Defense Inventory: Approach for Deciding Whether to Retain or	 
Dispose of Items Needs Improvement (25-MAY-01, GAO-01-475).	 
								 
As of September 1999, the Department of Defense (DOD) reported	 
that it owned secondary inventory worth about $64 billion and of 
that about $9.4 billion of that inventory is more economical to  
retain than to dispose of and possibly repurchase later. This	 
report focuses on whether DOD's economic retention decisions are 
sound. GAO found that military components (other than the Air	 
Force) have developed models designed to make economic retention 
decisions on secondary inventory. However, none of the components
now use their economic retention models. Instead, they and the	 
Air Force use ceilings to limit the amount of economic retention 
inventory they hold. Components have not properly documented	 
their approaches to economic retention decisions. For example,	 
common model factors vary and assumptions lack consistency and	 
are not current. In addition, DOD lacked sound analytical support
for the maximum levels it now uses. As a result, the components  
cannot depend on their models or ceilings to determine retention 
inventory levels without review and improvement. They also have  
not reviewed their approaches annually . Because the ceilings	 
lack analytical support and the model factors and assumptions	 
vary without explanation and are out of date, the Department does
not have a sound basis for its approach to manage items held in  
economic retention status. Consequently, the Department cannot	 
guarantee that inventories held in economic retention are the	 
right amount.							 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-01-475 					        
    ACCNO:   A00844						        
  TITLE:     Defense Inventory: Approach for Deciding Whether to      
             Retain or Dispose of Items Needs Improvement                     
     DATE:   05/25/2001 
  SUBJECT:   Cost effectiveness analysis			 
	     Defense economic analysis				 
	     Internal controls					 
	     Inventory control systems				 
	     Property disposal					 
	     Spare parts					 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-01-475
     
Report to the Ranking Member, Committee on Armed Services, U. S. Senate

United States General Accounting Office

GAO

May 2001 DEFENSE INVENTORY Approach for Deciding Whether to Retain or
Dispose of Items Needs Improvement

GAO- 01- 475

Page i GAO- 01- 475 Defense Inventory Letter 1

Appendix I Changes in Secondary and Economic Retention Inventory Levels 13

Appendix II Information on Component Economic Retention Models 20

Appendix III Scope and Methodology 27

Appendix IV Comments From the Department of Defense 28

Tables

Table 1: Component Ceilings on Economic Retention Inventories 8 Table 2:
Retention and Disposal Cost Factors in Components?

Economic Retention Models 24 Table 3: Application of Economic Retention by
Type of Item, by

Component, as of September 30, 1999 (number and dollar value) 25

Figures

Figure 1: Department of Defense and Components? Economic Retention
Inventories, September 30, 1991, and September 30, 1999 5 Figure 2: How
Ceilings Reduce an Economic Retention Level 9 Figure 3: Changes in
Department of Defense and Components?

Secondary Inventory, September 30, 1991, and September 30, 1999 (billions of
dollars) 13 Figure 4: Components? Shares of Department of Defense

Secondary Inventory, September 30, 1999 (percent) 15 Figure 5: Share of
Secondary Inventory Held in Economic

Retention Status, by Component and Department- wide, September 30, 1991, and
September 30, 1999 16 Contents

Page ii GAO- 01- 475 Defense Inventory

Figure 6: Components? Shares of Department of Defense Economic Retention
Inventory, September 30, 1999 (percent) 17 Figure 7: Components? Shares of
Department of Defense

Consumable Economic Retention Inventory, September 30, 1999 (percent) 18
Figure 8: Components? Shares of Department of Defense Reparable

Economic Retention Inventory, September 30, 1999 (percent) 19

Abbreviations

DLA Defense Logistics Agency DOD Department of Defense

Page 1 GAO- 01- 475 Defense Inventory

May 25, 2001 The Honorable Carl Levin Ranking Member Committee on Armed
Services United States Senate

Dear Senator Levin: As of September 30, 1999, the Department of Defense
reported 1 that it owned about $64 billion of secondary inventory 2 and
that, of that amount, components have determined that about $9. 4 billion is
more economical to retain than to dispose of and possibly need to
repurchase. We previously reported that $39. 4 billion (about 60 percent) of
the Department?s $65. 8 billion secondary inventory as of September 30,
1997, exceeded operating requirements and that the Department?s cost of
operations could be reduced by disposing of inventories if it was economical
to do so. 3 Through disposal, the Department eliminates some storage costs
and the government recoups a small fraction of an item?s purchase price.
However, if stock is disposed of before it is economical to do so,
inadequate supplies of the item could reduce readiness and costly
repurchases might be necessary.

The Department divides secondary inventory into two broad categories.
Supplies that are expected to be used within 2 years or that meet war
reserve requirements are categorized as active inventory; 4 amounts beyond
that level are placed in inactive status. Components may retain inactive
inventory if it is economical to do so; inventory retained for this reason
is

1 The analysis in this report is based on information provided from
Department of Defense reporting systems, which have significant
deficiencies. See our report Major Management Challenges and Program Risks:
Department of Defense (GAO- 01- 244, Jan. 2001). However,

this information is the best available on Department inventory levels and is
therefore used in this report. 2 Secondary inventory includes spare and
repair parts and other items that support the

Department?s operating forces. 3 Defense Inventory: Status of Inventory and
Purchases and Their Relationship to Current Needs (GAO/ NSIAD- 99- 60, Apr.
1999).

4 Department of Defense Materiel Management Regulation 4140. 1- R, appendix
16, Office of the Under Secretary of Defense for Acquisition and Technology,
May 1998.

United States General Accounting Office Washington, DC 20548

Page 2 GAO- 01- 475 Defense Inventory

referred to as economic retention inventory. Department regulations state
that the economic retention level for an item should be based on an analysis
that balances the costs of retention and disposal. 5 In doing the analysis,
the Department?s regulations state that components 6 should consider factors
such as storage costs, potential long- term demand, potential repurchase
costs, and the expected life of the weapon system the item is to support. 7
As agreed with your office, we determined whether the Department components?
approaches for making economic retention decisions are sound.

The Department?s components do not have sound analytical support for
determining when it is economical to retain or dispose of the $9. 4 billion
in inventory the Department is holding for economic reasons. The components?
decision- making approaches for retaining economic retention inventory have
evolved from the use of economic models to the use of judgmentally
determined levels. In the mid- 1990s, to meet Department inventory reduction
goals, each component lowered the maximum levels set for holding economic
retention inventory. The factors that support these models and the current
maximum levels being used by the components have various weaknesses. For
example, there are variations in common model factors without explanation,
and the assumptions used lack consistency with governmentwide and
Departmentwide guidance. In addition, the Department did not have sound
analytical support for the maximum levels they selected. Also, although the
Department requires annual reviews of the analyses supporting economic
retention decisions, the components have generally not done such reviews. As
a result of these weaknesses, the Department is vulnerable to retaining some
items when it is uneconomical to do so and disposing of others when it is
economical to retain them.

We are recommending that the Secretary of Defense direct the Under Secretary
of Defense for Acquisition, Technology, and Logistics, in

5 Department of Defense Materiel Management Regulation 4140.1- R, section
C4. 2. 1. 2. 1, Office of the Under Secretary of Defense for Acquisition and
Technology, May 1998. 6 The Air Force, the Army, the Navy, and the Defense
Logistics Agency manage almost all (over 99 percent) of the Department?s
secondary and economic retention inventories. Since the Marine Corps manages
such a small portion of inventory, it was not included in this review. 7 See
footnote 5 above. Results in Brief

Page 3 GAO- 01- 475 Defense Inventory

consultation with the Secretaries of the Army, Navy, and Air Force and the
Director, Defense Logistics Agency, to develop and implement sound
approaches for making economic retention decisions. We also recommend that
the components annually review their approaches as required by Department
regulations.

The Army, Navy, and Defense Logistics Agency approach to complying with
Department of Defense requirements was to develop economic models to
determine the maximum amount of inactive inventory they could retain. The
Air Force did not develop an economic retention model; rather it employed a
model based on historical usage patterns. Most recently, all the components
lowered their maximum levels (referred to as ceilings) for items in economic
retention during the 1990s to help them meet inventory reduction targets.

The Department requires that an economic model calculating whether to retain
an inventory item or dispose of it should compare the cost of retention with
the cost of disposal and select the option with the least cost. 8 As the
amount of inactive stock increases, the cost of retention increases (more
items cost more to hold) and the cost of disposal decreases (with greater
amounts of an item on hand, the likelihood of having to repurchase it
becomes less). Equilibrium is reached when the additional cost of retention
equals that of disposal. This equilibrium level of inventory is the economic
retention level- the largest retention amount of an inactive item that can
be justified by economic analysis. Any amount of inventory over this level
would become eligible for retention on a contingency basis or disposal.

Management of the Department?s secondary inventory is a complex process, and
effectively implementing systemwide improved management approaches has been
a long- standing challenge for the Department. 9 However, following the end
of the Cold War, the Department recognized that it had unnecessarily high
inventories as a result of major reductions in

8 See footnote 5 above. 9 In 1990, we identified the Department?s management
of secondary inventories as a highrisk area because inventory levels were
too high and management systems and procedures were ineffective. Some
improvements have been made, but these conditions still exist and this area
remains on our high- risk list. See Major Management Challenges and Program
Risks, Department of Defense (GAO- 01- 244, Jan. 2001). Background

Page 4 GAO- 01- 475 Defense Inventory

its force structure, and it directed the components to take action to lower
inventory levels, including economic retention inventory. In response to
this directive, during 1994, the Army, Navy, and Defense Logistics Agency
chose to lower economic retention inventory levels by placing a preset
maximum retention level that generally fell below the levels calculated by
the models. The Air Force lowered its maximum level in 1996.

The Department reported its economic retention inventory fell from about
$13.8 billion to about $9. 4 billion (about 32 percent) between fiscal year
1991 and 1999 (see fig. 1) and by 40 percent when adjusted for inflation.
The latter part of this period covers the years when lower ceilings were put
in place by the services and Defense Logistics Agency. Although all three
services reported reductions in economic retention inventory levels during
this time, the changes were uneven. The Air Force reported the smallest
decrease in economic retention inventory- from about $5. 1 billion to about
$4. 5 billion (about 12 percent). In contrast, the Army reported a decrease
from about $1. 3 billion to about $600 million (about 54 percent) and the
Navy from about $5. 6 billion to about $1. 6 billion (over 71 percent). On
the other hand, the Defense Logistics Agency reported an increase in
economic retention inventory from about $1. 8 billion to about $2. 7 billion
(about 50 percent) as a result of a decision to consolidate management
responsibility for all consumable items within the Agency. According to
Department data, if it had not required the services to transfer management
of large quantities of inventory to the Defense Logistics Agency during the
1990s, the Defense Logistics Agency inventory would have decreased by over a
billion dollars. (See app. I for more details on how the composition of
secondary and retention inventory changed between fiscal year 1991 and
1999.)

Page 5 GAO- 01- 475 Defense Inventory

Figure 1: Department of Defense and Components? Economic Retention
Inventories, September 30, 1991, and September 30, 1999

Source: Our analysis of the Department of Defense Supply System Inventory
Report?s data for fiscal years 1991 and 1999.

It is also important to note that, in response to a congressional mandate,
the Department is conducting an independent study of secondary inventory and
parts shortages as required by section 362 of the National Defense
Authorization Act for Fiscal Year 2000. 10 As described in the authorization
act, the independent study is to include analyses of the appropriate levels
and use of secondary inventories, alternative methods for disposing of
excess inventory, and the application of private sector cost calculation
models in determining the cost of secondary inventory storage. According to
the Department, the study is scheduled for completion by the end of August
2001.

The services and Defense Logistics Agency do not have sound analytic support
for their approaches for determining whether they are holding the correct
amount of items in economic retention inventory. While components (with the
exception of the Air Force) developed individual economic models designed to
place inactive inventory in economic retention status as early as 1969, they
have not used them since 1994. Instead, components lowered maximum levels of
inventory that could be held (referred to as ceilings) to make economic
retention determinations that would help achieve agency inventory reduction
goals. Agency

10 Public Law 106- 65. Economic Retention

Inventory Decisions Lack Sound Analytical Support

Page 6 GAO- 01- 475 Defense Inventory

information indicates that this approach has helped to reduce inventory
levels. However, the components have not annually reviewed the analyses used
to support their economic retention decisions, as required by the
Department, and therefore have no assurance that the inventories held in
economic retention status are appropriate.

Although the components were not using their economic retention models to
manage inventory levels, we did generally review the models. We noted that
factors and assumptions within the models differed and were not current
without explanation for the differences. Given the differences we found in
these models, such as varied and outdated cost factors and assumptions and
the lack of support for these factors and assumptions, it is uncertain
whether they determine an accurate retention level.

A methodology for determining how many items are to be kept in economic
retention status, which the Department of Defense requires, should compare
the costs of retention to the costs of disposal of an inventory item. 11 The
Army, Navy, and Defense Logistics Agency developed models designed for
making economic decisions that consider the costs of retention and the costs
of disposal. The Air Force does not compare retention and disposal costs in
determining economic retention inventories. Instead, the Air Force employs
historical usage levels to determine economic retention levels.

The components developed their models in different ways and use different
factors and assumptions in their models without detailed documentation. The
amount of inventory to hold in economic retention varies by model depending
on the factors and assumptions in the models. The economic retention models
of the three components generally meet the Department?s requirements to
compare retention costs to disposal costs, but the factors and assumptions
in them vary across the components. For example, the Army and Navy use a
factor of obsolescence and the Defense Logistics Agency does not. In
addition, the values for similar factors used in economic retention models
varied among components. For example, the Army?s value for loss rates (loss
through theft or decay) is 1 percent, the Navy?s value is 4 percent, and the
Defense Logistics Agency does not use a loss rate.

11 See footnote 5 above. Component Retention

Model Cost Factors and Assumptions Are Different and Not Current

Page 7 GAO- 01- 475 Defense Inventory

Furthermore, the components have not appropriately updated their model
assumptions. For example, prescribed discount rates, a key assumption in all
economic retention models, vary across components? models. The Navy uses a
10- percent discount rate and the Army and the Defense Logistics Agency use
a 7- percent rate for computing net present values. 12 Neither value matches
discount rates recommended by the Office of Management and Budget for a
cost- effectiveness analysis, which is the decision to retain or dispose of
inventory. 13 The rates are also inconsistent with our guidance on discount
rates. 14 For example, for the year 2000, the Office of Management and
Budget discount rate for a cost- effectiveness analysis was 4.2 percent for
a 30- year analysis; the rate was 2.9 percent in 1999.

Further information about component economic models can be found in appendix
II of this report.

The use of a maximum level to manage economic retention stocks (commonly
called ceilings by components) makes the Department vulnerable to retaining
items when it is uneconomical to retain them or disposing of items that are
economical to keep. The components judgmentally developed their ceilings for
economic retention inventory, which differ and have yielded lower levels of
economic retention inventory than the levels calculated by the economic
retention models.

During 1994- 96, the components established different ceilings for items in
economic retention. A ceiling imposes an upper constraint of years of
demand- the quantity needed on an annual basis to meet requirements- on how
much inactive inventory can be retained. Prior to the 1990s, the components
had set ceilings on retention inventory that varied but that generally
exceeded higher levels determined by their economic retention

12 A discount rate of 10 percent implies that 91 cents would grow at a 10-
percent interest rate to $1 in a year ($ 0. 91 times 1.1 = $1). Thus, the
value today (the present value of $1 received one year from now) is 91
cents. The calculation of the present value of $1 received 2 years from now
would be 1 divided by 1.1 squared, or 1 divided by 1. 21, which equals 82. 6
cents. 13 Office Of Management and Budget Circular A- 94, 2000 Discount
Rates for the Office of Management and Budget Circular No. A- 94, appendix
C, February 9, 2000.

14 Discount Rate Policy (GAO/ OCE- 17. 1. 1, May 1991, pp. 25- 26).
Comparing the estimated costs of retention to the costs of disposal in
deciding whether to keep or dispose inventory is a cost- effectiveness
analysis. The discount rates that are to be used in such analyses are
updated every year and are published in appendix C of the Office of
Management and Budget Circular A- 94. Components? Ceilings Are

Different and Judgmentally Based

Page 8 GAO- 01- 475 Defense Inventory

models. While component ceilings varied in the span of years of demand, they
also varied in the total years of inventory covered. The Army ceiling is
applied to inventories above active inventory requirements. 15 The Air
Force, Navy, and Defense Logistics Agency ceilings apply to their entire
inventory requirements, including active inventory. Table 1 summarizes the
maximum levels used by each component during the 1990s.

Table 1: Component Ceilings on Economic Retention Inventories Component
Ceilings for inventory to be held

Air Force 13 years of total maximum demand for all items Army 7 years of
demand above requirements for serviceable reparables, 6

years for unserviceable reparables 5 years of demand above requirements for
all other items (as of Dec. 1999, levels left up to management) Navy 12
years of total demand for new weapons systems

8 years of total demand for ?steady? weapon systems 4 years of total demand
for weapons systems approaching obsolescence Defense Logistics Agency 6
years of total demand

Source: Our analysis of component information.

According to component officials the components now hold fewer items in
economic retention status because the ceilings established in the 1990s
replaced the levels calculated by economic retention models. The components?
more stringent ceilings result in smaller inventory level determinations
than would be calculated with economic retention models. Figure 2
illustrates the level a model might calculate and how a more constrained
ceiling would override it. For example, a component?s ceiling for economic
retention stock is 6 years of demand above requirements. If the model
computed an economic retention limit (e. g., 8 years of demand of 25 items a
year- 200 items) that exceeded the maximum level (e. g., 6 years of demand-
150 items), the ceiling (150 items) would be selected as the retention
level. The additional stock (50 items) would be moved to other inactive
categories (contingency or reutilization status) or be disposed of.

15 As defined in the Supply System Inventory Report (September 30, 1999, p.
25), active inventory requirements are materiel expected to be used within
the budget year and materiel that has been purchased to meet specific war
reserve requirements.

Page 9 GAO- 01- 475 Defense Inventory

Figure 2: How Ceilings Reduce an Economic Retention Level

Source: Our analysis.

Components have not reviewed their economic retention models or the
judgmentally established ceilings annually, although required to do so by
the Department. 16 The lack of the components? reviews of their retention
analyses, either their models or ceilings, raises further questions about
the cost- effectiveness of either approach.

The Department?s required annual reviews are to focus on

 improving analyses supporting retention decisions by accounting for
potential upward or downward trends in demand and/ or the uncertainties of
predicting future long- term demand based on historical data and

 improved estimates of costs used in retention decision- making. 17 All
components have conducted studies of their economic retention analyses, and
initiatives were undertaken to meet inventory reduction goals during the
1990s, such as constraining economic retention model

16 Department of Defense Materiel Management Regulation 4140.1- R, section
C4. 2. 1. 1. 4, Office of the Under Secretary of Defense for Acquisition and
Technology, May 1998. 17 Department of Defense Materiel Management
Regulation 4140.1- R, section C4. 2. 1. 2. 3.,

Office of the Under Secretary of Defense for Acquisition and Technology, May
1998. Limited Reviews of

Approaches for Making Economic Retention Decisions

Page 10 GAO- 01- 475 Defense Inventory

determinations with ceilings. However, no studies had been conducted to
determine if economic retention models could be used to establish
appropriate Department inventory levels on an economic basis, rather than
through the use of ceilings. There was little documentation available
supporting the selection of the factors and assumptions used in economic
retention models, such as obsolescence rates and discount rates. The various
factors and assumptions might be appropriate, but in most cases the
components lacked documentation describing why they were selected for use.

Furthermore, the limited information that is available about the impact of
ceilings indicates that they could be causing uneconomical disposals. For
example, the Army Audit Agency produced a March 2000 study that suggested
that, while the Agency found no instances in which an item was disposed of
when it was more economical to retain it, it concluded, based on the
statements of inventory managers, that maximum levels resulted in the
disposal of items that were still economical to retain. 18 Inventory
managers also told us that maximum levels also caused disposal of items that
were still economical to retain, but components were unable to provide data
about repurchases of disposed items because of limitations in component
databases.

Components (other than the Air Force) have developed models designed to make
economic retention decisions. However, none of the components currently use
their economic retention models. Instead, they and the Air Force use
ceilings to limit the amount of economic retention inventory they hold.
Components have not properly documented their approaches to economic
retention decisions. For example, there are variations in common model
factors and assumptions lack consistency and are not current. In addition,
the Department did not have sound analytical support for the maximum levels
they currently use. As a result, the components cannot currently depend on
their models or ceilings to determine retention inventory levels without
review and improvement. They also have not annually reviewed their
approaches. However, the Department is currently conducting a mandated study
of secondary inventory and spare parts shortages. Because the ceilings lack
analytical support, and the model factors and assumptions vary without
explanation and are out of date, the

18 Audit of Repair Parts Disposal Policy and Procedures, U. S. Army Audit
Agency (AA 00203, Mar. 13, 2000). Conclusions

Page 11 GAO- 01- 475 Defense Inventory

Department cannot provide reasonable assurance that inventories held in
economic retention are the right amount.

We recommend that the Secretary of Defense direct the Secretaries of the
Army, Navy, and Air Force and the Director of the Defense Logistics Agency,
in consultation with the Under Secretary of Defense for Acquisition,
Technology, and Logistics, to take the following actions:

 Taking into consideration the results of the congressionally mandated
study, establish milestones for reviewing current and recently used
approaches for making decisions on whether to hold or dispose of economic
retention inventory to identify actions needed to develop and implement
appropriate approaches to economic retention decisions.

 Annually review their approaches to meet Department regulations to ensure
that they have sound support for determining economic retention inventory
levels.

In written comments on a draft of this report, the Department of Defense
partially concurred with our recommendations. The Department agreed with our
recommendation that its components needed to annually review the
appropriateness of their economic retention inventory levels.

Regarding our draft recommendation that the components review their
approaches to determining economic retention levels, the Department stated
that the need for components? further review of retention decisions would be
determined after the completion of an independent study in August 2001 of
secondary inventory and parts shortages required by section 362 of the
National Defense Authorization Act for Fiscal Year 2000. The results of the
study could affect component approaches to making economic retention
decisions. The study is to report on such issues as the appropriate levels
of secondary inventories, alternative methods for disposing of excess
inventory, and the application of private sector cost calculation models in
determining the cost of secondary inventory storage. Our recommendation
focuses on reviewing the approaches for setting economic retention levels to
minimize the possibility of inappropriate retention or disposal decisions.
How the study results will affect how the Department should address our
recommendation remains to be seen. Therefore, we modified our draft
recommendation. We are now recommending that the Department establish
milestones for taking action on the study?s recommendations as they relate
to the economic retention issues that we raised in this report. The
Department?s comments are reprinted in appendix IV. Recommendations for

Executive Action Agency Comments and Our Evaluation

Page 12 GAO- 01- 475 Defense Inventory

Appendix III describes our objectives, scope, and methodology for this
report.

We are sending copies of this report to the appropriate congressional
committees; to the Honorable Donald H. Rumsfeld, Secretary of Defense; the
Honorable Joseph W. Westphal, Acting Secretary of the Army; the Honorable
Robert B. Pirie, Jr., Acting Secretary of the Navy; the Honorable Lawrence
J. Delaney, Acting Secretary of the Air Force; Lieutenant General Henry T.
Glisson, Director, Defense Logistics Agency; and the Honorable Mitchell E.
Daniels, Jr., Director, Office of Management and Budget.

Please contact me at (202) 512- 8412 if you have any questions. Key
contributors to this report were Charles Patton, Donald Snyder, Scott
Pettis, and Charles Perdue.

Sincerely yours, David R. Warren, Director Defense Capabilities and
Management

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Page 13 GAO- 01- 475 Defense Inventory

After the end of the Cold War and a subsequent reduction in force structure,
the Department of Defense recognized that it had high inventory levels and
took action to reduce them. Department secondary inventory levels were
reduced by about a third between 1991 and 1999 when adjusted for inflation.
All four components reviewed reduced their secondary inventory levels during
this time, although the level of reduction varied by component. The
percentage of secondary inventory held in economic retention status also was
reduced during this time, although there were fluctuations in inventory
levels among components. There were also sizable shifts in the amount of
consumable and reparable inventories managed by each component. The
following sections provide details of Department and component secondary and
economic retention inventory trends.

The Department reported reductions in its secondary inventory during the
1990s. The amount of secondary inventory fell from $88 billion in 1991 to
$64 billion in 1999 (a decline of 27 percent and a 36- percent reduction
when adjusted for inflation). As shown in figure 3, component performance in
reducing inventory levels was uneven but generally reflected the Department-
wide performance.

Figure 3: Changes in Department of Defense and Components? Secondary
Inventory, September 30, 1991, and September 30, 1999

Legend DLA = Defense Logistics Agency DOD = Department of Defense

Source: Our analysis of the Department Supply System Inventory Report?s data
for fiscal years 1991 and 1999.

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Secondary Inventory Levels Have Decreased

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Page 14 GAO- 01- 475 Defense Inventory

All four components reduced their secondary inventory levels during this
time, but component performance in reducing inventory levels varied. The
three services reported reductions in their levels of secondary inventories
by amounts ranging from 24 to 38 percent. The Defense Logistics Agency
realized a slightly smaller decrease of about 10 percent, primarily because
the Department transferred management of many consumable inventories from
the services to the Defense Logistics Agency during this time. 1 This
transfer helped the services meet their inventory reduction goals.

In 1999, the portions of secondary inventory managed by components varied,
with the Air Force managing the largest share of the Department?s secondary
inventory, as figure 4 shows.

1 In November 1989, the Deputy Secretary of Defense directed a review of
management practices in the Department. The result was a number of decisions
called Defense Management Report Decisions. Decision 926, Nov. 9, 1989,
?Consolidation of Inventory Control Points,? recommended the transfer of the
management of nearly all consumable items from the services to the Defense
Logistics Agency. Phase 1 of the Consumable Item Transfer program involved
the transfer of 760, 000 consumable supplies and spare parts, began in
August 1991, and concluded in November 1995. Items affected by phase 2 were
unique end items that had critical applications or required intensive
management. Under phase 2, which began in January 1996, 142, 706 items were
transferred to the Defense Logistics Agency. When the transfer ended in
1998, the Defense Logistics Agency was responsible for managing about 85
percent of the Department consumable item inventory.

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Page 15 GAO- 01- 475 Defense Inventory

Figure 4: Components? Shares of Department of Defense Secondary Inventory,
September 30, 1999

Legend DLA = Defense Logistics Agency

Source: Our analysis of the Department Supply System Inventory Report?s data
for fiscal year 1999.

The percent of component secondary inventories held in economic retention
status Department- wide fell slightly, from 15. 7 percent to 14. 7 percent
between fiscal year 1991 and 1999. However, there were sizeable shifts in
the percent of secondary inventory held in economic retention status by each
component. For example, shares of secondary inventory held in economic
retention status by the Army and Navy fell while the Air Force and Defense
Logistics Agency portions increased (see fig. 5). Economic Retention

Shares of Secondary Inventory Have Decreased Overall

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Page 16 GAO- 01- 475 Defense Inventory

Figure 5: Share of Secondary Inventory Held in Economic Retention Status, by
Component and Department- wide, September 30, 1991, and September 30, 1999

Legend DLA = Defense Logistics Agency DOD = Department of Defense

Source: Our analysis of Department Supply System Inventory Report data for
fiscal years 1991 and 1999.

There was variance in the share of the Department?s economic retention
inventories managed by Defense components, with the Air Force holding almost
half, as figure 6 shows.

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Page 17 GAO- 01- 475 Defense Inventory

Figure 6: Components? Shares of Department of Defense Economic Retention
Inventory, September 30, 1999 (percent)

Legend DLA = Defense Logistics Agency

Source: Our analysis of Department of Defense Supply System Inventory Report
data for fiscal year 1999.

Currently, the services manage mostly reparable items because management
responsibilities for nearly all consumable items have been transferred to
the Defense Logistics Agency. 2 On September 30, 1999, over 34 percent ($ 3.
2 billion) of the Department?s consumable inventory was in economic
retention status. The Defense Logistics Agency managed $2. 7 billion (85
percent) of the Department?s consumable economic retention inventory. The
percentage of total Department consumable and reparable stock in economic
retention status as of September 30, 1999, varied widely by component.

Department consumable inventory in economic retention status fell by about
42 percent between September 30, 1993, and September 30, 1999. Each service
reduced its consumable inventory in economic retention by an average of 73
percent between 1993 and 1999. However, the Defense

2 As defined in Department Regulation 4140. 1- R, appendix 16, consumable
items are items of supply (except explosive ordnance and major end items of
equipment) that are normally expended or used up beyond recovery in the use
for which the item is designed or intended. The Department defines reparable
items as being items of supply subject to economical repair and for which
the repair is considered in satisfying computed requirements at any
inventory level. The majority dollar value of Department inactive inventory
is reparable inventory. Trends in Consumable

and Reparable Inventories in Economic Retention Status

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Page 18 GAO- 01- 475 Defense Inventory

Logistics Agency?s consumable economic retention inventories were reduced by
only 26 percent during this time. By 1993, the Defense Logistics Agency
consumable retention inventory had more than doubled to $3. 7 billion from
its 1991 amount of $1. 8 billion due to the Consumable Item Transfer
program. The 1993 amount subsequently fell to $2.7 billion in 1999. After
the transfer of most of the services? consumable items, the Defense
Logistics Agency held about 85 percent of the Department?s consumable
inventory (see fig. 7).

Figure 7: Components? Shares of Department of Defense Consumable Economic
Retention Inventory, September 30, 1999

Legend DLA = Defense Logistics Agency

Source: Our analysis of the Supply System Inventory Report?s data for fiscal
year 1999.

The Air Force?s share of Department of Defense reparable items in economic
retention status increased between 1993 and 1999. The Air Force reparable
inventory in economic retention status increased by $526 million (about 14
percent) during this time. In contrast, the Navy reduced reparable inventory
in economic retention status by 55 percent 3 and the Army reduced its
inventory in economic retention status by 63 percent between 1993 and 1999.
By 1999 the Air Force managed over 70 percent of

3 The Navy redefined the status of its reparables in economic retention
status in September 1994; according to Navy officials this approach resulted
in more stock being put into disposal or contingency status and reduced
inventories by $1 billion.

Appendix I: Changes in Secondary and Economic Retention Inventory Levels

Page 19 GAO- 01- 475 Defense Inventory

total Department reparable inventories in economic retention- up from 48
percent in 1993 (see fig. 8). 4

Figure 8: Components? Shares of Department of Defense Reparable Economic
Retention Inventory, September 30, 1999

Note: The Defense Logistics Agency does not hold any inventory described as
reparable in its accounting systems.

Source: Our analysis of Supply System Inventory Report data for fiscal year
1999.

4 The Department?s Supply System Inventory Report included consumable and
reparable inventory data separately for the first time in 1993.

Appendix II: Information on Component Economic Retention Models

Page 20 GAO- 01- 475 Defense Inventory

The following sections provide more detail on the general factors and
specific models the components use to determine economic retention limits
and the ceilings the components use.

An economic retention methodology for determining how many items are to be
kept in economic retention status, which the Department of Defense requires,
should compare the costs of retention to the costs of disposal of an
inventory item. 1 In practice, the components configure their models in
different ways and often use distinct values for the factors and assumptions
in their models. The solution the models seek, the economic retention limit,
depends on how the models are set up and factors used in the models when
calculating retention and disposal costs.

The principal factors in calculating the cost of retention include the
estimated costs of operating storage facilities based on the estimated value
of the item in storage and the probability of damage, theft, or loss. The
component responsible for managing the item estimates the cost factors by
taking a percentage of the value, or price, of the item. Components
typically determine that the appropriate price to use in this calculation is
the last price paid for an item. If an item needs repair, this cost is also
added. The percentages for all the cost factors are then multiplied by an
item's price and totaled to obtain the estimated cost of retention. The end
result of these calculations is that the model's estimate of retention cost
rises for each additional unit of an item held.

Another consideration for determining retention cost is the possibility that
an item may become obsolete. An item could be replaced by a new item or the
weapon system the item supports could be discontinued. To address this
possibility, the value of the item is reduced based on estimates by the
managing component of the likelihood it would not be used. For example, if
an item is valued at $100 and there is a 0. 9 probability that the item
would be needed in a future period, then the $100 value of the item would be
multiplied by 0. 9 to yield a future value of $90, based on the 10- percent
possibility of obsolescence.

Almost two- thirds of the Department?s economic retention stock consists of
reparable items. Some reparable items in inactive status are in disrepair

1 Department of Defense Materiel Management Regulation 4140.1- R, section
C4. 2. 1. 2. 1., Office of the Under Secretary of Defense for Acquisition
and Technology, May 1998. Appendix II: Information on Component

Economic Retention Models General Factors in an Economic Retention Model

Retention Cost Factors

Appendix II: Information on Component Economic Retention Models

Page 21 GAO- 01- 475 Defense Inventory

or need to be upgraded. In these cases, a repair cost is factored into the
calculation, which is typically based on the item?s estimated value in its
unrepaired condition. Components also subtract an item?s repair cost from
the price of a new item. Additionally, delivery costs (based on a percentage
of the item?s price) for moving the item to a repair facility are estimated
and added to the computation of retention costs.

A key factor in determining the disposal cost of an item is the chance of
having to repurchase the item later. To calculate the disposal cost, a
component estimates the probability of repurchase and typically multiplies
it by the item?s estimated price, which is the estimated purchase price in
the future. Some components include estimates of administrative costs
associated with procuring an item (such as contract costs) and the cost of
starting up a production line to manufacture the item. Estimated future
prices can be higher than original purchase prices because additional costs
could be incurred, such as the administrative costs involved in contracting
with a manufacturer, and setup costs. However, items may become obsolete in
the future. This estimated reduction in demand caused by obsolescence is
also factored into the retention cost. Since the probability of repurchase
declines as the amount held increases, the estimated disposal cost of the
item declines as more items are held.

Additional adjustments to the disposal cost include estimated expenses
associated with disposal (such as transportation to a disposal facility and
conversion of sensitive items for sale) and administrative costs incurred in
selling an item. These costs are estimated as a percentage of the price of
an item and added to the disposal cost. Finally, an item?s estimated salvage
value would be deducted from the estimated disposal cost to obtain the net
disposal cost for each unit.

The time value of money is also considered in determining whether to retain
an item because the costs associated with storing or disposing of an item
are incurred in the future. The costs and benefits of retaining or disposing
of an item are computed on a present value basis (their value today) to
equalize all of the costs and benefits. To make the present value
calculation, a discount (interest) rate is used to factor in the time value
of money. For example, if the model computes a benefit in one year that
equals one dollar, discounting by 10 percent would make that present
discounted value today equal to 91 cents.

Economic retention models use additional factors to determine whether to
accept items offered for return from depots or other activities. If a return
is accepted, then the estimated cost associated with that action would
Disposal Cost Factors

Other Factors

Appendix II: Information on Component Economic Retention Models

Page 22 GAO- 01- 475 Defense Inventory

have to be included. For example, estimated transportation costs to return
an item would be added to estimated storage costs and compared to estimated
disposal costs to decide whether to accept the return of an item.

The Army, the Navy, and the Defense Logistics Agency use economic models to
calculate the economic retention limit for each item they analyze; the
components then apply the ceiling limits described in this report. The Air
Force model does not compare the costs of retention to the costs of
disposal. Instead, the Air Force employs an item?s usage history and its
retention ceiling to determine whether to dispose of an item. The specific
factors and assumptions the three components? models use and the Air Force
computation are described in the following sections.

The Army model compares retention and disposal values to determine the
economic retention limit. In calculating disposal costs, the Army model
includes any potential disposal value. In calculating retention costs, the
model includes the benefits of retention, i. e., not having to reorder an
item. The model also includes the probability of obsolescence. The Army
model allows for the possibility of obsolescence to vary depending on the
age of the item. The Army uses a 7- percent discount rate to calculate the
present value of disposal and retention values.

The Navy model compares what it calls holding costs to buy back costs to
determine the economic retention requirement. Holding costs include a
disposal value minus disposal transportation costs. The model also factors
in an inventory loss rate and an obsolescence rate in computing storage
costs. The major factors of buy back costs are those costs that would be
incurred if the item had to be reordered and include item replacement price,
the administrative costs of procurement, and manufacturer's set- up costs to
manufacture unique items (if needed). The model computes buy back and
retention costs on a present value basis (using a 10- percent discount rate)
because reorder costs would be incurred in the future. If the item is
reparable, repair costs (if appropriate) are included in the computation to
account for the costs that would be incurred for making the item ready for
issue.

The Defense Logistics Agency's holding costs are determined by computing
storage costs times the probability of needing the material. Holding costs
are compared to expected disposal costs. If holding costs are less than
disposal costs (expected possible costs of reprocurement minus disposal
proceeds), the item should be retained. The model can also evaluate whether
an item manager should accept a return and pay the associated costs to bring
an item back from retail operations or deny the Component Economic

Retention Models Army

Navy The Defense Logistics Agency

Appendix II: Information on Component Economic Retention Models

Page 23 GAO- 01- 475 Defense Inventory

return and take the chance of reprocurement in the future. The costs of
accepting a return are added to the expected holding costs, and the sum is
compared to the expected cost to dispose and repurchase. The analysis is
done on a present value basis using a discount rate of 7 percent because
both costs involve potential outlays in the future.

The Air Force model does not compute the costs of disposal or storage as
required by the Department. The Air Force computation of economic retention
limits is based on an item's usage history. The current Air Force model
distinguishes between inactive items that have experienced zero demand over
a 5- year span, and have no foreseeable demand, and demand items- those with
a demand history over the past 5 years.

If an item has been categorized as inactive, the retention limit is based on
a non- demand amount- called either an insurance or numeric stock objective
limit. A maximum of five items is held for numeric stock objective items and
a maximum of two items is held for insurance items. According to component
officials and other experts, most Air Force items are in these categories.

For items that are categorized as demand items, the economic retention limit
is determined through several steps. The system computes gross retention
levels first by adding 9 years of demand to peak requirements (the highest
quarterly demand level for the item from the prior 25 quarters). Second, the
minimum retention level for stock needing repairs (unserviceable items) is
computed by adding (1) condemned stock for the past 9 years, (2) stock held
at supply facilities and (3) peak requirements identical to the factors
computed for the gross retention level. Third, these two levels are then
adjusted to establish maximum (gross) and minimum inventory levels, which
are compared to the number of assets in inventory. Unserviceable items are
retained only when the number of serviceable assets falls below the minimum
retention level.

Table 2 summarizes the factors and assumptions used in component economic
retention models. Air Force

Appendix II: Information on Component Economic Retention Models

Page 24 GAO- 01- 475 Defense Inventory

Table 2: Retention and Disposal Cost Factors in Components? Economic
Retention Models Component Retention factors Disposal factors

Air Force None None Army Retention values:

unit price obsolescence rate storage cost

Disposal values: unit price disposal value

Navy Holding cost: item replacement price storage cost rate obsolescence
rate loss rate

Buy back cost: item replacement price manufacturer?s setup cost procurement
order cost (administrative) obsolescence rate Defense Logistics Agency Hold
option:

item unit price storage cost rate probability of needing the material

Disposal benefit and possible repurchase cost: item unit price repurchase
price rate (percent of item unit price) cost to reorder (administrative) net
disposal rate (salvage value) Defense Logistics Agency disposal cost
(administrative) probability of needing the material

Source: Our analysis of component information.

Economic retention models and ceilings are to be applied only to items with
predictable and steady annual demand. Limited- demand items- those with no
or infrequent demand in a year- are also held in economic retention status.
They are retained even though the probability of demand is low because the
lack of the items would seriously hamper the operational readiness of a
weapon system. None of these inventory items would be considered for
disposal unless they exceeded Department inventory requirements for limited-
demand items.

The majority of items held in economic retention status fell into
limiteddemand status. Table 3 details the number and dollar value of items
that the retention models and ceilings would analyze and the items that
would not be analyzed due to low or no demand. Many Inventory Items with

Limited Demand Not Subject to Economic Retention Model Determinations

Appendix II: Information on Component Economic Retention Models

Page 25 GAO- 01- 475 Defense Inventory

Table 3: Application of Economic Retention by Type of Item, by Component, as
of September 30, 1999 (number and dollar value)

Component Economic retention formula

applies (number/ dollar value)

Economic retention formula does not apply (number/ dollar value)

Air Force 10,874 items $2. 0 billion

56,370 items $2. 4 billion Army 23,721 items

$0. 8 billion 1,259 items

$0. 014 billion Navy 37,035 items

$0. 6 billion 270,156 items

$1. 6 billion Defense Logistics Agency 2. 8 million items

$2. 7 billion Source: Component officials.

As discussed in the body of this report, components used a variety of
maximum levels of demand (commonly called ceilings by components) to further
reduce inventory levels (low- demand items noted in table 3 above were not
affected by ceiling limits). Three of the four components reviewed imposed
ceilings on the number of years of demand for all items, and most inventory
control points in the fourth component (the Army) also used ceilings.

Through most of the 1990s, the Army set a different ceiling on how many
years of demand can make up the retention limit, depending on the
characteristics of the item. According to Army officials, in 1992 the Army
adjusted its factors to a ceiling of 7 years of demand above requirements
for essential items (items that directly support critical parts of a weapon
system). The Army used a ceiling of 4 years of demand above requirements for
nonessential items (items that do not directly support the critical parts of
a weapon system). In December 1999 the Army revised the ceiling limits to 7
years of demand above requirements for serviceable reparables, 6 years of
demand above requirements for unserviceable reparables, and 5 years of
demand above requirements for all other items. At the same time the Army
decided to let each inventory control point set its own ceilings. An Army
official at one of its five inventory control points stated that his
inventory control point is using the determinations of the economic
retention model for setting retention inventory limits without the ceilings.
Army officials at the other four inventory control points stated that they
have not changed their ceilings. The Structure of

Component Ceilings

Appendix II: Information on Component Economic Retention Models

Page 26 GAO- 01- 475 Defense Inventory

In 1994, the Navy chose to implement ceilings of years of demand depending
on the expected future growth of the weapon system. The Navy computation
system for inventory calculates a single inventory limit that includes
active requirements and inactive inventory. As a result, the demand ceiling
(4, 8, or 12 years of total demand) applies to the entire computed demand
limit- not just economic retention limits. The ceiling is 12 years of total
demand for items supporting new weapon systems. A ceiling of 8 years of
total demand is applied to items supporting weapons systems in common use
and 4 years of total demand is applied to items supporting weapons systems
approaching obsolescence.

The Defense Logistics Agency ceiling for items in economic retention,
implemented in 1994, is 6 years of total demand. The prior ceiling on years
of demand for inactive inventory was 10 years of total demand.

The Air Force ceiling for inventory in economic retention status was reduced
from 20 years of total maximum demand to 13 years of total maximum demand in
1996, according to Air Force officials.

Appendix III: Scope and Methodology Page 27 GAO- 01- 475 Defense Inventory

To determine component approaches to making economic retention decisions, we
interviewed Department and component officials who managed economic
retention models and ceilings in the Defense Logistics Agency, the Army, the
Navy, and the Air Force and reviewed documents they provided. Because the
Marine Corps had less than 1 percent of the Department?s total economic
retention inventory, we did not include the Marine Corps in our analysis. We
did not independently test or validate the component models or inventory
systems.

We interviewed officials and gathered relevant documentation for our review
at the following locations:

 The Office of the Deputy Under Secretary of Defense (Logistics),
Washington, D. C.

 The Defense Logistics Agency, Ft. Belvoir, Virginia.

 The Army Material Systems Analysis Activity, Aberdeen, Maryland.

 The Air Force Materiel Command, Wright- Patterson Air Force Base, Ohio.

 The Navy Inventory Control Point, Mechanicsburg, Pennsylvania. To support
our analysis of trends in component inventory levels, we analyzed
information in the Department of Defense Supply System Inventory Report for
September 30 of fiscal years 1991 through 1999. Our analysis included data
for fiscal year 1991 through 1999 for trends in economic retention. We
reviewed data from these time periods because the Department changed the way
dollar estimates of inventory were calculated in 1990; inventory data was
reported consistently in the Supply System Inventory Report from 1991
through 1999. Component officials stated that there were differences in the
categories of inventory components reported in the Supply System Inventory
Report. We did not adjust the Supply System Inventory Report data for this
analysis. We analyzed data for reparable and consumable items for fiscal
years 1993 through 1999 because the Department first separately reported
data on reparable and consumable items in its 1993 Supply System Inventory
Report. Our review focused on the models and items with predictable demand.
We did not analyze items with limited demand (see app. II). We did not
independently test or validate the accuracy of the data reported in these
inventory systems. We adjusted Department data to account for inflation as
part of our analysis of component performance in realizing inventory
reductions over time.

We conducted our work between May 1999 and January 2001 in accordance with
generally accepted government auditing standards. Appendix III: Scope and
Methodology

Appendix IV: Comments From the Department of Defense

Page 28 GAO- 01- 475 Defense Inventory

Appendix IV: Comments From the Department of Defense

Appendix IV: Comments From the Department of Defense

Page 29 GAO- 01- 475 Defense Inventory (709423)

The first copy of each GAO report is free. Additional copies of reports are
$2 each. A check or money order should be made out to the Superintendent of
Documents. VISA and MasterCard credit cards are also accepted.

Orders for 100 or more copies to be mailed to a single address are
discounted 25 percent.

Orders by mail:

U. S. General Accounting Office P. O. Box 37050 Washington, DC 20013

Orders by visiting:

Room 1100 700 4 th St., NW (corner of 4 th and G Sts. NW) Washington, DC
20013

Orders by phone:

(202) 512- 6000 fax: (202) 512- 6061 TDD (202) 512- 2537

Each day, GAO issues a list of newly available reports and testimony. To
receive facsimile copies of the daily list or any list from the past 30
days, please call (202) 512- 6000 using a touchtone phone. A recorded menu
will provide information on how to obtain these lists.

Orders by Internet

For information on how to access GAO reports on the Internet, send an e-
mail message with ?info? in the body to:

Info@ www. gao. gov or visit GAO?s World Wide Web home page at: http:// www.
gao. gov

Contact one:

 Web site: http:// www. gao. gov/ fraudnet/ fraudnet. htm

 E- mail: fraudnet@ gao. gov

 1- 800- 424- 5454 (automated answering system) Ordering Information

To Report Fraud, Waste, and Abuse in Federal Programs
*** End of document. ***