Document created: 21 April 03
Air University Review, March-April 1976

An Operations Cost Index

Answer to the question "How Much More?"

For Almost every Air Force operation there is some expert--a commander, comptroller, or financial analyst-who is able to answer the question "How much does this operation cost?" The same question put to the overseas analyst logically suggests this follow-up question: "How much more does this cost overseas than it would if it were in the CONUS?" The way to answer this second question is simple. Cost it out for overseas, cost out a similar operation for continental U.S., and subtract the second from the first. Simple, right? Well, as you will see, there is just a little more to it than that.

Right now about half the people who are reading this article are saying: "Just what I would expect from one of those comptroller weenies. He takes a relatively simple problem and makes it so difficult that only his comptroller brethren will understand the answer. I can see the terms coming now: EEIC, PPBS, FYDP, Y = aX + B! (Shudder!)"

On the contrary, dear reader, you will learn no deep dark secrets of the cost trade today. Instead you will be introduced to a problem that requires no more mathematics than arithmetic and little more economics than common sense. All you get to see is how the Alaskan Air Command (AAC) answers the question "How much more?" With any luck, someone will see applications for other overseas operations. With a lot of luck, someone will offer a better solution.

From every mission or operation, some benefit must be derived, e.g., readiness, training, coordination, etc. When an overseas mission or operation is established, some additional benefit should accrue relative to the conduct of the same mission from a CONUS location. For example, stationing fighter interceptors in Alaska rather than Duluth gives us an improved intercept posture that might be measured in crucial minutes. Placing a garrison of soldiers in Berlin also yields additional benefits, albeit not so readily quantified. We will leave this side of the cost/benefit analysis to the think tanks and the whiz kids in the basement of the Pentagon. Let us turn our attention to the cost side.

For each additional benefit of an overseas operation, there is some corresponding incremental cost. If we are to understand this cost, we must first understand the factors causing it. To simplify the discussion, reasons for cost differential are lumped into two categories, physical and man-made.

Every overseas location seems to have some unique climate. Northern Europe has the rain and fog; Southeast Asia, typhoons and monsoons; Alaska, ice and snow. Sidestepping the point that all these "weather pleasures" can be found in the good ole U.S. of A., one can easily see how these climatic features could add to equipment and support costs. The same is true for the various terrains, such as dense jungles or high mountains, and of course the mere fact of being overseas means that logisticians must cope with the problems of geographical separation. Clearly, the physical environmental differences of overseas locations add to costs, but what about man-made reasons?

The varying political systems and economies of some of our overseas locations also add to costs. Foreign nationals employed in positions that might overwise be occupied by U.S. citizens entail an indirect cost. On the other hand, cheaper labor sources might bring about negative incremental costs (bureaucratese for savings). The gold flow problem causes still another incremental cost. In some overseas locations more costly bases might be maintained partially for reasons of local economic impact or diplomatic value. The list of manmade causes for extra costs is at least as long as that of environmental causes.

Of the several methods for dealing with these CONUS-versus-overseas cost differences, three approaches representing various stages of sophistication come to mind:

Cost-be-damned. During World War II the nation was not particularly concerned with the CONUS/overseas cost relationship. The main thing was that the fighting was taking place overseas rather than in the United States itself. Clearly the benefit side dominated the issue.

Modified subjective (skeptic). As time passed, threats seemed to diminish, and taxpayer cost awareness grew. Skeptics began to doubt the worth of the benefits. This marks the beginning of the numbers game insofar as incremental costs are concerned. All of us have heard such official statements as: "Costs of maintaining forces overseas are 'several times' greater than in the CONUS." Perhaps the phrase was the more precise but equally inaccurate "three times greater." Refinement was clearly needed.

Objective approach. Today we are faced with any number of indices that tell us how much more it costs overseas than in the CONUS. For example, the Bureau of Labor Statistics tells us that the Anchorage intermediate family budget is 31% greater than the U.S. urban average. Another set of indices contained in the military construction program tri-service guidances tells us that military construction in Anchorage, Fairbanks, and Point Barrow is 70%, 90%, and 250% greater, respectively, than military construction in Washington, D.C. These indices represent improvements but do not solve our problems so far as operations costs are concerned.

Obviously, we cannot afford the luxury of World War II just to ease the costing problem. Nor can we afford to keep reacting to unsupported contentions of the uninformed. This leaves us two choices: either use some indices that are handy or develop some indices tailored for operations overseas. We might as well choose the latter alternative because sooner or later someone in authority is going to ask us what the cost of raising a family in Anchorage has to do with the fuel costs of F-4s on alert at Galena AFB, Alaska, 200 miles north. As a matter of fact, that is a fairly close description of how the question was actually posed.

An Alaskan Operations Index

The idea of constructing a model portraying the Alaskan cost differential for Air Force operations crystallized in the comptroller shop of the Alaskan Air Command during the late 1960s. The AAC Comptroller and his staff started the ball rolling by posing the following thesis.

Granted, it costs more to conduct Air Force business in Alaska than it might cost in some other location in the South 48, but it does not cost nearly as much as is commonly supposed.

The thesis, through its suggestion of exaggerated costs associated with Alaskan Air Command operations, was posed primarily to point out that the strategic value of Alaska could be negated by cost myths. A study was then started to follow up this thesis.

After the idea was kicked around in-house and at RAND, the following study objectives and guidelines were established.

1. The purpose of the study was to construct a simple model that would enable one to compare operations costs in Alaska to operations costs in the CONUS.

2. Operations costs are defined as those costs incurred in the conduct of ongoing operations from facilities already in place. This includes pay, operations, and maintenance monies and related expenses but excludes investment items such as RDT&E, major procurement, and military construction.

3. The model should differentiate between main base and remote station operations in Alaska.

4.The Alaskan main base situation is a theoretical main base near a large urban center. The remote station is a theoretical typical radar station. These are obviously generalizations based upon composite attributes of the many Alaskan bases and stations.

5. Alaskan main base costs should be compared to similar hypothetical main base situations in Washington, D.C. Remote station costs are relative to a hypothetical station somewhere in Montana.

6. The model should reflect the cost differential in index fashion relative to a base of 100%. For example, an index of 116 for Alaska relates to the CONUS base index of 100, meaning that the Alaskan cost is 16% greater.

7. The model should express some range to account for generalization and make allowances for confidence of results.

The format chosen to present the final indices is represented by the following matrix:

Thus, when the numbers are filled in, one can state generally that a main base in Alaska is expected to cost X% more than its CONUS counterpart. Alternatively, one could say that a particular base in Alaska is expected to cost more than X% (low estimate) but less than Y% (high estimate).

With the index goals established, the analysts then turned to the actual work involved in calculating the indices required. This task was accomplished in three stages. First, a list was composed of those factors expected to influence Air Force operations in Alaska. Second, the specific costs affected by these influences were identified. Last, the actual indices--low, high, and best--were computed for each cost category defined.

Alaskan Cost Influences. The list of cost-influencing factors is seemingly endless, but for costing purposes the relevant reasons were reduced to the following list: climate and terrain, distance from major supply centers, availability of surface transportation, limited labor market, and competition for resources. A review of these forces highlights their impact on Air Force operations in Alaska.

Climate and Terrain. Although winter in the Anchorage area and along the Aleutian chain may be no more severe than in some of the northern CONUS locations, it is certainly something to be reckoned with, and the interior and northern Alaskan bases and stations are subjected to some of the harshest conditions in the world. To combat the effects of the long, cold, and windy winters, special measures are required for personnel, equipment, and buildings alike. In addition to being costly, the special equipment is subject to unusual wear and tear.

Distance from Major Supply Centers. The distance from the major CONUS supply centers affects cost in three ways. First, there is the cost for the extra distance over which the goods must be transported. Second, because Alaska is separated from the CONUS by Canada or a large stretch of water, there are some peculiar shipping tariffs applicable to Alaska. For example, it costs more to ship goods to Seattle if Alaska is the final destination than it does to Seattle if Seattle is the final destination. Third, because the time between deliveries is significantly longer, larger inventories must be kept on hand. This point is particularly noticeable at the remote stations.

Insufficient Surface Transportation. Because there are only 4000 miles of paved road, 3000 miles of secondary road, and 600 miles of railroad, it is frequently necessary to ship by air. Two of the Alaskan Air Command remote stations, Sparrevohn and Indian Mountain, as well as several communications sites, rely 100 percent on air delivery.

Limited Labor Market. The population of Alaska is only slightly more than 350,000. There are distinct limits to both the quantity of labor and the expertise available to employers. Recruitment from the Lower 48 is commonplace, and the employer (the Air Force) must bear the extra recruitment cost.

Competition for Resources. At one time the numerous federal agencies dominating the Alaskan economy were actually driving up the prices through interagency competition for resources. Because of new purchasing techniques and better cooperation, this problem has largely disappeared. However, another agency is now bidding up the prices. The oil and gas industry in Alaska is garnering Alaskan and West Coast resources that might otherwise be available to the military in Alaska.

Cost Categories. Obviously the total cost of conducting Air Force business in Alaska is comprised of cost in several categories, e.g., costs for pay, costs for flying, costs for supplies, etc. The analysts were faced with the decision as to which costs were pertinent to the model and which costs would give the best overall picture. Experts representing almost every major functional portion of operations in AAC were queried as to what costs in their functions were affected most by Alaskan influences. The answers ranged from the usual, such as more fuels for heat and power, to the unexpected, such as the costs associated with power poles slipping out of sight into the tundra. One person even brought up the cost associated with an airman's lessened morale because he is stationed at Shemya rather than Las Vegas.

To reduce the list to a manageable number, three selection criteria were used:(1) the degree to which Alaskan influences could be applied, (2) the total dollars involved, and (3) susceptibility to being measured. The following cost categories were thus selected: military personnel compensation, Civil Service personnel compensation, military family housing, facility maintenance and repair, supplies other than fuels, fuels for heat and power, purchased services, transportation of things, aircraft operations, operation and maintenance of vehicles, travel and per diem, equipment, and communications.

Indexing. For each cost category listed above, three indices were computed: low, high, and best. The low represents the best possible situation (all economies taken advantage of); the high represents the highest cost case. The best estimate lies somewhere in between, sometimes the average, more frequently the situation as actually perceived. The indices are each relative to CONUS counterparts as explained earlier. The method of computation differs according to the circumstances surrounding each category. Military pay, for example, presents little problem. The low estimate assumes that some Alaskan troops receive exactly the same pay as their CONUS counterparts. The high estimate represents the case in which the Alaskan blue suiter receives special cost of living, housing allowance, or family separation allowance not paid in CONUS locations. The best estimate was the actual case somewhere in between the two extremes. The indices are computed by dividing AAC pay averages for each of the three conditions by average CONUS pay. Detailed financial statements and population figures allowed such computations to a reasonably accurate degree. "Supplies" represents a slightly more difficult category. Here one must consider the quantity variance (more volume) as well as the price variance (per unit cost). Furthermore, there are thousands of different types of supply items. A third example, perhaps the most difficult case, is the area of communications. Since there is no CONUS counterpart of the government-owned/contractor-operated military communication system in Alaska, our analysts had to be very ingenious. Being true economists, they simply assumed such systems existed in the CONUS and then costed out the assumptions. By now it is obvious that the final index results could not be totally objective. Rather, they must be a mixture of expert opinion and objective values, summarized in an objective, quantitative manner.

Weighting. All of the indices were combined into one total for each estimate (low, high, and best) based on a weighted average. Each separate index was weighted in proportion to its contribution to the total budget. Each contribution was made as "net" as possible, by considering each separate from the influences of the others. Supplies, for example, were considered deleting the civilian pay and military pay involved in handling them.

After applying the indexing and weighting to the cost categories, the following results were achieved:

Main Base O & M Cost Index

 Remote Station O&M Cost Index

       When the appropriate weighted averages are computed, the final matrix is then complete.

From these results one can say that the average main base and remote station operations in Alaska cost, respectively, 24% and 31% more than the CONUS counterparts. Reapplication of the methodology two years later yielded respective indices of 125 and 123.

Postmortem #1. The overall results achieved from the first try at an operations index seemed satisfactory. There are many other indicators that suggest a 20 to 30% cost differential for Alaskan business, and the analysts felt their results were realistic. The methodology, however, caused two problems, one procedural and one practical.

Redundancy. It was impossible to eliminate all redundancy in the cost categories. For example, not much is left over for a vehicle operations and maintenance index when pay, supplies, and equipment are netted out. The same is true for aircraft operations, this problem being the result of the mixture of missions and budgets as cost categories--apples and oranges as it were.

Applicability. The greatest difficulty in "selling" the index to other commands and agencies also stems from particular cost categories selected. These categories do not lend themselves to operations of other services. Admiral Noel Gayler, CINCPAC, after receiving a briefing on the Alaskan indices, desired that his own staff pursue such a model for Pacific forces. In subsequent dialogue between AAC and PACOM, interpretation of the term "communications" was a symptom of the problem. The Alaskan interpretation of "communications" was as a cost category; the Pacific analysts viewed it as a mission comprised of several other categories. Matters were not helped by the fact that the communications area was and still is the toughest area to define. In another attempt to apply the methodology to larger operations, an Alaskan Command (ALCOM index was attempted, based on a combination of Air Force in Alaska (primarily AAC) and Army in Alaska (USARAL) indices. The USARAL analysts first objected to the basis of measurement of "mission." Whereas AAC had specified no particular mission other than the one in being, USARAL desired to address specific missions such as artillery and field forces, specifying all indices on a man-year supported basis. Obviously, USARAL had some problems coping with our arrangement of cost categories. The problems were not solved, and a satisfactory joint index was never obtained.

In 1975 a revised model was constructed. The basic approach previously described was retained, but one more criterion was added--applicability. The 1975 version was to be one that could be used by other commands, particularly the joint commands. Instead of having the mixture of cost categories and functional categories used in the first try, the 1975 version develops indices generally along traditional DOD budget lines for the following operations cost elements: Personnel Expenses including military personnel, military permanent change of station, Classification Act employee compensation, Wage Board employee compensation, and military family housing; and Operations and Maintenance Expenses including travel and per diem, transportation of things, utilities, rents, leases, communications, purchased and other services, aviation POL, aircraft depot maintenance and spare parts, supplies, and equipment and related maintenance.

Hopefully the use of DOD-wide cost categories, rather than one command's interpretation of costs, will enhance the usability of the indices. As a side benefit, this array also highlights the dominance of personnel-related costs or operations costs.

Results. New indices were developed for each category represented, and they were weighted in the fashion previously described, using actual AAC fiscal year 1974 costs. This led to the following results:

Main Base Estimates

 Remote Station Estimates

The final cost matrix for the 1975 model is as follows:

Postmortem #2. Having constructed a better model, we were now faced with another challenge. It had taken us two years to convince our boss that a 25 percent differential was just about right for main bases. Now that he finally believed it, we had to tell him about the new numbers we came up with in the 1975 model.

Actually the differences between the first try and the 1975 model can be attributed to three factors. First, we are getting smarter. Many problems brought out with earlier versions have been solved, and most estimates are getting better and better. Second, costs change over time, and exactly duplicate results would not be expected. Third, it is possible that we have not hit upon the best methodology for treating some of the various cost categories. This applies to both Alaska and CONUS counterpart categories. One big influence on the 1975 model, for example, is PCS costs. Lumped together with pay in the earlier models, the PCS factor was probably understated in the first version. It is possible that the reverse is true now. Right now AAC Management Analysis studies are under way in several areas of the index. These studies should lead to an even clearer picture.

In summary, we have constructed a simple index model that tells planners how much more Air Force operations in Alaska cost than similar operations in the CONUS. Planners can use this model to judge the relative value of the benefits gained. Furthermore, a general methodology has been established that should allow analysts to develop similar indices for other overseas areas.

Perhaps the biggest change in the index over the years since its origin has been the evolution of purpose. At first the index was perceived as an educational effort designed to reorient collective thinking about costs of conducting Air Force business in Alaska. While there still exists plenty of naïveté about the high costs in Alaska, this reeducation effort has been successful at many levels. However, even after the record on Air Force costs in Alaska is set straight, the need for such an index continues. The purpose of the index now is to provide a reliable management tool that can be used in making broad decisions regarding the placement and operations of forces in Alaska.

The basic concepts that justify pursuit of an Alaskan Air Command Operations Cost Index are as real today as they were six or seven years ago. There will always be a need to educate people—particularly those in high--level decision-making positions--on the costs of operations in Alaska. Likewise there is a continuing need to provide the objective economic and cost information necessary to judge the merits of maintaining operations in that state. Future efforts on Alaskan operations indices are planned to incorporate regional differences and discriminate between missions.

Because the basic concepts and the methodology should apply to other commands as well, there is no reason that similar indices could not be computed for other U.S. overseas operations. As mentioned earlier, some work in this area has already been accomplished in the joint Pacific Command. Also, comptroller grapevines tell us similar efforts are under way in the European Command. This high-level interest will surely spur some interest in operations cost indices, but high-level interest need not be the only driving force.

Although the operations index has been primarily a comptroller-related project, the technical expertise required in costing out the indices described in this article is general enough to invite studies from persons outside the comptroller field. New ideas are needed more than financial expertise. The subject is particularly adaptable to thesis-level studies by AFIT students or bootstrappers or to projects by professional education study groups. A fresh look at the problem by persons not immediately connected with it could lead to improvements.

Captain Willard L. Mason (M.B.A., Florida State University) recently completed a tour as Chief, Cost Analysis Division, DCS/Comptroller, Hq Alaskan Air Command. A former technical sergeant, he was commissioned through the Airman Education and Commissioning Program. Captain Mason is a graduate of Squadron Officer School and has completed Air Command and Staff College and Industrial College of the Armed Forces.

Air & Space Power Home Page | Feedback? Email the Editor