Research Brief

Military personnel are frequently rotated in and out of positions and are removed from their positions for training and other military duties. For these reasons, the Department of Defense (DoD) frequently discusses civilianization—the transfer of functions performed by military personnel to civil service personnel—as one way to save costs while affecting force effectiveness minimally.

However, while conventional wisdom argues that civil service workers are cheaper than their military counterparts, little analysis has been done to determine whether this is really so.

RAND researchers Susan M. Gates and Albert A. Robbert conducted an analysis to shed light on this issue, starting first by estimating the costs of civil service and military workyears (full-time employment of one worker for one year) at specific grade levels. Given the results of that analysis, they then analyzed two approaches to substituting civil service for military personnel and the policy implications of the different approaches.

Estimating the Relative Cost of A Civil Service Workyear and a Military Workyear

Before the researchers could determine whether the DoD can save on costs by substituting civil service for military personnel, they needed to know how much a civil service workyear costs relative to a military workyear at specific grade levels. To find out, Gates and Robbert updated previous RAND costing work done in the late 1980s,[1] focusing their efforts on the cost elements that differ between civil service and military personnel, not on developing a comprehensive estimate of all costs related to a military or a civil service workyear.

In this analysis, the researchers focused on civilianization actions involving a relatively small proportion of the workforce, and examined the costs associated with adding or removing a single employee—the incremental costs of a civil service workyear. Researchers basically followed the previous RAND approach while updating the data. However, in deriving the cost of an incremental military workyear—the sum of base pay, the cost of benefits, and the cost of other pay—researchers expanded on the previous approach by changing the way health benefits are estimated and by adding in the costs of military accession and training, which the previous approach treated as one-time costs.

The health-benefits analysis was done because, since the late 1980s, significant changes have occurred in how health care is provided to military personnel, retirees, and dependents. Accession and training were added because the researchers believe military training and accession costs should be treated as incremental costs: In a steady state, the DoD must access a certain number of personnel each year to maintain the size of the force.

The Results of the Substitution Analysis

General Civil Service and Military Grade Equivalencies

Military Grade Equivalent Civil Service Grade
O-10 x - - - -
O-9 x - - - -
O-8 x - - - -
O-7 x - - - -
O-6 - 15 - 15 14–19
O-5 - 13, 14 - 15 14–19
O-4 - 12 - 15 14–19
O-3 - 10, 11 12–15 6–14 8–13
O-2 - 8, 9 12–15 6–14 8–13
O-1 - 7 12–15 6–14 8–13
W-4 - 8, 9 12–15 6–14 8–13
W-3 - 8, 9 12–15 6–14 8–13
W-2 - 7 12–15 6–14 8–13
W-1 - 7 12–15 6–14 8–13
E-9 - 6 9–11 1–5 1–7
E-8 - 6 9–11 1–5 1–7
E-7 - 6 9–11 1–5 1–7
E-6 - 5 9–11 1–5 1–7
E-5 - 5 9–11 1–5 1–7
E-4 - 4 1–8 - -
E–3 - 1–3 1–8 - -
E-2 - 1–3 1–8 - -
E-1 - 1–3 1–8 - -

SOURCE: With the exception of SES categories, this table represents grade equivalencies found in DoDI 1000.1 (1974).

NOTE: O=Officer; W=Warrant Officer; E=Enlisted; SES=Senior Executive Service; GS=General Schedule; WG=Wage Grade; WL=Wage Leader; WS=Wage Supervisor.

With the results of the cost estimation in hand, the researchers analyzed the costs and benefits of substituting civil service for military personnel. Any such analysis fundamentally rests on knowing the answers to such questions as, Which service grades are substituted for which military grades? Do the aggregate grade distributions change in the event of civilianization? and Is the substitution one-for-one? Unfortunately, no systematized information is available on these issues. Thus, cost-comparison studies like this one must rely on assumptions.

The two approaches examined relied on differing assumptions. The first approach is the traditional one currently used by the DoD and other government agencies for replacing military personnel with civil service workers. It is based on comparing the cost of military and civil service personnel at comparable grade levels. It assumes that civilianization leads to a change in the aggregate grade distributions and that the grade substitutions follow the guidelines set forth in DoD Instruction (DoDI) 1000.1,[2] which (except for the Senior Executive Service [SES]) is presented in the table.

Using the researchers' cost estimates, this approach leads to the conclusion that civil service employees are less costly than military personnel.

However, the grade equivalencies shown in the table were developed for administrative purposes; thus, there is no evidence that they reflect comparability of work done by individuals in these grades. Ideally, equivalencies would be developed by comparing the nature of work done by individuals in the specific grades or by empirical evidence on actual substitutions.

To illustrate the importance of understanding the true nature of these grade-by-grade comparisons, the researchers posed an alternative approach, which—like the traditional approach—assumes the substitution is one-for-one and that the civil service grade structure is altered as civil service workers in specific grades are hired to fill the civilianized positions; however, unlike the traditional approach, it assumes that civilianization results in no change to the military grade structure.

Military–Civil Service Conversion Cost Break-Even Points When Military Grade Distributions Are Not Adjusted

When marginal manpower changes are evaluated this way, a military-civil service conversion yields a savings if the grade-specific civil service manpower cost is less than the average cost of the corresponding flag officer (O-7 through O-10; $155,919), other commissioned officer (O-1 through O-6; $83,063), warrant officer (W-1 through W-5; $69,372), or enlisted manpower (E-1 through E-9; $43,479) type it replaces. When these average costs are compared with the costs for civil service personnel generated in the earlier cost analysis, we find the break-even points shown in the figure.

Civil service personnel in grades above the line are more expensive than the average military workyear of a given type. For example, commissioned officers (other than flag officers) are less expensive than GS-14, GS-15, and SES civil service personnel, but more expensive than most other civil service personnel. No break-even is presented for flag officers, because they are more expensive than all civil service personnel, including SES employees.

The researchers believe that the alternative approach best reflects the actual military personnel management and budgeting process. Under these circumstances, cost-effective civilianization would require the DoD to limit substitution to positions that could be filled with lower-grade civil service workers. While such a policy might generate substantial cost savings, it could create personnel-management problems with both workforces.


From these findings, the researchers concluded that civilianization can produce cost savings under many, but not all, circumstances. As a result, they recommend that the Office of the Secretary of Defense (OSD) modify its current guidance on military/civil service position assignments. Revised guidance should specify that assignment decisions be based on three considerations: military necessity, cost, and career-progression opportunities.


This report is part of the RAND Corporation research brief series. RAND research briefs present policy-oriented summaries of individual published, peer-reviewed documents or of a body of published work.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.