Project on Defense Alternatives

The Readiness Crisis of the U.S. Air Force:
A Review and Diagnosis

Project on Defense Alternatives Briefing Report #10
Carl Conetta and Charles Knight
22 April 1999

This Publication Available in Print
$18.00 buy

[Table of Contents] [Executive Summary] [Sections 1-3] [Conclusion] [Bibliography]

4. The Long View: Air Force Readiness in the Post-Cold War Era

Our review of the Air Force's recent readiness problems traces them to two principal sources: (i) discrete miscalculations of spare parts requirements in 1996 and 1997, and (ii) imprudent or inflexible management of the pilot inventory. A prominent, contrasting perspective is that the Air Force's recent troubles reflect something more fundamental -- a mismatch between missions, force structure, and resources affecting all of America's armed forces and rooted in a post-Cold War retrenchment policy that has gone too far. In this view, America's military has been asked to do too much while being given resources insufficient to support it even in its essential, warfighting function.

The two views lead in very different policy directions -- the first prescribes change in how the Air Force manages its resources, organizes itself, and conducts its business. The second emphasizes a need to either significantly curtail the Air Force's involvement in peace operations, or substantially increase the resources at the service's disposal, or both. The following sections of this report examine key readiness indicators and issues over a longer time frame than the past few years. The aim is two-fold. First, to ask, Is there evidence that the post-Cold War draw-down has created chronic readiness problems in the Air Force? And, second, to assess those problems that the Air Force has encountered in terms of appropriate solutions. Where are new resources needed? And in what instances is innovation the required remedy?

4.1 A twenty year perspective on mission capable rates

The recent dip in mission capable and unit readiness rates gives a distorted impression of the Air Force's post-Cold War condition. By most measures -- and especially those that consider the whole Air Force -- the years 1992-1996 compare fairly well with the decade preceding them. The common choice of 1991 or 1990 as a baseline for evaluating today's force also presents problems. Mission capable rates reached historic peaks in these years for a unique reason -- during and immediately after the Gulf War there was a concerted effort to bring all participating combat aircraft up to prime operating condition. It was an unusually comprehensive effort, facilitated by support systems that had been geared-up for the war and, later, by allied financial contributions to the United States for its war effort. A more meaningful baseline for evaluating and interpreting today's MCRs is the period preceding the Gulf War and stretching back to the "hollow force" years.

Two sets of DoD data assembled by the Congressional Budget Office help put the recent readiness condition of USAF aircraft in perspective. One set presents annual average MCRs for the period 1980-1993; the other, annual MCR's for the period 1980-1996 indexed to 1981 (so that the MCR for any year can be read as a fraction of the 1981 MCR).54 Although the two data sets are not entirely in accord, they both show a surge in MCRs during 1990. And, they both show MCRs in subsequent years remaining above the levels that prevailed in the late 1980s. One implication is that the recent dip in this rate, temporary or not, represents a decline from an unusually high starting point -- not only 1990 but also the six years that followed. If evaluations of the Air Force's current condition were to use 1989 as a baseline -- instead of 1990, 1991, or 1992, as is usually the case -- the recent dip to a 75% MCR would register as a decline of less than 6%.55 And the years 1992-1996 would all register above the value for 1989.

Table 3: Mission-Capable Rates for Aircraft (Active and Reserve Components)

  1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Index Value 1.01 1 1 1.02 1.06 1.13 1.23 1.29 1.29 1.27 1.44 1.36 1.35 1.32 1.29 1.32 1.34
All Aircraft 66.3 66.1 66.8 68.2 70.9 74.7 77 80.2 81 85.1 88.1 85.1 83.8 82      
Fighters 65 64.4 66.3 68.3 72.6 76.1 77 80.2 81 85.1 88.1 85.1 83.8 82      
Bombers 51.6 50.1 44.1 42.7 40.9 45.5 69.1 72.1 71.4 86.6 88.1 76.6 84.8 80.3      
Tankers 67.5 67 65.3 65 65.5 68.9 72.6 78.8 79.6 84.2 82.5 76.6 74.6 82.3      
Other 69.1 70 71.2 73 75.1 79.7 83.9 83.8 84.3 84.4 85.7 84.6 84.3 85.4      
Notes: (1) Index values are based on combined rates for fighters and bombers, taking into account the different sizes of the two fleets. Indexed values are expressed in terms of fractions of the 1981 rates.

Source: Index values are derived from Congressional Budget Office, Paying for Military Readiness and Upkeep: Trends in Operation and Maintenance Spending (Washington, DC: CBO, September 1997) p.29, and consultations with the author. Other data is from Congressional Budget Office, Trends in Selected Indicators of Military Readiness, 1980 through 1993 (Washington, DC: CBO, March 1994) p.70-71.

A third set of data (page 32) comes from the Air Combat Command. A successor to the strategic and tactical air commands, ACC today controls the Air Force's bomber fleets, most of its special mission aircraft, and two-thirds of its operational fighters including all Reserve fighters. Like the CBO data (which covers a broader swath of the Air Force), the ACC data on operational fighters shows the Gulf War period to be a historic peak. Indeed, the whole period 1986-1992 registers quite high for ACC with an average MCR of 87.4. The years 1993-1996 average only a few percentage points lower -- 83.8. The ACC data also allow some quantification of the "hollow force" period of 1975-1980. The average MCR for the command's operational fighters during these years was 57% and the trough was 54% -- which is a long way from today's 75%.56

4.2 Cannibalization in context

An important component of recent readiness concerns has been the increase in aircraft cannibalization rates. A close look at the issue soon brings into focus some post-Cold War changes in the Air Force's circumstances -- specifically, in the structure of its air fleets -- that are relevant to interpreting not only cannibalization rates, but a variety of other readiness indicators as well: the size of war stocks, for instance, and the increasing average age of the air fleet.

Table 4: Air Combat Command Mission Capable Rate History

  1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Fighters 59.1 66.6 71.7 74.6 82.6 83.8 86.0 87.6 88.2 87.8 86.9 88.4 87.2 85.5 84.0 83.6 82.2 78.6 74.8 74.8
Bombers                     71.2 68.0 72.4 77.4 78.2 72.5 74.5 69.0 63.9 67.4
Special Mission                     84.3 82.0 83.7 82.3 82.4 83.2 81.5 77.4 75.1 75.6

Figure 8: Air Combat Command Mission Capable Rate History

Source: Director of Logistics, Ten Years Look Back: Standards and Performance FY89-FY98 (Langley AFB: ACC, November 1998).

During his January Senate testimony General Ryan reported a 78% jump in cannibalization rates since 1995 -- a period mostly comprising the years of spare parts shortage. Indeed, 28 of those percentage points had been added since September 1998. The rise had been from a 1995 airforce-wide rate of eight acts of cannibalization for every 100 sorties flown. This rose to 12 in September 1998 and then to 14 in January 1999. What is most disconcerting about these events is the rate of increase. Fortunately, this has a cause that is identifiable and discrete. But what should we make of the general magnitude of cannibalization today? How much of a buffer exists to safely absorb fluctuations in the supply of or demand for spare parts? One yardstick for measuring today's situation is the experience of the 1970's "hollow force" period, during which cannibalization rates peaked at levels 50% higher than today, persisting at significantly higher levels for more than a decade.57

Interpreting cannibalization rates requires resisting the notion that "cannibalization" is plainly bad, if not degenerate. In fact, it "has always been used as a production maintenance tool to compensate for parts shortages."58 At some low rate, the practice need not cut into a unit's basic fighting strength if the unit has more planes on hand than called for in its structure -- as all unit's do. Historically, a rule of thumb has been that wings retain backup aircraft on a ratio of one to three -- but there are other categories of "spares" as well. These serve multiple purposes, including standing-in for primary aircraft when they need maintenance. They can also serve as a source of spare parts or, for that matter, as flying "readiness spares packages."59

Clearly, there are limits to the utility of cannibalization. First, it makes no sense to procure extra planes solely for their parts -- and the other reasons for buying them would be defeated if the planes were always missing something vital. Second, cannibalization increases maintenance costs. Finally, cannibalization, like dipping into war stocks, always constitutes borrowing on the Air Force's strategic buffer. And the risk rises with the rate. At too high rates, it would involve borrowing dangerously on the capability of the force. What would constitute an ideal limit on cannibalization would vary from period to period, depending on the dynamics of spare part production, the efficiency of the supply system, and the size of the existing buffer of stockpiled parts and extra planes relative to possible war-time requirements.

Table 5: Air Combat Command Cannibalization Rates

  1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Bombers                       71.3 49.7 31.8 33.3 41.6 46.3 66.4 66.2 69.3
                      18.8 12.5 12.3 8.4 8.4 6.8 8.4 10.0 12.0
22.5 22.8 18.0 19.1 13.6 12.5 8.5 7.8 8.6 11.9 11.7 12.2 13.4 10.6 11.6 12.7 13.4 15.3 16.1 16.7

Figure 9: Air Combat Command Cannibalization Rates

Source: Director of Logistics, Ten Years Look Back: Standards and Performance FY89-FY98 (Langley AFB: ACC, November 1998).

Historically, the predecessors to Air Combat Command achieved an average cannibalization rate of 8.3 acts per 100 sorties during the best three years of the last two decades -- 1986-1988.60 Looking at more recent experience (but prior to today's unique spare parts shortage), ACC had an average rate of approximately 12.5 acts per 100 sorties for the years 1993-1995, inclusive. This suggests that ACC's recent cannibalization baseline is 50% higher than its best three years historically. (A straight-forward comparison of the past two decades, however, shows the 1980s to have a slightly higher rate than the 1990s). To make sense of comparisons like these, which span different historical periods, requires appreciating the different circumstances of those periods.

Complicating any direct comparison between the current cannibalization rate and that during the late 1980s is the fact that the ratio between force structure and aircraft inventory has changed. Regarding fighter and attack aircraft, there is a greater portion of the fleet outside of official Fighter Wing Equivalents (FWEs) than was the case in 1989. This means that the Air Force has more planes to draw on as spares or backups, at least theoretically. And this alters the implication of any particular rate of cannibalization.

FWEs are based on a count of combat-coded primary mission aircraft. These are drawn from the Primary Aircraft Inventory (PAI), which also provides similar planes -- in this case F- and A- planes -- for other purposes including research, development, and training. Out of this inventory, wings may also get some spares that do not count toward the FWE calculation. The next inventory level is the Total Active Inventory which includes the PAI, backup aircraft for wings, and the attrition reserve. Table 6 shows the change in the ratio among active-component inventories of attack, fighter, and interceptor aircraft.

The bottom row indicates how many fighter/attack aircraft the active-component keeps alive for each one that it counts towards its fighter wing equivalents. As the chart shows, since 1989 the ratio between primary mission aircraft and total inventory has grown by almost 20%. For some fighters the current ratio is even greater. For instance, in 1997, active-component fighter wings had slots for 454 F-16s. But the active-component's primary aircraft inventory held 666 F-16s and its total active inventory held 802.61

Table 6: USAF Active-Component Attack, Fighter and Inceptor Aircraft

  1989 1992 1997
Primary Mission
Aircraft Inventory (combat-coded) (1)
2005 1212 936
Primary Aircraft Inventory (2)   1673 1354
Total Active Inventory (3) 2896 1966 1610
PAI as factor of PMAI   1.38 1.44
TAI as factor of PMAI 1.44 1.62 1.72
Notes: (1) The Primary Mission Aircraft Inventory (PMAI) comprises those aircraft that count toward the authorized strength of the Air Force's squadrons -- in this case, active-component fighter, attack and interceptor squadrons.
(2) The Primary Aircraft Inventory (PAI) includes the PMAI as well as other sub-inventories covering functions such as training, testing and development. The PAI can also supply squadrons with "spares" that stand outside their authorized strength. (3) The Total Active Inventory includes the PAI, the Backup Aircraft Inventory, the attrition reserve, as well as some other categories of aircraft. The USAF also maintained an "inactive" inventory.

Sources: Secretary of Defense, Annual Report to the President and Congress (Washington, DC: US Government Printing Office, 1998); Secretary of Defense, Annual Report 1993; Office of the Under Secretary of Defense, Manpower Requirements Report (Washington, DC: Department of Defense, 1993); Manpower Requirements Report 1992; "Almanac: The Air Force in Facts and Figures," Air Force Magazine (May 1998); "Almanac," AFM (May 1993)

Spare aircraft and readiness spares, including war stocks, constitute material reserves that buffer the force against both unpredictable events -- such as military operations -- and fluctuations in the supply system. When supply unexpectedly runs short or demand runs high, some combination of three things may occur: increased aircraft cannibalization, increased use of parts stockpiles, or increased maintenance backlogs. The assessment of change in any one of these areas depends on an appreciation of both change in all three and the spare parts/spare planes safety margin that exists. Absent from the recent discussion of readiness has been a recognition that the Air Force's buffer of "spare planes" is substantially greater today than during the Cold War. The increase in the size of that buffer -- at least with regard to fighter, attack, and interceptor aircraft -- is on the order of five times the number of planes lost in the Gulf War. This should substantially transform the interpretation of cannibalization rates and spare parts shortages.

4.3 Optempo and the burden of operations overseas

4.3.1 Gauging the increase in operational tempo

The Air Force's post-Cold War readiness dilemma is supposed to have two horns. Resource constraints constitute the first. The second is increased Optempo. But how much has operational tempo increased since 1989? This question is key to determining the severity of the Air Force's long-term dilemma. A set of data assembled by the Congressional Budget Office gives a provisional answer for an important segment of the force -- all active-component bombers and fighter/attack aircraft, which together typically account for 30% of all USAF flying hours.

Flying hours

Table 7: Flying Hours Indexed to 1981
(Active-Component, Fighter Bomber Crews)

  1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Index Value 0.95 1 1.04 1.08 1.1 1.1 1.08 1.12 1.08 1.12 1.19 1.35 1.17 1.09 1.14 1.16 1.16
Source: Congressional Budget Office, Trends in Selected Indicators of Military Readiness, 1980 Through 1983 (Washington, DC: CBO, March 1984).

Table 7 shows the change in average crew flying hours during the period 1980-1996 for active-component combat aircraft, indexed to the average for 1981. Comparing the five years preceding the Gulf War period 1990-1991 to the five years after provides a good indication of change. During the years 1985-1989, USAF combat aircraft flew an average of 110% as much as in 1981, the base year. For the post-Cold War period 1992-1996, the average was 114.4%. Thus, between the pre- and post-Gulf War periods there was an increase in average flying hours per crew of 4%.

Operational deployments

Another common measure of increased Optempo is the number of personnel deployed on a temporary basis in military operations -- deployment tempo or Deptempo.62 This calculus is as familiar as it is simple: 3,500 personnel were so deployed in 1989 versus 13,700 in 1996 -- a four-fold increase. By September 1998, the number deployed on Temporary Duty had declined to 11,577, although the average for FY1998 was 14,000, reflecting the fluctuation in demand for operations against Iraq. During the same period the number of active-duty air force personnel declined by 35.6%; reserve component personnel have declined by 9.6%. Total Air Force uniform personnel has declined 29%. Within these statistics it is not difficult to see the contours of an Optempo crunch: a smaller force attempting to do more than its larger predecessor. But the picture is not complete because operational deployments reflect only a small fraction of the USAF's involvement overseas.

Foreign presence & deployments

  • In 1989, the Air Force had 126,400 people stationed or deployed overseas -- more than 22% of the active component.

  • In September 1998, by comparison, the Air Force had 62,538 people stationed or deployed in foreign countries -- 17% of its reduced active force.

Table 8: Change in USAF Personnel Overseas

  1977 1978 1979 Avg. 85-89 1990 1991 Avg. 92-96 1997
Total USAF worldwide 570,695 569,712 559,455 580,057 592,815 510,432 426,081 377,385
Total USAF foreign 99,765 105,862 107,402 119,196 117,206 99,311 71,020 60,450
Total USAF foreign as % of worldwide USAF 17.48% 18.58% 19.20% 20.54% 22.00% 19.46% 16.60% 16.00%
Sources: Directorate for Information Operations and Reports, Statistical Information Division, Active Duty Military Personnel Strengths by Regional Area and by Country (Washington, DC: Department of Defense, 1997); Office of the Comptroller, National Defense Budget Estimates for FY 1999 (Washington, DC: DoD, March 1998).

More telling than the difference in absolute numbers of troops in foreign territories is the reduction in the portion of the USAF deployed overseas. Broadening the basis of comparison shows the reduction to be characteristic of the post-Cold War transition:

  • During the last five years of the Cold War, the Air Force stationed or deployed an annual average of 22% of its personnel overseas; for the period 1992-1996, the average percentage was just under 17%.

Thus, a paradox of the Air Force's current readiness situation is that the service is more of a "stay-at-home" force today than during the Cold War. Qualifying this is the fact that the pattern of deployment also has changed: currently, a greater proportion of those USAF personnel overseas are on temporary duty. And temporary deployments are more difficult to manage than the more permanent ones -- although neither is inexpensive or easy.63

One way to compare the circumstances of the USAF in 1989 to those now is to imagine a force of today's size but with the deployment pattern that prevailed in 1989. This reduced version of the 1989 force can then be compared to today's actual force.

  • If the 1989 Air Force were scaled down to the size of today's while retaining the 1989 deployment pattern, it would have 79,000 personnel stationed "permanently" overseas and another 2,500 in temporary deployments.

  • Instead, today's actual force has 49,000 in permanent foreign bases and 13,500 (circa 1996) temporarily deployed.

The latter (actual) arrangement is probably less expensive in terms of incremental costs than would be the former. This suggests that, in terms of the financial burden from overseas presence, today's Air Force is better off than yesterday's. This estimate is based on the history of average incremental costs for the foreign stationing and deployment of U.S. troops (all services). The annual incremental cost to the United States of stationing troops in Europe is approximately $75,000 per person, averaged across services. (Another $20,000 is contributed by allies.)64 By comparison, the incremental cost of recent contingency operations ranges between $100,000 and $160,000 per person per year.65 War is another matter -- the Gulf War involved the equivalent of 280,000 troop/years at an incremental cost of almost $60 billion -- or approximately $215,000 per troop/year.66

Setting aside the issue of budgetary cost, temporary deployments are certainly more disruptive to military organizations than the routine stationing of troops abroad -- because they are less predictable in size, character, duration, and timing. A balanced portrayal of the change since 1989 in the demands associated with foreign commitments should encompass both long-term and temporary deployments, and it should convey a sense of scale:

  • In 1989, 21.5% of USAF personnel were stationed at overseas bases and another 0.6% were employed in temporary operations; today, 13.3% are stationed overseas, while another 3.7% are engaged in overseas operations.

The current pattern may not impose a higher budgetary cost, but it certainly poses a more difficult management problem.67 A mitigating factor is the absolute number of troops involved, which is small. Policy-makers should keep scale in mind. The current discussion revolves around the difficulty of managing an operational deployment rate of less than 75 out of every 1000 deployable air force personnel.

Hours spent flying support for operations

The Optempo burden of contingency operations also can be measured in terms of flying hours. During the years 1991-1995, USAF aircraft flew 800,000 hours in support of contingency operations -- out of a total of 13.5 million. Thus, contingency operations consumed about 6% of the Air Force's flying time in this period. With the exception of FY1992, annual effort fluctuated in the 150,000-170,000 hour range. In no year did it exceed 8% of total USAF flying hours.68 This measure gives an initial sense of the extrinsic challenge that recent contingency operations impose on the Air Force as a whole. However, there also are challenges intrinsic to the Air Force concerning how the service manages the demands that operations impose on it

4.3.2 Managing operational tempo

Uneven burdens: select USAF commands are under stress

The weight of contingency operations has fallen more heavily on some parts of the service than others. During the 1991-1995 period, the active component flew about 90% of all peace operations sorties. With few exceptions, however, less than 10% of the annual sorties generated by the active component's fighter, tanker, C-130 cargo, and special operations units were devoted to supporting peace operations.69 The electronic combat and surveillance fleets faced a higher demand. On average, electronic combat squadrons devoted between 12% and almost 25% of their total sorties to such operations during this period. And the surveillance aircraft fleet gave even more -- between 25% and 35% of their total sorties.70

A 1998 RAND Corporation study, Air Force Operations Overseas in Peacetime, shows how recent contingencies weighed differently on the different major commands. The study measures activity level in terms of Temporary Duty (TDY) rates -- days spent away from home station. It tracks the experience of Air Force flying units during four years -- 1994-1997 -- distinguishing the units by type of aircraft flown and by command. In every case it uses as a yardstick the goal of keeping temporary duty assignments limited to no more than 120-days per year.

Looking at fighter and attack aircraft units the study finds that all types belonging to the Pacific Air Forces (PACAF) were able to average under the 120-day threshold for the entire period. Indeed, for most years, the Pacific units averaged substantially under 100 days. The experience of attack and fighter units belonging to ACC or to USAF Europe (USAFE) was quite different. In 1994 and 1995 most USAFE unit types averaged close to the 120-day limit, and A-10 units averaged significantly above it. USAFE found some relief in 1996 and 1997, with all unit types retiring to below the 100-day level.

The experience of Air Combat Command was a mirror image of USAFE. In 1994, its attack and fighter aircraft averaged safely below the 100 day threshold. By 1995, however, its F-16s and F-15s were flying close to 120 TDY days. Although the situation for most types of ACC fighters had improved by 1997, its A-10's climbed upward toward an average of 150 TDY days -- just as USAFE's A-10s were climbing down.

Figure 10: Discrepancy in Decline of ACC Mission Capable and Unit Readiness Rates, Indexed to 1995

Sources: Reproduced from David E. Thaler and Daniel M. Norton, Air Force Operations Overseas in Peacetime (Santa Monica, CA: National Defense Institute-RAND, 1998) p.9.

The RAND data make clear that F-15 and F-16 units could have averaged significantly below the 120-day threshold for the entire period -- if Optempo had been more evenly distributed. In other words, the fact that some units spend excessive days away from home station is partly a problem of Optempo management. (For A-10 units, the problem seems to be a straight-forward shortage in the supply of units and crews). The RAND study notes that between 1995 and 1997 the Air Force made progress in distributing burdens more evenly across the active force. Indeed, USAFE was substantially relieved. But relief for ACC was less evident -- and this undoubtedly contributed to the 1997-1998 meltdown in its mission capable rates.

PACAF units on average, although more active in peace operations by 1997, remained substantially below 100 TDY days. As the RAND authors note, "the CINC in the Pacific theater severely restricts the use of forces based in Korea and Japan for TDY operations in other regions, resulting in lower TDY rates on average for PACAF forces than for ACC or USAFE forces."71 Indeed, the Pacific Command's TDY goal is 90 days rather than 120, although a variety of its units exceed this limit today.72

Toward balanced utilization of USAF fighter assets

The 1998 RAND study concluded that the Air Force had sufficient fighters to meet current contingency requirements while observing a limit of 120 average TDY days for each type of unit. The study set contingency requirements to accord with the deployment level of September 1996 -- 146 aircraft or 2.03 Fighter Wing Equivalents (FWE). Against this goal, it found that the USAF could field 2.14 FWEs continuously. This conclusion rested on several key assumptions:

  • All PACAF aircraft in Korea and Japan were assumed unavailable for contingencies in other regions; and,

  • U.S.-based aircraft were assumed to give 70 TDY days (out of a total of 120) to contingencies, using the rest for training, competitions, and exercises. Europe-based units were assumed to give only 60 TDY days to contingencies; Reserve units, only 10.

The calculus changes, however, if the 120-day TDY limit is accepted as a maximum for individuals, and not just as an average applied to unit-types. Only a perfectly efficient organization could ensure that all crews posted no more than the average number of days for their unit-types. To allow for some inefficiency, the RAND authors lowered the permissible unit average to 100 TDY days for the active component. The 20 reduced days were subtracted entirely from the pool allotted to contingency operations. Under these assumptions, today's force cannot meet the assumed requirement. But the RAND authors suggest two policy innovations that could close the gap and allow the force to meet operational requirements while observing a 100-day TDY limit for units -- first, aircraft in Korea and Japan could be made available for 40 TDY days outside their area; and second, Reserve aircraft could be made available for 20 TDY days per year.

The idea of increasing PACAF and Reserve participation is a good one. Indeed, the Air Force has increased its reliance on reserves since the RAND study was conducted. However, the authors' conclusions are too pessimistic, for several reasons.

First, it is not necessary to set aside 50 TDY days for U.S.--based units to participate in off-station training, competitions, and exercises. Less will do; and, thus, more is available to meet contingency needs. (This proposition is explored further in Section 4.4.1).

Second, the 120-day limit for individuals can afford some individual variation. What the Air Force must preclude is an Optempo rate that requires individuals to serve in excess of 120-days on TDY every year for several years. A study of the most severely stressed air units, conducted during 1995-1997 by the Air Force Studies and Analyses Agency, found little near-term correlation between Perstempo and separation from the service: "Where Perstempo begins to wear out the carpet is after three or four years in a row."73 This suggests that units might alternate high and low rates on a biennial basis. In addition, some individuals may be willing to serve more often; within safe limits, they should be allowed to do so.

Loosening the assumptions of the RAND study while adopting its proposals for increasing Pacific and Reserve participation in contingency operations opens the possibility of creating substantial breathing room for the Air Force's fighter fleets.

Air Lift optempo

Figure 11: Temporary Duty Rates of USAF Active Lift/Refueling Crews

Sources: Reproduced from David E. Thaler and Daniel M. Norton, Air Force Operations Overseas in Peacetime (Santa Monica, CA: National Defense Institute-RAND, 1998) p.12.

Reviewing TDY days for airlift and refueling aircraft during the years 1994-1997, the RAND study shows all unit types in all three major commands as well as the Air Mobility Command (AMC) remaining below the 120-day threshold for FY1995 and FY1996. However, in FY1994 ACC C-130 crews flew 125 days as did AMC C-130 crews in FY1994 and FY1997. Notably, in both these years, European and Pacific C-130 crews flew less than 100 TDY days -- substantially less in the case of USAFE.

The AMC's C-141, C-5, and C-17 crews were able to keep below 100 TDY days all four years. AMC achieved this by having Reserve associate crews augment the active component -- a good example of creative Optempo management. Nonetheless, the TDY trend was upward from the beginning of the period to the end.

How does the activity level of strategic airlift in 1989 compare with that of the post-Cold War period -- 1994-1997 -- covered in the RAND report? Looking back to 1989, Table 9 compares the activity level of C-141's and C-15's in that year and 1994, with a focus on pilot stress.

Table 9: Flying Hours for C-5 and C-141 Aircraft and Pilots,
1989 and 1994

  1989 (MAC) 1994 (AMC)
flying hours (000) 49.5 58.2
number of pilots 342 439
hours per pilot 145 133
flying hours (000) 258 103.7
number of pilots 1199 842
hours per pilot 215 123
(corrected for 1/4 of fleet being Not Mission Capable)   (164)
MAC=Military Airlift Command; AMC=Air Mobility Command

Source: Derived from a table based on USAF data that appeared in James Kitfield, "Airlift at
High Temp." Air Force Magazine (January 1995).

This comparison shows a decline in flying hours per pilot between 1989 and 1994. It may be that in the years subsequent to 1994 the demands placed on C-141 and C-5 pilots again rose toward 1989 levels. The independent RAND study certainly shows that crew TDY's rose. An important proviso to the 1989 data is that it reflects support for Operation Just Cause, during which these and other lifters moved 21,000 tons of cargo. This would have elevated 1989 flying hours by a few percent. Nonetheless, what neither this data nor the RAND study supports is the proposition that airlift crews on average face a burden today that is substantially greater than that of the Cold War period. Where and when excessive burdens have occurred, their source has been the maldistribution of Optempo among the service's assets and commands.

Uneven burdens: select systems under stress

Turning finally to the experience of electronic warfare, reconnaissance, and command and control units: Most of these belong to ACC, and the RAND study graphically illustrates the demand placed on them by contingency operations. In this case, almost all unit types -- except a handful assigned to PACAF -- averaged between 120 and 180 TDY-days during 1994 and 1995. The experience of these aircraft, as well as that of A-10s, illustrates another way in which Optempo is unevenly distributed: by system. In these cases, the problem is a straight-forward mismatch between supply and demand, and therefore less amenable to simple managerial solutions. Instead, solutions would have to be sought along either one of two avenues: (i) restrict the use of these assets or (ii) add aircraft, units, or crews to the force. The RAND authors conclude that "Specialized aircraft are experiencing a rate of utilization well beyond the level that the current force structure would seem able to support on a long-term basis."74 Nonetheless, since 1995 the TDY trendline for these systems has been downward -- with the exception of the U-2.

Figure 12: Temporary Duty Rates of USAF Active C2/EW/Recce Crews

Sources: Reproduced from David E. Thaler and Daniel M. Norton, Air Force Operations Overseas in Peacetime (Santa Monica, CA: National Defense Institute-RAND, 1998) p.12.

The Air Force was attuned to the problem of "high demand/low density" (HD/LD) systems as early as 1994, setting in motion a study the following year under the auspices of the Air Force Studies and Analyses Agency.75 Since then, these "stressed systems" have become emblematic of the readiness crisis, although their situations are in many respects unique.76 At the center of concern were a variety of specialized craft: U-2s, EC-, HC-, and MC-130s, and E3 (AWACs). One early focus of the study was the RC-135 Rivet Joint whose crews had logged average TDY's in 1994, 1995, and 1996 of 152, 163, and 151 days, respectively -- compared with a pre-Desert Storm norm of 116 days. The Cold War norm was based on a requirement for 2200 crew days per year allocated among nineteen crews flying ten planes. Moving into the 1990s the requirements for this system actually increased in absolute terms -- to 3,057 crew days in 1995.77 This absolute increase in requirements is a feature that distinguished the RC-135 and some of the other specialized aircraft as well.

While adaptation to the new strategic environment was requiring the Air Force to implement painful reductions in personnel and its fighter fleet, conditions also called for increases in some types of support aircraft or, at least, in their crew ratios. To bring RC-135 units within acceptable TDY limits would have required the addition of seven crews. Another possible approach to reconciling supply and demand was a redefinition of how these assets were to be used under new conditions -- a move that would have screened the demand for these planes or refigured their modes of operation. Initially, change did not occur and, as a result, the Perstempo rate shot upward. After 1995, however, the Air Force was able to gradually reduce RC-135 TDY's through a combination of measures. By 1998, almost 65% of all crews logged less than 90 TDY days; about 18% remained above the 120-day limit.78 The annual average TDY rate had fallen to below 105 days. The addition of two more RC-135s in the FY1999 budget should further reduce the stress on these systems.

The Stressed System Study looked not only at low-density systems that had been overwhelmed by demand, but also at several occupational specialties -- such as Tanker Airlift Control Elements -- that seemed short staffed. Finally, it looked at units that were registering elevated stress factors, such as very high TDY rates, even though the systems they supported did not fit the HD/LD profile. In these cases, the question was: What combination of factors had pushed the units into the red zone? One example was the 50th Airlift Squadron, whose operations personnel averaged 136 TDY days in 1996. In this case a key factor was participation in major exercises. The unit had taken part in more than nine such exercises, although two or three would have been an acceptable standard.79

Another telling case is that of the A/OA-10, which won the distinction of being the service's most stressed "shooter." Although by no means a "low density" system, the demand for A/OA-10s outstrips their supply. Despite its performance in both the Gulf War and current peace operations, the status of the A-10 in the Air Force simply does not match its proven utility.

By 1997 the second phase of the Stressed System Study found discernible improvement in the problem: only four systems surpassed the 120-day limit in 1997, as opposed to 13 in FY1994. This was accomplished principally by increasing crews and convincing regional CINCs to do with less in some operations.80 Finding that the number of AWACs systems and crews averaging more than 100 days away had increased from 17 in 1994 to 24 in 1996, the Air Force quickly added crews and formed an Air reserve associate wing. The Air Force also turned to "outside" assistance to relieve stressed systems, relying more on Navy EA-6B's and NATO AWACs.

4.3.3 Optempo and training readiness

A central element of USAF readiness concerns is the effect of high operational tempo on training. Concerns about training derive from a perceived decline in the combat skills of pilots returning from overseas deployments. Instructors at the Air Force Weapons School, for instance, reported a decline between 1995 and 1996 in the basic maneuver skills of pilots who had recently completed tours in Bosnia and Iraq.81 Operational deployments purportedly diminish skills by preempting training events or disrupting their cycle, while failing to add much to pilots skills. Another possible source of reduced training is the diversion of funds by wing commanders from training accounts to meet base operating expenses, although this applies principally to the special circumstances of 1997 and 1998, and not the longer-term issue of training readiness.82 The real test of training readiness (as well as readiness overall) is performance in war. For obvious reasons, the nation cannot afford to take a "wait and see" stance. Nonetheless, performance in on-going operations also provides an indisputable reality check on the readiness condition of the military. Another indicator of problems in the Air Force would be a deterioration in the rate of flying mishaps, which correlates both with increased fatigue and weakening of training regimes.83 In this regard, the 1994 shoot-down of two Army helicopters by USAF F-15s over Iraq was interpreted by some as indicative of readiness problems.84 However, looking at the history of flying accidents for the Air Force as a whole shows no notable deterioration of rates. (See Table 10. Figure 13). Except for a brief tick in FY1994, the transition from a Cold War to post-Cold War posture has continued the downward trend in accident rates. This evidence alone cannot preclude the possibility that problems are in the offing, but it should mitigate concerns about the extent of problems so far: they have not been sufficient to affect accident rates. What might eventually lead to performance problems, however, is the curtailment of training regimes. Hence, its worthwhile to take a closer look at how today's Optempo is impinging on training.

Table 10 and Figure 13: Aircraft Mishap and
Destroyed Rates Per 100,000 Flying Hours

  Class A Mishaps Fatal Mishaps Aircraft Destroyed
CY 1974 2.9 1.3 2.4
CY 1975 2.8 1 2.4
CY 1976 2.8 1.1 2.2
CY 1977 2.8 1.2 2.5
CY 1978 3.2 1.3 2.9
CY 1979 2.9 1.5 2.6
CY 1980 2.6 1.3 2.3
CY 1981 2.4 1.2 2.3
CY 1982 2.3 1 2.3
FY 1983 2 1 2
FY 1984 1.5 0.8 1.5
FY 1985 1.8 0.9 1.6
FY 1986 1.5 0.6 1.6
FY 1987 1.6 0.9 1.6
FY 1988 1.6 0.7 1.4
FY 1989 1.6 0.7 1.4
FY 1990 1.5 0.6 1.5
FY 1991 1.1 0.3 1
FY 1992 1.7 0.7 1.8
FY 1993 1.4 0.4 1.4
FY 1994 1.5 0.5 1.5
FY 1995 1.4 0.5 1.3
FY 1996 1.3 0.3 0.9
FY 1997 1.4 0.4 1.2
Source: Assistant Secretary of the Air Force, United States Air Force Statistical Digest FY 1997 (Washington, DC: SAF/FMC, Air Force, Pentagon, April 1998).

A 1997 RAND study, Preparing the U.S. Air Force for Military Operations Other Than War, examined the possible effect of contingency operations on training readiness. The study reviews average training hours during the period 1988-1995 for various types of aircraft and also looks at the hours they spent supporting contingency operations.85 To assess recent training trends the authors establish a "Cold War standard" for training: average flying hours for the years 1988 and 1989. While the adoption of a comparative approach to today's problems is laudable, their method suffers from two weaknesses:

  • First, it treats all flying hours during 1988 and 1989 as making an equally important contribution to skills development. This masks real differences between dedicated flight and weapons training, various types of exercises, channel flights by lift aircraft, and simple transits.

  • Second, while their putative "Cold War standard" for annual flying hours is the average of 1988 and 1989, the difference in how many hours were flown each year was substantial: generally speaking, units spent less hours flying in 1989 than in 1988. Because of this variance, a more reliable "Cold War standard" would have had to include several more years in the average. Absent this, the two years could be taken to establish an acceptable range of flying hours.

Figure 14: Flight Hours for Operational Training and Peace Operations Relative to Those for the Cold War Standard, All F-16s: 1988-1995

Sources: Reproduced from David E. Thaler and Daniel M. Norton, Air Force Operations Overseas in Peacetime (Santa Monica, CA: National Defense Institute-RAND, 1998) p.12.

Figure 15: Flight Hours for Operational Training and Peace Operations Relative to Those for the Cold War Standard, All F-15s

Sources: Both charts are from Alan Vick, et al., Preparing the U.S. Air Force for Military Operations Other Than War (Santa Monica, CA: National Defense Institute-RAND, 1997) p.31 and p.179.

These objections do not detract from the value of the study. By contrasting 1988 and 1989 with the years that followed, the study helps put today's experience into perspective.

The recent training history of the Air Force's F-16s, F-15s, and A-10s is especially relevant to current readiness concerns. Reviewing the total force of these units shows their training hours during 1990-1995 to be near, at, or above the 1988-1989 average -- with one prominent exception: During 1993-1995 the A-10 fleet fell toward 80% of the Cold War standard.86

Figure 16: Flight Hours for Operational Training and Peace Operations Relative to Those for the Cold War Standard

Sources: Alan Vick, et al., Preparing the U.S. Air Force for Military Operations Other Than War (Santa Monica, CA: National Defense Institute-RAND, 1997) p.181

Disaggregating the averages for the total force reveals substantial variations in the experience of different USAF commands and components. And this is where real problems may reside. While PACAF and Reserve units generally flew well above the putative "Cold War standard" for training hours, ACC and USAFE averages for F-16 and A-10 units often fell below it. Indeed, European A-10s in 1993 and 1994 fell as low as 50% the standard, before recovering to 80% in 1995. Otherwise, average training hours seldom fall below 80% of the standard, regardless of component or command. F-15 units in all subdivisions achieve averages comparable to or better than the standard.

The conclusion that average Air Force training hours remained close to late-Cold War levels during the period 1992-1997 does not contradict evidence -- anecdotal or statistical -- of localized declines, including the oft-quoted 1996 report from the Air Force Weapons School. Instead, it simply argues against generalizing this evidence to infer a service-wide "demand-supply" problem -- ie. more demand for operational activity than the service's resource base can possibly support without cutting seriously into training time. There is nothing contradictory in the idea that an organization can be adequately resourced for its mission and, nonetheless, fail to allocate those resources internally in a way that prevents localized shortages.

Reducing the draw on training time

One solution to localized shortfalls in training hours is to better distribute those activities that cut into training time -- for instance, hours spent in conducting peace operations. Sharing operational burdens implies sharing the opportunities for training time. In this way, all the subordinate parts of the Air Force might come to approximate the total force average for training hours, which is reasonably close to the Cold War standard.

Another possibility is to re-engineer Air Force peace operations so that they consume fewer flying hours. Along these line the RAND authors propose that continuous air patrolling in some peace operations might be replaced by a combination of measures:87

  • greater reliance on remotely-piloted aircraft and a network of air-delivered ground sensors; and

  • a "cop-on-the-beat" approach to using regular aircraft, whereby smaller random patrols aim to deter prohibited activity rather than interdict every instance of it. A more robust response would be held in reserve, both in-theater and back home, to be used as needed.88

In order to keep the training hours crunch in perspective, however, policy makers should also ask, How has the Air Force's need for training hours changed since the Cold War's end? Certainly the demise of a peer adversary adept at air defense must make a difference. Has increased use of simulation had any impact at all on training hour requirements? And finally, Does Air Force activity in peace operations make any positive contribution to flying skills?

The 1997 RAND report recognizes that peace operations afford pilots the opportunity to practice some routine activities. The authors' concern, however, focuses on more specialized abilities: "Peace-operations sorties provide fighter crews with virtually no opportunity to maintain their proficiency in many of their most important and perishable combat skills."89 This view, which has many adherents, tends to treat all current contingency operations as identical, which they are not. The most demanding set of current Air Force operations are those in Southwest Asia. Between the end of the Gulf War and mid-1997, the USAF flew approximately 110,000 sorties in contingency operations around Iraq; By comparison, over Bosnia it had flown only 30,000.90

Today Southwest Asia claims two-thirds of all USAF personnel deployed in contingency operations. The operations in Southwest Asia involve a broad range of reconnaissance, deterrence, interdiction, and defense activities. These include fairly conventional preparations for border defense -- not unlike some activities that preoccupied the U.S. military for fifty years along the divides in Europe and Korea. Today the Persian Gulf is one of two areas that concern U.S. policy-makers as potential near-term sites for major regional conflict. It would be myopic to ignore the value of having a large cadre of pilots who are intimately familiar with the details of the region.

The proposition that all the hours spent flying support for peace operations cannot simply substitute for combat training is a modest and acceptable one. The issue is whether these hours have any training value at all. If they do, they may close the narrow gap between today's total force training hours and the Cold War standard, allowing policy-makers to focus undivided attention on the real issue: not peace operations, but Optempo management.

Turning from the issue of attack and fighter assets, the RAND study points out that the activities of tanker, airlift, and surveillance units in peace operations are much like those in war. In the case of these units, the hours spent flying support might substitute for some training time. This would require reconfiguring their training regimes to "fill in" whatever gaps remain. Although the training averages for most of these unit-types have been near or above the Cold War standard, they all tend to carry a heavy Optempo burden. That burden could be mitigated by reducing and reconfiguring their training routines.

The single-minded focus on peace operations as the cause of training shortfalls diverts attention from both other sources of Optempo pressure and from the option of improving the efficiency of training hours. Although, historically, all flying hours were counted as contributing to readiness, this mode of measurement was never considered precise. Rather, there was a recognition that within every hour of flying there was some valuable opportunity to enhance skills. Inadvertently, the recent determination to divide flying time into "relevant" and "non-relevant" hours has highlighted the fact that what matters is not the quantity of hours, but their quality. Along these lines the 1997 Marine Aviation Campaign Plan sought to refocus training efforts on objectives rather than the accumulation of hours. The deputy chief of staff for USMC aviation, Lt. Gen. Frederick McCorkle reported in late 1998 that the effort had dramatically improved the efficiency of flight training: "I go out now and fly a sortie and accomplish probably twice as far as being capable to go to war."91

4.4 Toward a broader view of optempo

How Air Force wings and squadrons spend their time does not divide neatly into training and support for operations, of course. Broadening our view of "tempo" disposes of the simple, binary opposition between "contingency operations" and "readiness" or "retention." The experience of the 50th Airlift Squadron is instructive in this regard. As the Stressed Systems Study found, the key contributor to this unit's excessive amount of TDY days in 1996 was participation in too many exercises.

The USAF manages a heavy schedule of exercises and peacetime engagement activities other than contingency operations. The 1999 USAF posture statement asserts that in 1998 service members "participated in over 1,600 exercises in 35 countries, and conducted almost 300 military-to-military contact visits in Europe and the Pacific."92 Many of these exercises serve something more or something other than training needs -- notably, "environment shaping" objectives. A July 1998 GAO study of the 1,405 JCS exercises conducted or planned for the period 1995-2002 found that 44% were primarily directed toward engagement activities with foreign military forces.93 In September 1998 testimony before the U.S. House subcommittee, the commander of Pacific Air Forces, General Richard Myers noted that the Pacific Command alone conducts an intensive exercise program involving 23 Pacific nations.

Despite a recent decision to reduce by 15% USAF participation in JCS exercises, the cost to the United States of combined exercise may increase. In his congressional testimony, General Myers warned that economic turmoil and hardship is pushing the region's nations toward limiting the burden that exercises place on them -- and increasing the burden carried by the United States:

Specifically, we see our counterparts asking for our help in reducing their exercise costs and in reducing the appearance of extravagant spending on the part of their militaries. We are starting to see requests for additional supply support such as fuel, munitions, and rations, during exercise execution. Additionally, our partners are asking us to increase our share or assume the total cost for basic infrastructure support, facility, and equipment maintenance like aircraft arresting system repairs, ramp maintenance, and the aircraft ground equipment maintenance required to conduct exercises.94

Many peacetime engagement activities and exercises are small. But the stress that units suffer is due not only to the magnitude of the demands placed on them, but also to the quantity, variety, and frequency of them. These demands arise at levels throughout the service hierarchy and also from outside the service -- vertically from the Joint Chiefs and laterally from regional CINCs -- but they accumulate at wing and squadron level.95 Notably, the levels at which all these demands converge are not levels empowered to simply set priorities among them. And so wing and squadron commanders must often take time and resources out of those activities that they can control -- home-station training and base operations.

4.4.1 "Tempo" at wing level

A 1998 RAND study of three USAF wings examined the full array of demands that drive Optempo and weighed their likely effects on readiness and quality of life.96 The study was based on interviews with 500 commanders and supervisors representing 80% of the units comprising the wings. Unit leaders and supervisors were asked to specify the amount of effort expended in each of ten activity areas and to estimate their impact on readiness, professional development, and quality of life. The activity areas and percentage of total time consumed by each were:

  • Routine peacetime activities and training (53.5%);
  • Off-station training (4.5%);
  • Local tasking for activities such as air shows (4.5%);
  • Higher headquarters' taskings for special activities (4%)
  • Inspections and preparations for inspections (8.5%);
  • Wing exercises with multiple squads (10.5%);
  • Command exercises involving units from other wings (3%);
  • Joint exercises involving units from other services (3%);
  • Combined exercises involving units from other countries (1.5%); and
  • Peacekeeping and other contingencies (7%).

Figure 17: Average Time Allocation: Three Wings, 8th Air Force

HHQ=Higher Headquarters
Sources: Thomas Fossen, et al., What Helps and What Hurts: How Ten Activities Affect Readiness and Quality of Life at Three 8AF Wings (Santa Monica, CA: National Defense Institute-RAND, 1997).

Figure 17 clusters these activities in five major categories. Of the three wings examined in the study, one was a fighter wing -- the 27th. It devoted the greatest amount of time to contingency operations -- 10.5% of its total. Interestingly, it also spent more than the average time engaged in off-station training -- 5.5%. It compensated by spending less time in routine activities and slightly less in inspections and exercises. At the other end of the spectrum was the 5th Bomber Wing, which flies B-52s. It spent less than 4% of its time engaged in Missions Other Than War (MOOTWs) and more than 40% engaged in inspections and exercises.

Evaluating the impact of these activities on readiness and quality of life, the respondents judged that:

  • Inspections and wing exercises were the most intensive activities of the group, followed by contingency operations and joint exercises;
  • Special tasking by local or higher headquarters had the most adverse affect on readiness, followed by wing exercises and contingency operations; and
  • Contingency operations and wing exercises were the most detrimental of the activities to professional growth and family life. Inspections also scored poorly for its impact on family life.

Although inspections and exercises are designed partly to enhance readiness, they earned low ratings because the respondents felt that they too "often entail fruitless activities that consume inordinate amounts of time."97 Thus, the critical assessment of these activities concerned their efficiency. Contingency operations, inspections, and some types of exercises earned low ratings for their impact on family life and professional development, either because they intensified work schedules or required extended days away from home.

Another issue raised by respondents was the frequency of inspections and exercises, which some felt was counter-productive because it left too little time to "identify mistakes...and then develop and implement corrective action."98 Thus, the practices were felt to be disruptive to the process they were meant to serve. Similar criticisms have been raised about the practice of sending units into intensive training for Operational Readiness Inspections soon after they return from contingency operations.99 Both cases highlight failures in how the cycle of activity is managed, and both illustrate how, under some circumstances, simply "doing more" readiness-related activity can actually degrade readiness.

Figure 18: Theoretical Annual Schedule: F-15 Squadron, 1st Fighter Wing

TDY=Temporary Duty
Sources: Michael C. Ryan, Military Readiness, Operations Tempo and Personnel Tempo: Are U.S. Forces Doing Too Much? Washington, DC: Congressional Research Service, 14 January 1998).

Figure 19: Actual Schedule: F-15 Squadron, 1st Fighter Wing

ORE=Operational Readiness Exercise; TDY=Temporary Duty
Sources: Michael C. Ryan, Military Readiness, Operations Tempo and Personnel Tempo: Are U.S. Forces Doing Too Much? Washington, DC: Congressional Research Service, 14 January 1998).

The experience of the 1st Fighter Wing, which has gained some prominence for excessive TDY days, provides another perspective on how squadrons spend their time.100 Figures 18 and 19 depict two views of a fighter squadron's annual time allocation -- one ideal and the other actual.101 The actual schedule reflects the typical experience of F-15 squadrons in the 1st Fighter Wing. (Note that the standard of measure is days, not weeks; thus, the chart is not directly comparable to those based on the RAND study.) The "Other TDY" category encompasses some off-station training, competitions, and exercises.

Figure 19 shows the effect of unplanned additional exercises (some TDY and some not), inspections, and days supporting contingency operations. Essentially, unplanned events steal 41 days from the schedule -- six from free time and 35 from routine training and base activities. Moreover, in order to fit training into its reduced 125-day time slot, the squadron would undertake between 10 and 50 surge days, during which workloads would be substantially increased. Figure 19 also shows that contingency operations are neither the only source of TDY days, nor the only source of unplanned activities. They account for 73% of the former and 19.5% of the latter. Figure 20 presents a potential solution to the squadron's Optempo problem that allows for some reform in each major area of activity.

4.4.2 Managing optempo

The first prerequisite of finding comprehensive and robust solutions to the Air Force's Optempo-related problems is defining these problems and their sources correctly. To date, a major impediment to progress has been the subsumption of the Optempo problem in a political debate over national security strategy, focusing specifically on peace operations.102 That debate is properly political, but it bears no necessary relationship to the more technical question of whether today's Optempo problems are solvable within existing resource constraints. As a result, outside the Air Force itself, the Optempo issue has been approached too narrowly as a side-effect of peace operations. A more appropriate perspective would review all the activities of the force, and it would allow that the process of adapting to the new era is not yet complete -- perhaps not even substantially complete.

Figure 20: Potential Solution: Fighter Squadron Time Allocation

A second necessary adjustment to problem definition is the recognition that the Optempo issue is principally one of resource management. Only exceptionally are today's problems strictly due to resource shortages. Of course, adding resources can mitigate problems of all types -- but this does not mean that opening the budget spigots is either necessary or efficient. To its credit, the Air Force has focused most of its attention on tackling the Optempo issue as a management problem.

Since 1995 there have been a variety of remedial efforts and proposals for change. These can be divided into several categories distinguished by the aspect of the problem that they propose to attack and by their scope of application:

  1. Efforts to more fully utilize the entire pool of resources at the Air Force's disposal by distributing demand more evenly across the force and by finding alternate means of meeting surplus demand;

  2. Efforts to reallocate existing resources so that the supply of "high demand" capabilities and occupational specialties can better meet the demand for their employment without incurring pockets of severe stress;

  3. Efforts to minimize the disruptive effect of increased Optempo by protecting unit cohesion and lending predictability to the demands that accumulate at wing level; and

  4. Efforts to restrict and reduce the demands placed on units. These efforts have focused either on stressed systems and units, in which case the aim is to mitigate localized problems as they begin to reveal themselves, or on the functioning of the Air Force as a whole, in which case the aim is to alter those practices and protocols that generate problems. This category includes: (a) efforts to better screen the operational demand placed on units by higher-level authorities, especially regional CINCs, and to establish stricter priorities for the use of assets, especially low density ones; (b) efforts to limit various routine activities -- such as exercises and inspections -- in accord with demonstrable requirements, or to find more efficient ways of conducting these activities; and (c) efforts to set tighter limits at the unit level on how much activity can be imposed on units and their personnel.

  5. The final category comprises efforts and proposals to revise the doctrinal protocols that dictate how and how many assets are used in field operations. Here the aim is to increase the efficiency of Air Force operations by ensuring a closer fit to the real requirements of new era contingencies.

Specific measures have included:103

  • The goal of a 120 day limit on the amount of time individuals spend on TDY, established by former Air Force Chief of Staff General Fogelman;

  • The 1996 Global Military Force Policy, which sought to prioritize the use of special assets and to facilitate the sharing of comparable capabilities across services;

  • AMC's Phoenix Pace program, which sought to bring all wing personnel home for at least two-weeks a year in order to regroup as a unit;

  • Increases in pilot authorizations and crew ratios for a variety of unit-types including U2s, HC-130s, AWACs, and RC135s; and the procurement of additional RC-135s;

  • Reliance on Reserve support or Reserve Associate Wings for a variety of high-demand systems;

  • The scaling back of exercises and competitions beginning in 1997 and including the 1998 announcement of a servicewide plan for FY1999 and FY2000 to reduce by 5% the scale of USAF and Joint exercise and by 15% the number of USAF troops participating in JCS exercises;

  • The 1998 announcement of a reduction in Operational Readiness Inspections, with the aim of integrating these inspections with force deployments;

  • A limitation on deployment periods of no more than 45 continuous days for pilots supporting operations in Bosnia and Iraq, followed by guaranteed down-time for pilots on their return home and a period of refresher training; and

  • Prioritization of AMC air movements and increased reliance on the reserve and commercial airlines where suitable.

The most ambitious effort to tame the tempo problem is the reform of force employment practices along the lines of the Aerospace Expeditionary Force (AEF) concept.104 This would have most USAF combat and combat support assets distributed among ten AEFs. Although these would not replace wings as the organizational home of the Air Force's aircraft, they would function for the purposes of deployment as virtual composite super-wings, each comprising 175 aircraft. Two of these would be available for deployment at all times, with the remainder moving through cycles of down-time, maintenance, training, and exercises. A full AEF cycle would be 15 months, with deployment limited to three months preceded by a three month period of joint training. Units assigned to the same AEF could and would begin their coordination months before their joint training period began.

Each of the two AEFs available for deployment would station 75 aircraft overseas; 100 would remain on-call at home. Complementing these forces would be the specialized fleets of electronic warfare, command and control, and reconnaissance aircraft, available as needed. Each AEF would be prepared to work as a single unit or in smaller package, potentially deploying to more than one place at a time. Assets stationed in Korea and Italy would operate outside this scheme. And, of course, all forces would be expected to deploy as scheduled for major wars, should they arise.

The concept has three goals: first, to guarantee CINCs reliable access to balanced force packages; second, to minimize the disruptive effects of deployment by making deployment more predictable; and, finally, to structurally embed the goal of limiting personnel TDY to 120 days a year.

How well the AEF concept will work in practice remains to be seen -- it requires a unique degree of lateral coordination and communication. And because it is an overlay on the existing structure of commands, air forces, and wings, there is potential for friction. Nonetheless, it stands as a good example of reforming Air Force practice along the lines of an information-age military -- not because of the type of equipment it employs, but because of how it organizes itself.

Apart from the AEF initiative, the other, earlier efforts to better manage Optempo have been paying dividends -- at least in terms of reduced average TDY days and better distribution of effort. In September 1998, the Air Force reported that its LD/HD units were at or below steady state deployment limits. Of 22 systems receiving management attention for tempo, only one -- the U2 -- has a service-wide TDY average above 120 days. Most of the others are operating within or below the 100-110 day range, service-wide.105 Of course, there remains considerable potential for a maldistribution of this effort among the major tactical commands, although trend-lines through 1997 showed a narrowing of these differences.


54. U.S. Congressional Budget Office, Paying for Military Readiness and Upkeep: Trends In Operation And Maintenance Spending (Washington, DC: United States CBO, September 1997), pp. 29-31; U.S. Congressional Budget Office, Trends in Selected Indicators of Military Readiness, 1980 Through 1993 (Washington, DC: United States CBO, March 1994), pp. 68-71.

55. This calculation uses the 1989 MCR from the first C.B.O. table. Using the indexed MCR table would produce a similar result.

56. For a brief statistical overview of the condition of the USAF during the "hollow force" years see, Lt. Col. Daniel L. Cuda, "The Hollow Force That Was," Air Force Magazine (April 1994), pp. 69-73.

57. Director of Logistics, Ten Year Lookback: Standards and Performance FY89-FY98 (Langley AFB: ACC November 1998).

58. Air Force section in Senator John McCain, Going Hollow: America's Military Returns to the 1970s - An Update (Washington, DC: Office of Senator John McCain, October 1998), p. 23

59. Richard Betts, Military Readiness: Concepts, Choices, Consequences (Washington, DC: Brookings Institution, 1995), pp. 96-97.

60. Director of Logistics, Ten Year Lookback: Standards and Performance FY89-FY98 (Langley AFB: ACC November 1998).

61. Taking the fact of these different inventories into account would affect a number of USAF policy discussions -- for instance, the issue of air fleet age. While the average age of USAF fighter and attack aircraft in the total inventory (active and reserve) is 13 years, the USAF can today fill its wings with fighters that average 11 years in age -- and somewhat less if A-10s are not counted.

62. It has also become commonplace to measure the change in Optempo in terms of the growth in numbers of named operations. However, this measure is highly misleading and almost entirely lacking in analytical value. It tells nothing about the change in size or duration of missions. Without question, there has been a proliferation of named operations in recent years -- or, more precisely, an increased tendency to parse operations and assign names to all the parts. Thus, since 1992, there have been more than 40 US operations undertaken with regard to the situations in only two places: the former-Yugoslavia and Iraq. The fact that this measure of change can earn anything other than derision is a commentary on the sad state of the current readiness debate. For examples of its use see: Bradley Graham and Eric Pianin, "Military Readiness, Morale Show Strain; Budgets Contract; Deployments Expand," Washington Post, 13 August 1998; Richard Newman, "Can Peacekeepers Make War?," US News & World Report, 19 January 1998, p. 40; Sean D Naylor, "Creeping Hollowness," Army Times, 3 Feb 1997; and James Kitfield, "Fit to Fight," National Journal, 16 March 1996. A list and description of US military operations is available at the website of the Federation of American Scientists, Military Analysis Network:; internet.

63. This is one reason America's globally deployed military cost about 50% more per person than those of our comparably modern allies.

64. These estimates are based on the total incremental costs of foreign stationing, and not just the non-personnel costs. Secretary of Defense, Toward a New Partnership in Responsibility Sharing: An Overview of the 1995 Report on the Allied Contributions to the Common Defense (Washington, DC: Office of the Secretary of Defense, 1995); and, Institute for National Strategic Studies, Strategic Assessment 1996: Instruments of US Power, Strategic Assessment 1996 (Washington, DC: National Defense University, 1996), Ch. 10, section on "Responsibility Sharing."

65. This figure is based on estimated incremental costs of recent operations divided by the average number of persons deployed in those operations. Per person costs show variability by operation and also by year within operations. U.S. General Accounting Office, Bosnia: Cost Estimating Has Improved but Operational Changes Will Affect Current Estimates (Washington, DC: United States GAO, August 1997); U.S. General Accounting Office, Contingency Operations: Update on DOD's Fiscal Year Cost and Funding (Washington, DC: United States GAO, June 1996).

66. The estimate of 280,000 troop/years is based on average monthly deployments during a ten month period, August 1990-May 1991. The average for ten months is then multiplied by 0.833 to get a full-year equivalent. Cost figures are reported in U.S. Department of Defense, Conduct of the Persian Gulf War: Final Report to Congress (Washington, DC: United States DoD, April 1992), Appendix P: Responsibility Sharing, P-2. Of the total incremental cost of the Desert Storm, 51.7 billion was paid into the Operations and Maintenance account during FY1991 and FY1992.

67. Complaints about Optempo focus mostly on how operational demands are poorly integrated with and often disruptive of other, routine activities, including training. Thus, how these demands are managed and distributed seems to be more important than their absolute magnitude. Bruce Callander, "Dissecting the Tempo Problem," Air Force Magazine (April 1998); David Fulghum, "USAF Designs Enticements to Keep Younger Pilots," Aviation Week and Space Technology, 26 May 1997; and Art Pine, "Deployments Take Toll on U.S. Military," Los Angeles Times, 19 March 1995.

68. Alan Vick, et al., Preparing the U.S. Air Force for Military Operations Other Than War.

69. The exceptions were C-130 squadrons in 1991 and fighter squadrons in 1995, which respectively expended about 14% and 13% of their sorties in peace operations during the cited years.

70. Alan Vick, et al., Preparing the U.S. Air Force for Military Operations Other Than War, p. 20.

71. David E. Thaler and Daniel M. Norton, Air Force Operations Overseas in Peacetime (Santa Monica, CA: National Defense Research Institute-RAND, 1998), p. 10.

72. Richard Myers, testimony before the House Subcommittee on Military Readiness Concerning Quarterly Readiness Report to Congress, Washington, DC, 6 March 1998.

73. Major Robert A. Nuanes, AFSAA analyst, quoted in Peter Grier, "Stressed Systems," Air Force Magazine (July 1997).

74. David E. Thaler and Daniel M. Norton, Air Force Operations Overseas in Peacetime, p. 26.

75. Peter Grier, "Stressed Systems," Air Force Magazine (July 1997).

76. Julie Bird, "Washington, We Have a Problem: Doing Too Much With Too Little Is Taking Its Toll," Air Force Times, 24 March 1997; James Kitfield, "Airlift at High Tempo," Air Force Magazine (January 1995); Art Pine, "Deployments Take Toll on U.S. Military," Los Angeles Times, 19 March 1995; James Kitfield, "The Last Super Power," Government Executive (December 1994).

77. Michael C. Ryan, Military Readiness, Operations Tempo and Personnel Tempo: Are U.S. Force Doing Too Much? (Washington, DC: Congressional Research Service Report for Congress, January 1998), pp. 48-51.

78. U.S. Department of Defense, Quarterly Readiness Report to the Congress (Washington, DC: United States DoD, July-September 1998), p. AT-12.

79. Peter Grier, "Stressed Systems," Air Force Magazine (July 1997).

80. Julie Bird, "Washington, We Have a Problem," Air Force Times, 24 March 1997.

81. Peter Grier, "Readiness at the Edge," Air Force Magazine (June 1997); Floyd Spence, Military Readiness 1997: Rhetoric and Reality (Washington, DC: House Committee on National Security, 26 May 1997), pp. 14-15.

82. Air Force section in McCain, Going Hollow - An Update, p. 37

83. By the end of the "hollow force" period of the 1970s, class A mishaps for fighter/attack aircraft rose sharply, peaking in 1978 and 1979 at rates more than twice as high as those of the late-1980s. Lt. Col. Daniel L. Cuda, "The Hollow Force That Was," Air Force Magazine (April 1994), p. 72.

84. Art Pine, "Deployments take toll on U.S. Military," Los Angeles Times, 19 March 1995.

85. Data on the distribution of flying hours and sorties come from the Air Force's Reliability and Maintainability Information System (REMIS).

86. The authors' interpretation of their data is somewhat less sanguine. They observe that, in order to both support contingency operations and match the Cold War training standards, units must fly more today than they did during the Cold War. This, the study poses as a forced choice between less training and more Optempo. But this result is an artifact of the study's method, which counts all flying hours during 1988 and 1989 as training hours, while enforcing a strict separation of "training" and operational flight hours for the current period. It is therefore logically impossible for today's units to both match this standard and fly support for peace operations without exceeding the overall flying time for the baseline years. It would have been more useful to disaggregate the bloc of hours labeled "training." Then, potential trade-offs between a wider variety of activities -- dedicated training, exercises, competitions, peace operations -- could have been explored. What remains interesting, however, is that the data shows that peace operations have not added an enormous Optempo burden to units on average. When serious problems arise, they are due to a maldistribution of Optempo among units, so that some fly much more than average.

87. Alan Vick, et al., Preparing the US Air Force for Military Operations Other Than War, pp. 59-74.

88. Generally speaking, there seems room in the practice of peace operations to scale back on the use of some capabilities more suited to combat engagements with peer or near-peer adversaries. For instance, the Haiti operation should not have involved a flotilla of 70+ navy ships or the use of advanced air defense fighters.

89. Alan Vick, et al., Preparing the US Air Force for Military Operations Other Than War, p. 26.

90. Alan Vick, et al. Preparing the US Air Force for Military Operations Other Than War, pp. 14-15; also, Peter Grier, "Readiness at the Edge," Air Force Magazine (June 1997).

91. Quoted in Frank Wolfe, "Aviation Plan Slashing Wasted Flight Hours, General Says," Defense Daily, 8 December 1998. See also, Lt. Gen. Martin Steele, testimony before the House Subcommittee on Military Readiness, Washington, DC, 18 March 1998, p. 6.

92. United States Air Force, Posture Statement 1999 (Washington, DC: U.S. Department of the Air Force, 1999), p. 5.

93. U.S. Government Accounting Office, Joint Training: Observations on the Chairman, Joint Chiefs of Staff, Exercise Program (Washington, DC: United States GAO, July 1998), pp. 3-4.

94. Gen. Richard Myers, testimony before the House Subcommittee on Military Readiness Concerning Quarterly Readiness Report to Congress, Washington, DC, 6 March 1998, pp. 1-2.

95. The ACC has led the way in trying to monitor and manage these dynamics through its Scheduling Integrated Process Team, which prompted the Air Force to develop better service-wide scheduling procedures. There has been some consistent high-level concerns about this problem outside the air force as well. In March 1997 the head of the Atlantic Fleet, Admiral Paul Reason, argued before the House Readiness Subcommittee that "too many CINCs are placing demands on too few assets," leading to duplicative exercises for environment shaping purposes. Concerning administrative directives, the head of the Atlantic Command, Admiral Harold Gehman, has observed that "we have too many people (in) headquarters who invent things for forces to do." Bruce D. Callander, "Dissecting the Tempo Problem," Air Force Magazine (April 1998); Robert Holzer, "Structure of U.S. High Command Could Damage Force Readiness," Defense News, 17 March 1997, 20; John Omicinski, "Military Preparedness: Top Admiral Sees Too Many Cooks," Army Times, 2 December 1998.

96. The three wings were the 27th Fighter Wing, 314th Airlift Wing, and 5th Bomber Wing of the 8th Air Force. Bruce D. Callander, "Dissecting the Tempo Problem," Air Force Magazine (April 1998); Thomas Fossen, et al., What Helps and What Hurts: How Ten Activities Affect Readiness and Quality of Life at Three 8AF Wings (Santa Monica, CA: National Defense Research Institute-RAND, 1997).

97. Thomas Fossen, et al., What Helps and What Hurts, web summary available at; internet.

98. Thomas Fossen, et al., What Helps and What Hurts (Santa Monica, CA: National Defense Research Institute-RAND, 1997), p. 27.

99. David A. Fulghum, "USAF Designs Enticements to Keep Younger Pilots," Aviation Week and Space Technology, 26 May 1997.

100. Peter Grier, "Readiness at the Edge," Air Force Magazine (June 1997).

101. Reproduced from Ryan, Military Readiness. pp. 48-51. Their original source is a documented briefing by Air Combat Command.

102. Richard Newman, "Can Peacemakers Make War?," US News & World Report, 19 January 1998, pp. 38-44; Katherine McIntire Peters, "The Price of Peace," Government Executive (March 1997); David C. Morrison, "Republicans at War with Peacekeeping," National Journal, 11 March 1995.

103. Bruce D. Callander, "Dissecting the Tempo Problem," Air Force Magazine (April 1998); Gen. Michael E. Ryan, "Ryan Addresses Air Force Challenges," Air Force News, 16 January 1998; John Tirpak, "Working the Optempo Problem," Air Force Magazine (December 1997); John Tirpak, "Airpower in the European Theater," Air Force Magazine (October 1997); Christine Anderson, "ACC Announces Initiatives to Reduce Operational Tempo," Air Force News, 10 September 1997; Maj. Gen. Donald L. Peterson, testimony before the House Subcommittee on Military Readiness, Washington, DC, 11 March 1997; James Kitfield "Airlift at High Tempo," Air Force Magazine (January 1995).

104. United States Air Force, Posture Statement 1999 (Washington, DC: U.S. Department of the Air Force, 1999), pp. 13-14; General Michael E. Ryan, "Expeditionary Aerospace Force: A Better Use of Aerospace Power for the 21st Century," documented briefing, August 1998; David Fulghum, "USAF Embraces New, Fast-Moving Air Units," Aviation Week and Space Technology, 10 August 1998.

105. U.S. Department of Defense, Quarterly Readiness Report to the Congress (Washington, DC: United States DoD, July-September 1998), p. 5 and AT- 12.

[Table of Contents] [Executive Summary] [Sections 1-3] [Conclusion] [Bibliography]

Citation: Carl Conetta and Charles Knight, The Readiness Crisis of the U.S. Air Force: A Review and Diagnosis, Project on Defense Alternatives Briefing Report #10. Cambridge, MA: Commonwealth Institute, April 1999.

cover of report as published in Taiwan

 Cover of this report as published by
 Taiwanese military translation service.

E-mail This Article


powered by FreeFind

US Defense Policy | Regional Security | Terrorism
Iraq & Afghanistan | Military & Strategic Studies
Alternative Security & Defense | Chronological

Buy Publications | Home | What's New | About PDA
Links | Search This Site | E-mail PDA

War Report | RMA Debate Page
Defense Strategy Review Page | Chinese Military Power Page
Terrorism, Counter-Terrorism, Homeland Security | Occupation Distress

Become a PDA Sustainer

Donate Now to Support PDA

The Project on Defense Alternatives, The Commonwealth Institute
P.O.Box 398105, Inman Square Post Office
Cambridge, MA 02139, USA
Phone 617/547-4474, Fax 617/868-1267
Email: pda(at)

Copyright © The Commonwealth Institute. All Rights Reserved.