5 to YOS 25 period, the standard deviation is in the $5,000 to $6,000 range, with much of the variation coming from bonuses and much coming from S&I pays, especially sea pay. As in the Air Force, bonuses play a prominent role in the Navy in creating pay differentials during early and midcareer for enlisted personnel. Among officers, the standard deviation of RMC for the Air Force, Army, and Navy is nearly $8,000 in the first few years of commissioned service, then declines to less than $4,000 in YOS 4 12. The Marine Corps figures are a bit higher. S&I pays add about $1,000 to the standard deviation, and bonuses add substantially to pay variance. However, it is important to recall that bonuses are generally received by only a small percentage of officers, although as noted earlier, some bonuses amounts are quite substantial, especially the amounts given to doctors, aviators, and those in the nuclear fields. CONCLUSIONS The main conclusion stemming from our analysis is the high degree of similarity in cash compensation among military personnel at each year of service, regardless of branch of service or occupational group. In other words, although pay grows over a military career, pay is remarkably similar among personnel at the same year of service. A few notable exceptions exist, such as officers in the medical, aviation, and nuclear fields who receive S&I pays and bonuses that substantially increase their average compensation. Apart from these exceptions, equity in compensation seems

Summary xvii to prevail. Thus, despite the large number of S&I pays, they play a relatively small role in determining average total cash compensation. The small role of non-rmc components of pay in determining average total cash compensation is explained in part by the relatively small fraction of personnel who receive the non-rmc components of pay i.e., S&I pays, bonuses, and miscellaneous allowances and COLAs. Even in cases where the incidence of these pays is fairly pervasive, such as the Clothing and Uniform allowance, their average dollar values tend to be small. For example, nearly all personnel receive the clothing allowance, but the average amount is less than $400 for enlisted personnel and less than $600 for officers. The similarity in pay outcomes and relatively small overall role of non-rmc components also stem from the common foundation of basic pay and similarities in the services promotion systems in terms of promotion phase points and promotion criteria. These similarities in compensation across personnel and the similarities in the promotion systems provide tangible evidence of the military s commitment to equity of pay opportunities regardless of skill area. Nonetheless, that equity results in remarkable similarity in the retention profiles of personnel across the services and across occupational groups. In other words, the experience mix of occupational groups varies relatively little within each service, despite the enormous diversity in the skill requirements and duties of the personnel in these groups. Further, although notable differences exist in the experience mix across the services, with the Marine Corps having a more junior force and the Air Force having a more senior force, the similarities in the experience mixes of personnel in the different services are also notable. To the extent that the services would like to achieve more variable career lengths and more diversity in the experience mix of different occupations, the analysis in this report suggests that greater differentiation in military pay and changes in the structure of military compensation may be required.

ACKNOWLEDGMENTS This research would not have been possible without the compensation and personnel files supplied by the Defense Manpower Data Center, the major repository of data on military personnel and compensation. We are also deeply grateful to Curtis Gilroy and John Enns, director and co-director of the ninth Quadrennial Review of Military Compensation, for their guidance, support, and commitment to this work. We appreciate comments received on earlier versions of this research from members of the QRMC Working Group, as well as from our colleagues at RAND. John Warner kindly provided predicted percentiles of private-sector earnings. Susan Everingham, director of RAND s Forces and Resources Center, provided thoughtful observations on many aspects of our QRMC work and helped delineate connections between this and other personnel projects within the center. We are grateful to our reviewers, Charles Goldman at RAND and Glenn Gotz at the Institute for Defense Analyses. xix

ACRONYMS ACIP AFQT BAH BAS COLA CONUS DMDC FSA IT JUMPS MWR OCS/OCT OSD PERSTEMPO QRMC RMC ROTC S&I SRB YOS Aviation Career Incentive Pay Armed Forces Qualification Test Basic Allowance for Housing Basic Allowance for Subsistence Cost-of-living allowance Continental United States Defense Manpower Data Center Family Separation Allowance Information technology Joint Uniform Military Pay System Morale, welfare, and recreation Officer Candidate School/Officer Candidate Training Office of the Secretary of Defense Personnel tempo Quadrennial Review of Military Compensation Regular Military Compensation Reserve Officer Training Corps Special and incentive (pays) Selective Reenlistment Bonus Year of Service xxi

Chapter One INTRODUCTION Every four years, the level and structure of military compensation are reviewed to determine their adequacy for meeting the military personnel supply needs of the services. Military compensation presents a complicated picture because of the large number of special and incentive (S&I) pays and allowances that have been created over time. These pays and allowances address specific concerns about such factors as housing costs, dangerous duty, arduous duty, separation from family, and pay differences related to particular skill areas, such as medicine and aviation. It is possible that they create wide differences in military compensation among personnel. At the same time, military compensation is built on the foundation of basic pay. The Army, Navy, Air Force, Marine Corps, and Coast Guard share the same basic pay tables for officers and enlisted personnel. These common basic pay tables may lead to the expectation that the actual level of military compensation by year of service and pay grade (rank) is similar across the services. Further, if basic pay were the dominant element of military compensation, the fact that basic pay tables are common across the services implies that servicemembers face largely similar compensation incentives to remain in service. If so, one might expect a high degree of similarity in retention profiles by year of service, across the services. However, the services differ in their desired personnel force structure for example, the Marine Corps prefers a heavily junior force. Its first-term reenlistment rate is about half that of the other services. Thus, compensation incentives are not the only mediator of the personnel force structure. Even so, common basic pay tables and similar promotion practices would suggest a small amount of pay diversity among retained personnel, which is the opposite of what one might expect from the many S&I pays and allowances. The purpose of this report is to put S&I pays in perspective as elements of military cash compensation. We investigate how large a portion of military compensation they currently represent and how much they contribute to the variance in military compensation across personnel. Although military compensation is a pivotal factor in determining recruiting and retention outcomes, this report is intended to describe military compensation rather than relate it to behavioral outcomes. Given the recruiting difficulties experienced in the late 1990s by the Army, the Navy, and the Air Force and the retention problems experienced in some occupational areas, it is of interest to consider how military compensation differs among sub- 1

2 A Look at Cash Compensation for Active-Duty Military Personnel groups and how it compares to civilian pay. Furthermore, the dispersion of earnings in the private-sector economy has grown since the early 1980s, and differences in skill, education, and ability help explain that trend. Given the importance of privatesector earnings opportunities relative to military earnings opportunities in the retention decision, it is also of interest to understand the amount of variation in military earnings and the importance of skill versus other factors (such as location or duty type) in explaining that variation. A recent RAND study (Kilburn, Louie, and Goldman, 2001) examined patterns in the level and variance of enlisted compensation among some subgroups, but the study did not examine components of compensation among other subgroups, most notably in different occupational areas or among officers. This report provides information to fill that gap. We address the following questions: What is the relative size of different components of pay over enlisted and officer careers? The components include basic pay, allowances, S&I pays, bonuses, and the federal tax advantage. How do enlisted and officer pay profiles vary by service and by occupational area? How do the level and range of variation in military pay compare to the level and range in civilian pay? Chapter Two defines the elements of military compensation addressed in this report and describes the sources of data. Chapter Three presents a variety of pay comparisons in tables and figures, and Chapter Four contains our conclusions.

Chapter Two PAY DEFINITION AND DATA DEFINITION OF MILITARY PAY We focus on the cash elements of military compensation for active-duty personnel. These include basic pay, Basic Allowance for Subsistence (BAS), Basic Allowance for Housing (BAH), S&I pays, bonuses, miscellaneous allowances, and cost-of-living allowances (COLAs). The exceptions to limiting the analysis strictly to a member s cash pay are to assign BAS and BAH values to members who live in on-base housing and to include a tax advantage attributable to the fact that BAS and BAH are not subject to federal income tax. The allocation of BAS and BAH to members living on base assumes that BAS and BAH amounts are reasonable approximations of the value of meals and housing provided in kind. The tax adjustment puts BAS and BAH on par with other pays, all of which are pretax. Descriptions of pays and allowances may be found in the Uniformed Services Almanac (2001). Also, the Department of Defense is developing a military pay Web site (http://militarypay.dtic.mil). Cash pay does not include military health care, retirement benefits, educational benefits, and the in-kind provision of training and education. It also does not include in-kind services provided through morale, welfare, and recreation (MWR) accounts such as childcare, family counseling, and recreation facilities. Kilburn, Louie, and Goldman (2001) present estimates of the value of military health care by family size, measured in terms of the private-sector health insurance premium for similar coverage. They also estimate the present discounted value of a servicemember s expected retirement benefits, by year of service. DATA The military pay data used for this analysis cover the 1999 calendar year and indicate the incidence and level of all pays for each individual on active duty, as well as the individual s grade, occupation, year of service, and other characteristics. We created the data file from three sources. The first is the Joint Uniform Military Pay System (JUMPS) files for each of the 12 months of calendar year 1999, provided by the Defense Manpower Data Center (DMDC). The monthly JUMPS file has a record for each active-duty member who was paid. For each type of pay, the data indicate whether the individual received the pay in that month and the dollar amount. 3

4 A Look at Cash Compensation for Active-Duty Military Personnel In this report, we present tables and tabulations based on members who were in the military for all 12 months of calendar 1999. This approach captures payments made at irregular intervals during the year and provides information about the incidence, level, and variance of pays from the perspective of career members or those contemplating a career. However, another source of pay variation comes from the turnover of military personnel during the year, and it is possible that the inclusion of members who leave the military would present a different picture of compensation. For instance, leavers might have relatively lower ranks at a given year of service or receive fewer S&I pays. Therefore, to present a perspective on pay that includes the entire population of military personnel those who stay throughout the year as well as those who arrive or leave during the year we prepared tabulations for a given month during the year: June. These tabulations, which appear in Appendix A, show the incidence and amount of pays in June 1999 for members present all year and for members present in June whether or not they served the entire year. Although the percentages of members receiving S&I pays and certain allowances differ somewhat in the month view versus the all-year view, the message about the relative contribution of such pays is much the same. The second data source, also provided by DMDC, is known as the Proxy Personnel Tempo (PERSTEMPO) data file. It is built from the Active-Duty Master files and contains information on occupation and date of entry into the military. We used monthly information from the PERSTEMPO file for the first nine months of 1999; data were not available for later months. Such time-sensitive variables as occupation were taken as of September 1999. These data were then merged with the JUMPS records for each individual. The Directorate of Compensation, Office of the Secretary of Defense (OSD), provided the final source of data. These are estimates of the average amount of BAH, BAS, and the federal tax advantage for 1999, by pay grade, year of service, and marital status for officer and enlisted personnel. We applied the average BAH, BAS, and tax advantage data to officers and enlisted, given their pay grade and marital status. This allowed us to place members on a comparable footing. If we had instead relied on actual BAH, members living in military housing would have had zero values of BAH. Also, we wanted to include the tax advantage, the OSD computation of which depends on average BAS, so we included average BAS. (The JUMPS data contain actual BAS and housing allowances at the individual level for months in which servicemembers received those allowances.) To construct the analysis file, we merged the 12 JUMPS files for 1999, allowing us to compute total pay, by pay component, for each member. These data were merged with PERSTEMPO data to capture occupation and entry or commissioning date and OSD data on average BAH, BAS, and tax advantage by grade and marital status. Inflow and outflow of personnel during a year can affect the computation of annual pay. Because we restricted the analysis to members who served all 12 months of

Pay Definition and Data 5 1999, 1 we excluded new entrants in calendar year 1999 except those who entered in January 1999. This restricted the number of new entrants in our data, so we included in the analysis file those who entered in October 1998 through December 1998 i.e., who entered in the first quarter of FY 1999. Thus, the first year members of the analysis file are an undercount of personnel in their first year but should be sufficient to provide measures of annual pay comparable with those of personnel in later years of service. Annual pay was defined as pay during the 12 months in calendar year 1999 regardless of entry date. With these restrictions and inclusions, the analysis file contained 967,000 enlisted records and 96,000 officer records, out of the 1,490,000 in the JUMPS files. In separate tabulations of enlistment bonus incidence and amount, we included all personnel in the first year of service, not just those who served all twelve months of 1999. This ensured an accurate representation of the percentage of first-year personnel receiving an enlistment bonus. The analysis of enlisted personnel includes all individuals who meet the data restrictions described in the previous paragraphs. The analysis of officers also includes individuals who meet these restrictions as well as one additional restriction. Officers must have as commissioning source either a military academy or Reserve Officer Training Corps (ROTC). We exclude individuals who were direct appointments or commissioned via Officer Candidate School (OCS) or Officer Candidate Training (OCT). The reason is that OCS/OCT officers often have more years of service as a result of entering the military as part of the enlisted force, while direct-appointment officers enter at a higher rank and so often have a higher pay grade. Because most of our pay analysis is conducted by year of service, the inclusion of OCS/OCT or directappointment officers would yield misleading figures about how the average value of officer pay varied by YOS. Nevertheless, we recognize that OCS, OCT, and direct appointments are significant sources of officers. Therefore, we have prepared tables (Appendix B) that include all commissioned officers and all commissioned officers and warrant officers. These more comprehensive populations are companions to the service academy/rotc population discussed in the text. The incidence and amounts of pays are somewhat different for the more inclusive populations, but the main findings about the relative contribution of S&I pays and allowances are basically the same. The key variables in the analysis are years of service and the different components of military pay. Years of service are measured as of September 30, 1999 (i.e., the end of FY 1999). The analysis also focuses on seven categories of pay components. The first four are basic pay, BAS, BAH, and federal tax advantage. These four constitute Regular Military Compensation (RMC). The final three categories are S&I pays, bonuses, and miscellaneous allowances. (Bonuses are often counted among S&I pays, but we treat them as a separate category.) These three categories include many different pays, as Tables 3.1 and 3.2 reveal. 1 As mentioned, the tables in Appendix A allow for a comparison between an all-year population and one with an inflow and outflow of personnel.

6 A Look at Cash Compensation for Active-Duty Military Personnel Finally, the year-of-service numbering convention we use is as follows. YOS 1 refers to the period during the first year of service. This is the usual convention. It contrasts with the convention for stating one s age, where one-year-old refers to an infant in the second year of life. Hence, someone who enlists for a four-year term and extends for three months would make a reenlistment decision in YOS 5.

Chapter Three PAY LEVEL AND VARIANCE This chapter shows that the incidence and average amounts of non-rmc pays and allowances differ across the services. These differences are overshadowed, however, by similarity in the average amount of RMC. The similar values of average RMC at a given YOS and the fact that RMC accounts for 85 percent or more of total pay mean that average total pay is similar for individuals at a given YOS, regardless of service branch or broad occupational area. That is, we find that average total pay differences at a given YOS are relatively small. Several exceptions occur: doctors, aviators, and nuclear officers receive S&I pays and bonuses that increase their pay well above the average pay for officers in other occupational areas. THE INCIDENCE OF PAY COMPONENTS Table 3.1 shows the incidence and the average amount of military pay for enlisted personnel in 1999. Table 3.2 is a similar table for officers whose source of commissioning was either a military academy or ROTC. Care is needed in interpreting the enlistment bonus and Selective Reenlistment Bonus (SRB) figures because the averages in Tables 3.1 and 3.2 confound initial payment of the bonus, which may be large, with smaller anniversary payments. Enlisted RMC (Table 3.1) averaged $30,509 in the Army, $31,398 in the Air Force, $28,241 in the Marine Corps, and $30,655 in the Navy. The Army, Air Force, and Navy averages are close to one another, suggesting an overall similarity in the services YOS/pay grade mix. The Marine Corps, which has about 70 percent of its enlisted personnel in the first term of service, has an average about $2,000 lower. Officer RMC (Table 3.2) exhibits virtually no difference between the Army and Air Force: $61,689 and $61,599, respectively. The Marine Corps has the lowest average at $58,707, probably reflecting the more-junior nature of its officer corps. The Navy s average, $59,761, also suggests a more-junior officer corps than that of the Army and Air Force. As these tables make clear, the incidence and average amounts of S&I pays and of allowances varied considerably across branch of service in 1999. As expected, career sea pay is pervasive in the Navy. About 40 percent of enlisted personnel and 19 per- 7

8 A Look at Cash Compensation for Active-Duty Military Personnel Type of Pay Table 3.1 Incidence and Average Amounts of Enlisted Pay, 1999 Army Air Force Marine Corps Navy Basic Pay 100.0 19,542 100.0 20,371 100.0 17,611 100.0 19,757 BAH (Green Book) 100.0 6,497 100.0 6,559 100.0 6,245 100.0 6,453 BAS (Green Book) 100.0 2,738 100.0 2,738 100.0 2,738 100.0 2,738 Tax Advantage (Green Book) 100.0 1,732 100.0 1,731 100.0 1,647 100.0 1,707 Average RMC 30,509 31,398 28,241 30,655 Foreign-Duty Pay 28.1 73 25.2 65 10.3 35 5.3 90 Proficiency Pay 6.1 2,699 3.0 2,285 5.8 2,583 9.4 2,108 Overseas Extension Pay 0.4 696 0.1 434 1.5 1,212 0.4 675 Career Sea Pay 0.1 1,314 0.0 112 9.0 205 40.5 1,624 Career Sea Pay Premium 0.0 742 0.0 0.0 734 5.1 684 Hostile Fire Pay 15.7 633 19.8 570 12.1 468 26.1 511 Diving Duty Pay 0.1 1,744 0.3 1,687 0.3 1,800 1.7 2,007 Submarine Duty Pay 0.0 0.0 0.0 7.5 2,094 Foreign Language Pay (1) 1.5 675 1.5 806 0.7 620 0.5 715 Foreign Language Pay (2) 0.2 332 0.1 360 0.0 0.0 373 Flying Pay (Crew Member) 1.0 1,688 3.1 1,979 1.3 1,847 1.9 2,120 Flying Pay (Noncrew) 0.0 0.0 0.8 1,003 0.0 Parachute Duty Pay 10.1 1,471 0.2 1,078 0.7 1,095 0.3 1,417 Flight Deck Duty Pay 0.0 1,200 0.0 85 2.4 471 9.0 591 Demolition Duty Pay 0.4 1,567 0.4 1,641 0.3 1,475 0.5 1,406 Experiment Stress Duty 0.0 870 0.2 1,261 0.0 1,387 0.2 747 Toxic Fuels Duty Pay 0.0 261 0.3 1,507 0.0 0.0 303 Toxic Pesticides Duty 0.0 532 0.0 1,166 0.0 0.0 998 High-Altitude Low Opening 0.3 2,297 0.3 2,399 0.2 2,207 0.5 2,498 Chem Munitions Duty 0.1 927 0.0 813 0.0 0.0 546 Average S&I Pay a 482 301 317 1,345 FSA b I 1.4 181 0.7 308 0.0 0.8 180 FSA II 19.9 417 17.1 333 19.2 385 23.0 399 CONUS COLA 0.6 730 0.6 355 1.4 612 0.7 697 Overseas COLA 24.6 1,849 24.1 2,904 21.4 2,240 19.4 2,748 Clothing/Uniform Allowance 87.2 329 90.8 281 97.9 229 99.7 336 Avg Misc Allowance/ COLAs a 832 1,015 785 967 Enlistment Bonus 3.0 5,193 1.7 3,749 0.5 2,137 2.2 4,139 SRB 11.2 1,949 10.1 3,167 0.0 15.4 4,452 Average Bonus a 372 381 11 777 Average Annual Pay a 32,195 33,095 29,355 33,743 NOTE: Foreign Language Pay (2) is received by members who have mastered a second foreign language. FSA II is received by members who are involuntarily separated from their families. a Averaged over all members, including those not receiving these pays. b FSA = Family Separation Allowance.