Military Personnel Strengths in the Army National Guard 	 
(20-MAR-02, GAO-02-540R).					 
								 
The Army National Guard's funding requests for fiscal years 2000 
and 2001 were overstated by $42.9 million and $31.6 million,	 
respectively, because of inaccurate military strength and	 
participation rates used to develop projected and actual military
force levels. To correct these overstatements, the Guard is	 
placing more emphasis on an existing personnel database reporting
system that identifies the personnel assigned to a unit but who  
have not been paid for inactive duty training for three months or
more. The Guard also improved the method it uses to calculate	 
inactive duty training participation rates, now basing the rate  
on the number of people who have actually been paid for training,
rather than on expected program costs.				 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-02-540R					        
    ACCNO:   A02910						        
  TITLE:     Military Personnel Strengths in the Army National Guard  
     DATE:   03/20/2002 
  SUBJECT:   Army personnel					 
	     Budget administration				 
	     Military budgets					 
	     Military training					 
	     Reporting requirements				 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** 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-02-540R
     
GAO- 02- 540R National Guard Personnel Strengths

United States General Accounting Office Washington, DC 20548

March 20, 2002 The Honorable John McHugh Chairman The Honorable Vic Snyder
Ranking Member Military Personnel Subcommittee Committee on Armed Services
House of Representatives

Subject: Military Personnel Strengths in the Army National Guard The
accuracy of reported personnel strength and training participation rates has
a direct impact on the reliability of the Army National Guard?s budget and
the allocation of funds to individual states. If either the reported
strength levels or the participation rates for a given fiscal year are more
or less than the actual numbers, the funds required to pay Guard personnel
will be either overstated or understated. Congressional concerns about the
reported military personnel strengths of the Army National Guard have
emerged as a result of recent media coverage of the Guard?s socalled ghost
soldiers. 1

As a result of those concerns, you asked us to provide information on (1)
the Guard?s personnel strength levels and training participation rates and
(2) the Guard?s efforts to improve the accuracy of reported strength levels
and participation rates. To respond to your request, we drew on findings
from our annual review of the Department of Defense?s military personnel
budget requests and the Army National Guard?s military personnel data for
fiscal years 2000 and 2001 and the first quarter of fiscal year 2002. The
scope and methodology for our review is discussed on page 5.

Results in Brief

The Army National Guard?s fiscal years 2000 and 2001 funding requests were
overstated by $42.9 million and $31.6 million, respectively, because the
Guard used inaccurate military strength and participation rates to develop
its projected and actual military force levels. Additionally, to develop its
training budget needs, it used a mathematically derived training
participation rate based on expected program costs rather than on the actual
number of personnel being trained. By using these inaccurate figures, the
Guard overstated its overall military personnel strength and

1 ?Ghost soldier? is a slang term used for soldiers who remain on strength
reports but who are, in fact, no longer participating in training and who
should be removed from these reports.

GAO- 02- 540R National Guard Personnel Strengths Page 2 the amount of its
annual funding requests to Congress.

The Army National Guard is currently taking steps to correct these
overstatements. It is placing more emphasis on an existing personnel
database reporting system that identifies the personnel who are assigned to
a unit but have not been paid for inactive duty training for 3 months or
more. By doing this, the Guard can ensure that unit commanders remove these
personnel from unit strength reports if they are no longer determined to be
drilling reservists. 2 The Guard has also improved the method it uses to
calculate inactive duty training participation rates, now basing the rate on
the number of people who have actually been paid for training.

Personnel Strength Figures and Training Participation Rates Were Overstated

Our analyses of the Army National Guard?s military strength projections for
fiscal years 2000 and 2001 showed that the Guard overstated its personnel
strength because it relied on inaccurate military personnel strength data,
which included individuals who should not have been considered in the
calculation of strength numbers for inactive duty training. As a result, we
estimated that the budget requests for those two fiscal years were
overstated by $42.9 and $31.6 million, respectively.

The Guard can remove an individual from strength reports after 3 months if
it determines that the person is no longer in the program. In order to help
commanders identify these individuals, the Guard publishes a monthly Non-
Validation of Pay Report (NO- VAL). Unit commanders review the status of
individuals on this report and determine if they should be excused, removed,
or reclassified to a non- drilling status in the Guard?s strength reports.
Because each personnel action is unique, there is little guidance as to how
long a unit commander?s review and the processing of paperwork should take.
We used the 7- month rather than the 3- month period to estimate the
accuracy of reported strength for drilling personnel because there are a
number of circumstances that would cause a person not to be paid for more
than three months and still be included in unit strength figures. These
reasons include their movement from one unit to another, their inability to
perform training for medical reasons, and their being paid late for training
performed. Guard officials agreed that it would be reasonable to expect unit
commanders to adjust unit strength if an individual has not been paid for at
least 7 months or more.

Our analysis of the Army National Guard?s military personnel database used
to develop the NO- VAL showed that about 4,048, or 1.3 per cent, of the
301,140 drilling reservists should have been dropped from the fiscal year
2000 end strength and about 4,254, or 1.4 per cent, of the 296,430 drilling
reservists should have been removed from the fiscal year 2001 end strength.
Enclosure I shows the number of personnel,

2 An individual required to perform 2 weeks of annual training and weekend
drills (inactive duty training).

GAO- 02- 540R National Guard Personnel Strengths Page 3 by state, who were
not paid for 3 and 7 or more consecutive months as of the end of

fiscal years 2000 and 2001. In looking at the Army National Guard ?s method
for calculating its inactive duty training participation rates, we found
that in the past the rates were inaccurate because they did not correctly
identify the actual number of personnel who were, in fact, in training.
Instead, the Guard relied on a mathematically derived participation rate,
which was based on expected program costs, estimated training costs, and
military strength figures, to come up with a total number of military
personnel who were expected to train. This method resulted in inactive duty
training participation rates that were higher than they should have been.
For example, when we examined the Guard?s fiscal year 2001 budget, we found
that the Guard had determined- using mathematically derived rates from
fiscal year 1999 numbers- that about 91 percent of its officers and 84
percent of its enlisted personnel would participate in inactive duty
training. However, when we compared the number of personnel who had actually
been paid for inactive duty training in 1999 with the mathematically derived
numbers, we found that 88.7 percent of officers and 81.3 percent of enlisted
personnel had actually trained.

Steps Underway to Improve the Accuracy of Military Personnel Strengths and
Training Participation Rates

The Army National Guard?s methods of determining military personnel strength
and inactive duty training participation rates have improved.

In the course of our budget work we made a number of suggestions on how the
Army National Guard could improve its budget formulation methods. As a
result, the Guard has changed the method it uses to calculate inactive- duty
training participation rates and is now basing them on the number of people
who have actually been paid for training. In addition, the Guard has placed
more command attention on the accuracy of reported military personnel
strength and the number of NO- VAL personnel retained in the reporting
system. Between October 31, 1999, and December 31, 2001, the number of
individuals reported on the Guard?s NO- VAL report has declined from 16,264
to 9,627. Enclosure II shows this trend.

Our review of the December 2001 military personnel database indicates that
some state commanders are using the NO- VAL report to identify inaccuracies
in reported personnel strength. For example, between November and December
2001, the number of assigned drilling personnel was reduced from 297,846 to
297,226, or less than 1 percent, while personnel on the NO- VAL report
declined from 11,133 to 9,627, or about 14 percent. The state of Texas had
the largest decrease in both strength and NO- VAL personnel. Its assigned
drilling personnel strength numbers fell from 14,522 to 13,695, about 6
percent, and its personnel on the NO- VAL report declined from 1249 to 361,
a 70 percent reduction.

GAO- 02- 540R National Guard Personnel Strengths Page 4

Scope and Methodology

To provide information on the Guard?s personnel strength and participation
rates, we drew on our prior work and analyzed DOD?s military personnel
budgets, comparing requests for fiscal years 2000 and 2001 to actual
personnel data for October 1999 to December 2001. In addition, we obtained
and analyzed the database used to produce the monthly NO- VAL reports for
fiscal years 2000 and 2001. We also discussed our observations with Army
National Guard officials at the headquarters level and officials at the
Office of the Secretary of Defense, Reserve Affairs. Additionally, although
we utilized the Guard?s data in our analyses, we did not test this data to
ascertain its accuracy.

Agency Comments

We discussed a draft of this letter with Army National Guard officials. They
generally agreed with our observations and stated that, in the past,
reported personnel strength levels might have been unintentionally
overstated. The Guard stressed that it has recognized the problems it had in
calculating participation rates and in adjusting military personnel strength
levels and is taking action, as discussed above, to correct both.

The Office of the Secretary of Defense, Reserve Affairs, generally agreed
with our observations. We will continue to work with the Guard and the
Office of the Secretary of Defense, Reserve Affairs, to improve the accuracy
of reported strength and participation rates used in the budget formulation
process.

As arranged with your office, unless you announce its contents earlier, we
plan no further distribution of this report until 30 days after the date on
this letter. At that time, we will make copies of this letter available to
other appropriate congressional committees and place a copy on GAO?s home
page at http:// www. gao. gov. If you have any questions concerning the
information provided, please call me on (202) 512- 5559 or R. L. Furr on
(202) 512- 5426.

Derek B. Stewart Director, Defense Capabilities

and Management Enclosures - 2

ENCLOSURE I ENCLOSURE I Page 5 GAO- 02- 540R National Guard Personnel
Strengths

Number of Army National Guard Members Not Paid for 3 and 7 or More
Consecutive Months September 30, 2000 September 30, 2001 State Assigned a 3
months % 7 months % Assigned a 3 months % 7 months % Alabama 11,837 368 3.1
132 1.1 11,184 274 2.4 132 1.2 Alaska 1,457 179 12.3 77 5. 3 1,356 107 7.9
65 4. 8 Arizona 3, 276 70 2.1 7 0.2 3, 239 34 1.0 8 0.2 Arkansas 7,090 200
2.8 70 1. 0 7,139 246 3.4 118 1.7 California 13,965 444 3.2 112 0.8 13,918
524 3.8 238 1.7 Colorado 2,703 68 2.5 19 0. 7 2,586 50 1.9 11 0. 4
Connecticut 3,173 71 2.2 28 0. 9 3,193 177 5.5 90 2. 8 Delaware 1,410 42 3.0
11 0. 8 1,388 30 2.2 9 0.6 District of Columbia 1,333 83 6.2 35 2. 6 1,269
43 3.4 17 1. 3 Florida 8, 564 382 4.5 178 2.1 8, 485 276 3.3 142 1.7 Georgia
7, 556 340 4.5 114 1.5 7, 178 192 2.7 81 1. 1 Guam 534 4 0. 7 0 0. 0 548 4
0. 7 0 0. 0 Hawaii 2,427 74 3.0 34 1. 4 2,436 54 2.2 25 1. 0 Idaho 2,190 18
0.8 3 0.1 2, 156 28 1.3 10 0. 5 Illinois 8, 439 326 3.9 115 1.4 8, 162 242
3.0 116 1.4 Indiana 10,099 574 5.7 218 2.2 10,794 472 4.4 221 2.0 Iowa 6,191
120 1.9 43 0. 7 6,078 128 2.1 56 0. 9 Kansas 5,429 220 4.1 48 0. 9 5,128 169
3.3 85 1. 7 Kentucky 5,509 153 2.8 62 1. 1 5,563 107 1.9 30 0. 5 Louisiana
8, 317 122 1.5 50 0. 6 8,379 112 1.3 62 0. 7 Maine 1, 931 36 1.9 4 0.2 1,
930 43 2.2 16 0. 8 Maryland 5,434 261 4.8 100 1.8 5, 249 195 3.7 100 1.9
Massachusetts 6,534 245 3.7 82 1. 3 6,145 150 2.4 62 1. 0 Michigan 7,404 190
2.6 68 0. 9 7,441 172 2.3 77 1. 0 Minnesota 8, 145 177 2.2 30 0. 4 8,000 123
1.5 47 0. 6 Mississippi 8,015 140 1.7 50 0. 6 7,840 98 1.3 42 0. 5 Missouri
6, 574 227 3.5 78 1. 2 6,614 198 3.0 90 1. 4 Montana 2,099 44 2.1 16 0. 8
1,989 35 1.8 15 0. 8 Nebraska 2,835 142 5.0 41 1. 4 2,643 21 0.8 8 0.3
Nevada 1,389 46 3.3 7 0.5 1, 466 21 1.4 5 0.3 New Hampshire 1, 455 22 1.5 6
0.4 1, 431 19 1.3 8 0.6 New Jersey 6,170 682 11.1 282 4.6 5, 984 659 11.0
437 7.3 New Mexico 2,562 89 3.5 22 0. 9 2,601 103 4.0 27 1. 0 New York
10,368 460 4.4 172 1.7 9, 831 262 2.7 103 1.0 North Carolina 8, 572 393 4.6
132 1.5 8, 580 379 4.4 216 2.5 North Dakota 2,753 16 0.6 5 0.2 2, 728 17 0.6
7 0.3 Ohio 8,124 132 1.6 31 0. 4 8,594 110 1.3 25 0. 3 Oklahoma 6, 184 192
3.1 81 1. 3 6,055 132 2.2 42 0. 7 Oregon 5,046 182 3.6 82 1. 6 4,915 100 2.0
52 1. 1 Pennsylvania 13,748 349 2.5 140 1.0 13,719 362 2.6 204 1.5 Puerto
Rico 7,471 149 2.0 72 1. 0 7,497 148 2.0 68 0. 9 Rhode Island 2,198 110 5.0
38 1. 7 2,018 63 3.1 29 1. 4 South Carolina 8,279 397 4.8 208 2.5 7, 826 159
2.0 87 1. 1 South Dakota 2,917 34 1.2 8 0.3 2, 859 32 1.1 12 0. 4 Tennessee
9,422 236 2.5 107 1.1 9, 310 167 1.8 99 1. 1 Texas 14,546 1, 259 8.7 505 3.5
14,138 865 6.1 490 3.5 U. S. Virgin Islands 640 36 5.6 18 2. 8 606 25 4.1 16
2. 6 Utah 4,005 117 2.9 30 0. 7 3,923 105 2.7 39 1. 0 Vermont 2,689 136 5.1
58 2. 2 2,555 99 3.9 48 1. 9 Virginia 6,339 225 3.5 92 1. 5 6,256 165 2.6 95
1. 5 Washington State 4, 728 227 4.8 62 1. 3 4,680 175 3.7 82 1. 8 West
Virginia 3,404 66 1.9 18 0. 5 3,407 50 1.5 22 0. 6 Wisconsin 6, 418 104 1.6
29 0. 5 6,228 152 2.4 54 0. 9 Wyoming 1, 243 46 3.7 18 1. 4 1,193 28 2.3 14
1. 2

Total 301,140 11,025 3. 7 4,048 1.3 296,430 8, 701 2.9 4, 254 1.4 a Assigned
includes only Army National Guard members required to perform 2 weeks of
annual training and weekend drills.

ENCLOSURE II ENCLOSURE II Page 6 GAO- 02- 540R National Guard Personnel
Strengths

Army National Guard Monthly NO- VAL Reports on Individuals Not Paid for
Inactive Duty Training

for 3 Months or More (October 31,1999 to December 31, 2001)

Note: The graph shows a decline from 16,264 in October 1999 to 9, 627 in
December 2001. Source: U. S. Army National Guard monthly NO- VAL reports.

(350173)
*** End of document. ***