Defense Spending and Employment: Information Limitations Impede Thorough
Assessments (Letter Report, 01/14/98, GAO/NSIAD-98-57).

Pursuant to a congressional request, GAO examined defense and other
federal spending in the state of New Mexico, focusing on: (1)
characteristics of New Mexico's economy and changes in it; (2) the
amount of direct defense-related and nondefense-related federal spending
in the state and the direct federal employment associated with both,
over time; and (3) the extent to which available government data can
provide reliable information on defense spending and employment.

GAO noted that: (1) New Mexico is home to two Department of Energy (DOE)
national laboratories and four Department of Defense (DOD) military
installations, among other federal activities; (2) state officials
indicate that New Mexico's economy is "heavily dependent" upon federal
expenditures; (3) in 1996, New Mexico was fourth among states in the per
capita distribution of federal dollars and first in return on federal
tax dollars; (4) while parts of the state have relatively strong
economies, in 1994 New Mexico's poverty rate was the second highest in
the country and its per capita income was 48th in the country; (5)
although defense-related spending has been declining, New Mexico's gross
state product and total per capita income have been increasing,
indicating that the economy is growing and that efforts to diversify the
economy may be having a positive effect; (6) one can learn several
things from the available federal government expenditure and employment
data for New Mexico; (7) DOD and DOE expenditures have consistently
represented the largest share of all federal expenditures for
procurement and salaries and wages in New Mexico; (8) defense-related
employment has also consistently represented the largest share of total
federal employment in New Mexico, including retired federal workers; (9)
DOD and DOE do not contribute equally on types of defense-related
spending or defense-related employment, revealing relevant distinctions
between the types of direct economic contributions made by these
agencies; (10) DOE contributes most in federal procurement expenditures
and private contractor employment; (11) DOD contributes most in federal
salaries and wages and federal employment, namely active duty military
and retired employees; (12) existing government data, however,
contributes to only a partial understanding of the type of federal
dollars that enter a state's economy and the employment supported by the
expenditures; (13) GAO's research based on New Mexico shows that the
data have limitations that severely restrict the ability to determine
the total amount and distribution of federal funding and jobs in the
state; (14) key limitations include: (a) reporting thresholds that
exclude millions in procurement expenditures; (b) the reporting of the
value of an obligation, rather than the money actually spent; (c) the
absence of any comprehensive source of primary data that systematically
identifies private sector employment associated with federal contracts;
and (d) DOD's lack of data on subcontracts; and (15) since these data
sources are not unique to New Mexico, these limitations would also apply
to assessments of other states.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  NSIAD-98-57
     TITLE:  Defense Spending and Employment: Information Limitations 
             Impede Thorough Assessments
      DATE:  01/14/98
   SUBJECT:  Federal procurement
             Military facilities
             Federal facilities
             State governments
             Federal/state relations
             Defense procurement
             Economic growth
             Statistical data
             Economic analysis
             Economic indicators
IDENTIFIER:  New Mexico
             Federal Assistance Award Data System
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed 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.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Chairman, Committee on the Budget, U.S.  Senate

January 1998

DEFENSE SPENDING AND EMPLOYMENT -
INFORMATION LIMITATIONS IMPEDE
THOROUGH ASSESSMENTS

GAO/NSIAD-98-57

Defense Spending and Employment

(713003)


Abbreviations
=============================================================== ABBREV

  CFFR - Consolidated Federal Funds Report
  DOD - Defense of Defense
  DOE - Department of Energy
  FES - Federal Expenditures by State
  FPDS - Federal Procurement Data System
  FTE - full-time equivalents
  HHS - Department of Health and Human Services
  NASA - National Aeronautics and Space Administration
  OPM - Office of Personnel Management

Letter
=============================================================== LETTER


B-276819

January 14, 1998

The Honorable Pete V.  Domenici
Chairman, Committee on the Budget
United States Senate

Dear Mr.  Chairman: 

The size of the defense budget is a pressing issue in the wake of the
post-Cold War reduction in defense spending and the priorities of
balancing the federal budget.  Moreover, the difficulties of
determining the impacts and outcomes of budget decisions add to
existing challenges.  Congress and other decisionmakers have a
practical and continuous need for information on defense-related
expenditures and employment in states.  In response to your request,
and discussions with your office, we examined defense and other
federal spending in the state of New Mexico.  Specifically, we
examined (1) characteristics of New Mexico's economy and changes in
it; (2) the amount of direct defense-related and nondefense-related
federal spending in the state and the direct federal employment
associated with both, over time; and (3) the extent to which
available government data can provide reliable information on defense
spending and employment.  The last issue became a major component of
our study, since the quality of the data directly affects what can be
concluded about defense spending and employment. 


   BACKGROUND
------------------------------------------------------------ Letter :1

In 1996, the federal government spent $1.4 trillion in U.S.  states
and territories to procure products and services, to fund grants and
other assistance, to pay salaries and wages to federal employees, to
provide public assistance, and to fund federal retirement programs
and Social Security, among other things.  Some states rank relatively
high on the per capita distribution of different types of federal
dollars.  Government reports indicate that in 1996, Maryland,
Virginia, and Alaska were the only three states to rank among the top
five in each of the following categories:  (1) total federal
expenditures, (2) total federal procurement expenditures, and (3)
total salary and wage expenditures for federal workers.  The only
other state that ranked among the top 10 states in all these
categories was New Mexico. 

Interest in the economic magnitude of defense and other federal
expenditures in states has been amplified by concerns over
anticipated outcomes of the post-Cold War drawdown.  In hearings
before the Joint Economic Committee of the 101st Congress, 12 state
governors\1 submitted to the leadership of the Senate and House a
plan for responding to expected adverse economic impacts in states
that were believed to be particularly vulnerable to reductions in
defense spending.  In 1992, President Bush issued Executive Order
12788, requiring the Secretary of Defense to identify the problems of
states, regions, and other areas that result from base closures and
Department of Defense (DOD) contract-related adjustments.  The Office
of Economic Adjustment is DOD's primary office responsible for
providing assistance to communities, regions, and states "adversely
impacted by significant Defense program changes."

The federal government tracks defense-related and other federal
spending and associated employment through various sources. 
Centralized reporting of this information is done by the Census
Bureau in its Consolidated Federal Funds Report (CFFR) series.  The
CFFR includes the Federal Expenditures by State (FES) report and a
separate two-report volume that presents information at the county
and subcounty level.  The FES report presents the most comprehensive
information on federal expenditures at the state level that can
actually be attributed to specific federal agencies or programs. 
Agencies involved in collecting and reporting various types of
employment information include the Office of Personnel Management
(OPM) and the Bureau of Labor Statistics. 

Expenditure information reported in the CFFR also appears in
agency-specific publications or data sources.  DOD reports
information on its total procurement expenditures and the salaries
and wages paid to DOD personnel, by state, in the Atlas/Data Abstract
for the United States and Selected Areas.  In compiling information
for the CFFR, DOD's procurement data are first sent to the Federal
Procurement Data System (FPDS) and then sent to Census.  Therefore,
Census, DOD, and the FPDS can and do report DOD procurement
expenditures. 

Federal expenditure and employment data are available to users in and
outside the government and are regularly used in policy formulation
and evaluation.  DOD contractors, including the Logistics Management
Institute, have used federal government data in support of their work
for DOD on the economic impacts of base realignment and closure
actions.  The Office of Economic Conversion Information, a
collaborative effort between the Economic Development Administration
of the Department of Commerce and DOD, uses existing federal data to
provide information to communities, businesses, and individuals
adjusting to the effects of defense downsizing and other changing
economic conditions.  The Congressional Budget Office and the
Congressional Research Service have also used DOD procurement
expenditure data in examining the expected effects of planned
reductions in the national defense budget.  DOD uses its prime
contract award expenditure data to track the status and progress of
goals associated with contracts made to small businesses. 
Researchers at think tanks, universities, and state government
offices also use government data in a wide array of research projects
and publications. 


--------------------
\1 Governors from Arkansas, Colorado, Connecticut, Louisiana,
Maryland, Massachusetts, Michigan, Mississippi, New Jersey, New York,
Ohio, and Washington. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

New Mexico is home to two Department of Energy (DOE) national
laboratories and four DOD military installations, among other federal
activities.  State officials indicate that New Mexico's economy is
"heavily dependent" upon federal expenditures.  In 1996, New Mexico
was fourth among states in the per capita distribution of federal
dollars and first in return on federal tax dollars, receiving $1.93
in federal outlays for every $1.00 paid in federal taxes.  While
parts of the state have relatively strong and growing economies, in
1994 New Mexico's poverty rate was the second highest in the country
and its per capita income was 48th in the country.  Although
defense-related spending has been declining, New Mexico's gross state
product and total per capita income have been increasing, indicating
that the economy is growing and that efforts to diversify the economy
may be having a positive effect. 

One can learn several things from the available federal government
expenditure and employment data for New Mexico.  DOD and DOE
expenditures have consistently represented the largest share of all
federal expenditures for procurement and salaries and wages in New
Mexico.  Defense-related employment has also consistently represented
the largest share of total federal employment in New Mexico,
including retired federal workers.  DOD and DOE do not contribute
equally on types of defense-related spending or defense-related
employment, revealing relevant distinctions between the types of
direct economic contributions made by these agencies.  DOE
contributes most in federal procurement expenditures and private
contractor employment.  DOD contributes most in federal salaries and
wages and federal employment, namely active duty military and retired
DOD employees. 

Existing government data, however, contributes to only a partial
understanding of the type of federal dollars that enter a state's
economy and the employment supported by the expenditures.  Our
research based on New Mexico shows that the data have multiple
limitations that restrict the ability to determine the total amount
and distribution of federal funding and jobs in the state.  Key
limitations that generally apply include reporting thresholds that
exclude millions in procurement expenditures; the reporting of the
value of an obligation, rather than the money actually spent; the
absence of any comprehensive source of primary data that
systematically identifies private sector employment associated with
federal contracts; and DOD's lack of data on subcontracts.  Since
these data sources are not unique to New Mexico, these limitations
would also apply to assessments of federal funding and employment in
other states.  Existing data are not without value, but those who
rely on federal data need to be alert to their drawbacks and exercise
discretion when using them. 


   FEDERAL DOLLARS CONTRIBUTE TO
   NEW MEXICO ECONOMY, BUT ECONOMY
   IS DIVERSIFYING
------------------------------------------------------------ Letter :3

DOE and DOD military activities have contributed substantially to the
economy of New Mexico for about 50 years.  Government data show that
between 1988 and 1996, New Mexico was ranked second, third, or
fourth, among U.S.  states in per capita distribution of federal
dollars.  In terms of per capita federal procurement expenditures
only, New Mexico was ranked first among U.S.  states during 1988-94
and second in 1995-96.  In 1996, New Mexico was ranked first among
states in return on federal tax dollars, receiving $1.93 in federal
outlays for every $1.00 in federal taxes paid.  The state was also
ranked first in return on federal tax dollars in 1995.  In 1996, 5 of
the 6 major federal facilities were among the top
10 employers in the state. 

This federal revenue comes largely from the six major federal
facilities in New Mexico, including two DOE national laboratories,
Los Alamos National Laboratory and Sandia National Laboratory;
Cannon, Holloman, and Kirtland Air Force Bases; and White Sands
Missile Range, a test range that supports missile development and
test programs for all the services, the National Aeronautics and
Space Administration (NASA); and other government agencies and
private industry.  New Mexico's geography and climate, including
relative isolation from major population centers, year-round good
weather, and open airspace, have made the state attractive for some
military activities.  In May 1996, the Secretary of Defense and the
German Defense Minister activated the German Air Force Tactical
Training Center at Holloman Air Force Base in Alamogordo.  The
training opportunities provided by the vast airspace in and around
Holloman and its proximity to Fort Bliss, Texas--the headquarters
location for German air force operations in North America--were
factors in Germany's decision to invest in a tactical training center
at the base.  State officials estimate that the training center will
result in a population increase to the Alamogordo area of about 7
percent and investment by Germany of $155 million by 1999. 

Services and trade are distinct components of New Mexico's economy. 
In 1993, the largest employment sectors in New Mexico were services,
government, and trade:  these were reported as accounting for
approximately 76 percent of the total average annual state
employment.\2 Businesses involved in trade and/or services accounted
for 67 percent of all businesses in New Mexico in 1993.  Revenue from
the gross receipts tax is the highest source of tax revenue in New
Mexico, and in 1996, gross receipt taxes from services and trade
accounted for more than half of all gross receipts tax revenue.  DOE
reports show that between 1990 and 1995, it made more expenditures in
the services and trade sectors of the New Mexico economy.  New Mexico
Department of Labor projections indicate that by 2005, the services
sector will alone account for about 41 percent of total employment
while employment in the trade sector is projected to remain stable
and government employment is expected to decline.  The projections
indicate that jobs in services and trade will account for 70 percent
of the new jobs between 1993 and 2005. 

New Mexico state officials have been focusing on "achieving economic
diversification to protect against dramatic negative changes in the
state's economy," believed to be linked to changes in federal
spending in the state.  Efforts in 1996 to recruit select industries
to the state have initially resulted in at least 7 businesses
locating to New Mexico, creating 230 new jobs.  In terms of other
efforts, New Mexico was 8th among U.S.  states in high-technology
employment growth between 1990 and 1995.\3 The single leading
high-technology industry in the state is semiconductor manufacturing,
which accounts for 34 percent of total high-technology jobs.  Intel
Corporation has three advanced computer chip manufacturing sites that
employ at least 6,500 people making it the state's second- largest
private sector employer and contributing to the growth in New
Mexico's high-technology employment.  In 1995, Intel was also the
leading manufacturing employer in the state.  High-technology exports
account for the largest percentage of New Mexico exports to other
countries, with exports to Korea leading other nations.\4 Currently,
about 10 percent of all New Mexico manufacturers are exporting.  The
leading exporters in New Mexico are Intel, Motorola, and Honeywell
Defense Avionics. 

A comparison of the percent change in New Mexico's per capita income
and total defense-related spending (DOE and DOD) in the state during
1990-94 shows that real growth occurred in per capita income, while
total defense expenditures declined (see fig.1).  A comparison
between percent real growth in New Mexico's gross state product and
total defense-related federal expenditures reveals the same pattern,
suggesting that efforts to diversify the state's economy may be
having a positive effect (see fig.  2).  Based on the average rate of
growth in the gross state product during 1987-94, the Bureau of
Economic Analysis identified New Mexico as the third-fastest-growing
state. 

   Figure 1:  Change in Per Capita
   Income and Defense-Related
   Expenditures in New Mexico
   (1990-94)

   (See figure in printed
   edition.)

Source:  Our analysis of data from the Bureau of Economic Analysis,
Census Bureau, Department of Defense, and the Department of
Energy-Albuquerque Operations Office. 

   Figure 2:  Change in New
   Mexico's Gross State Product
   and Federal Defense
   Expenditures in New Mexico
   (1990-94)

   (See figure in printed
   edition.)

Source:  Our analysis of data from the Bureau of Economic Analysis,
Census Bureau, Department of Defense, and the Department of
Energy-Albuquerque Operations Office. 


--------------------
\2 The service sector includes employment associated with hotels and
other lodging places, personal services, business services, health
services, legal services, educational services, and others.  The
trade sector includes wholesale trade of durable and nondurable goods
and retail trade at general merchandise stories, food stores,
automobile dealers and service stations, eating and drinking places,
and other retail trade.  The government sector includes federal,
state, and local government employment. 

\3 Cyberstates:  A State-By-State Overview of the High-Technology
Industry.  American Electronics Association, Washington, D.C., 1997. 
There is no conventional, standard accepted definition of
high-technology or high-technology industry.  Thus, definitions of
high-technology industries can vary greatly depending upon what
combination of products and services are selected. 

\4 The Trade Division of the New Mexico Economic Development
Department defines high-technology industry as the composite of all
exports originating from three industry sectors:  electronic and
other electric equipment and components, except computer equipment;
industrial machinery and computer equipment; and measuring analyzing
and controlling instruments, photographic, medical, and optical
goods, and watches and clocks. 


   DOE SPENDS MORE ON PROCUREMENT;
   DOD SPENDS MORE ON WORKFORCE
   AND RETIREMENT
------------------------------------------------------------ Letter :4

Available federal data provides a segmented and rough snapshot of
federal money spent in states and the employment linked to those
expenditures that is relevant to gauging some trends and patterns. 
For example, government data indicates that in 1996, the federal
government spent about $12 billion in New Mexico.  Direct
expenditures for procurement, salaries and wages for federal workers,
and grants accounted for
60 percent, or about $7.3 billion, of the total.  Direct payments to
individuals, the single largest category of federal expenditures,
accounted for approximately 37 percent, or about $4.4 billion, of
total 1996 federal expenditures (see fig.  3). 

   Figure 3:  Federal Expenditures
   in New Mexico (1996)

   (See figure in printed
   edition.)

Source:  Our analysis of Census data. 

Appendix II includes additional descriptions of federal spending and
employment in New Mexico. 


      DEFENSE-RELATED EXPENDITURES
---------------------------------------------------------- Letter :4.1

The top five agencies making procurement expenditures in New Mexico
during 1993-96, were DOE, DOD, the Department of Interior, NASA, and
the Postal Service.  The defense-related agencies (DOE and DOD),
compared to the nondefense-related ones,\5 accounted for 90 percent,
or $14.1 billion, of the $15.5 billion total spent during 1993-96.\6

Specifically, DOE accounted for 80 percent of the total federal
defense-related procurement expenditures, or about $11.2 billion of
the 1993-96 total of $14.1 billion. 

Between 1993 and 1996, the top five federal agencies that accounted
for the largest dollar amount of expenditures to pay salaries and
wages of federal workers in New Mexico were DOD; the Postal Service;
and the Departments of Interior, Health and Human Services, and
Veterans Affairs.  Salaries and wages paid to federal employees of
the defense-related agencies account for about $7 billion, or 54
percent, of the total $13 billion spent in New Mexico.  Specifically,
between 1988 and 1996 DOD accounted for about $6.5 billion, or 93
percent, of the $7 billion total defense-related federal salaries and
wages.  Payments to workers retired from defense-related agencies
also accounted for more of the total annuities to retired federal
workers living in New Mexico during 1990-96.  Payments to retired
defense-related federal workers accounted for $3.2 billion, or
68 percent, of the total $4.7 billion in annuitant expenditures. 
Payments to former DOD workers accounted for 98 percent of the total
payments to retired defense-related workers.  Figure 4 shows the
percent of defense-related expenditures for procurement, federal
workers' salary and wages, and retirement payments accounted for by
DOE and DOD, respectively. 

   Figure 4:  Defense-Related
   Expenditures Accounted for by
   DOE and DOD

   (See figure in printed
   edition.)

Note:  Procurement expenditure data is the total during 1993-96,
salary expenditure data is the total during 1988-96, and retirement
payment data is the total during 1990-96. 


--------------------
\5 Agencies included in the nondefense category are the Departments
of Agriculture, Commerce, Education, Health and Human Services,
Housing and Urban Development, Interior, Justice, Labor, State,
Transportation, Treasury, Veterans Affairs, and other nondefense
offices. 

\6 The primary time period that we report on is 1988-96.  However,
data for the complete time period was not always available. 
Therefore, in some cases we report data only for the years it was
available. 


      DEFENSE-RELATED EMPLOYMENT
---------------------------------------------------------- Letter :4.2

Between 1988 and 1996,\7 the Departments of Defense, the Interior,
Health and Human Services, Veterans Affairs, and Agriculture were the
top five agencies in terms of total federal employees in New
Mexico.\8 Between 1988-1996, defense-related jobs were about 72
percent, or 300,000 jobs, of the total 420,000 federal jobs in New
Mexico.\9 Specifically, DOD accounted for 97 percent, or about
292,000 of these jobs, over the period 1988-96.  Thus, DOD federal
jobs were more of the total federal jobs and more of the
defense-related federal jobs in New Mexico.  Federal retirees of
defense-related agencies also comprised more of the retired federal
workers living in New Mexico:  68 percent of the total between 1990
and 1996.  Specifically, DOD accounted for 99 percent of all retirees
from the defense-related agencies.  Figure 5 shows the percent of
defense-related jobs and retirees in New Mexico accounted for by DOE
and DOD. 

   Figure 5:  Defense-Related Jobs
   and Retirees Accounted for by
   DOE and DOD (1990-96)

   (See figure in printed
   edition.)

The existing data provides information on federal employees only. 
This is an important point because although the overall ratio of DOD
federal workers to DOE federal workers was 44:1 between 1988 and
1996, our research also shows that more of the DOE employment is
linked to private contractors that manage and operate the
laboratories and other DOE facilities than to the number of DOE
federal employees.  Private contractors working on government
contracts are not considered or counted as federal employees. 
However, even when we compared the total DOE employment, which
included direct DOE prime contractor, subcontractor, and federal
employees, to the total DOD federal employment DOD's direct federal
employment was higher than DOE's in each year between 1990 and 1996. 

Of the DOD employment, more of the federal jobs were DOD military
than DOD civilians.  Between 1988-96 about 42 percent of the total
DOD federal jobs in New Mexico were held by active duty military
members, 33 percent were held by inactive duty military (national
guard and reserves), and 25 percent were held by DOD civilians.\10
Similarly, more of the federal wages were associated with active duty
military.  Active duty military members accounted for 55 percent,
inactive members accounted for 5 percent, and DOD civilians accounted
for 40 percent of the total salaries and wages between 1988-96. 

   Figure 6:  DOD Military and
   Civilian Jobs and Salaries in
   New Mexico (1988-96)

   (See figure in printed
   edition.)

A comparison of the occupations represented by the defense-related
federal jobs in New Mexico indicates that during 1988-96 the largest
number of jobs were blue-collar and technical.  This finding,
however, largely represents the patterns for the DOD active duty
employment in New Mexico, for which technical and blue-collar jobs
comprise about 70 percent of the total jobs.  Among DOD civilian
employees, the two categories that accounted for the largest number
of jobs over the period 1988-96 were professional (23 percent of the
total jobs) and blue-collar (20 percent of the total jobs).\11 The
two occupational categories that account for more of the DOE direct
federal employment in New Mexico are administrative (30 percent of
total jobs) and professional (37 percent of total jobs).\12


--------------------
\7 Data for 1989 was not available.  Government employment data
includes part-time and full-time workers. 

\8 We did not include Postal Service employment data in calculating
the top five because data for the Postal Service were only available
in 1988, 1990, 1992, and 1994.  If we calculated the top five for
only the years that Postal Service data were available, it would
consistently rank second. 

\9 On a yearly basis, there were approximately 52,500 total federal
jobs and 37,500 defense-related jobs. 

\10 The available DOD data did not make distinctions between
full-time and part-time DOD federal employees in New Mexico. 
Inactive duty military members, that is the reserve and national
guard, are typically employed less than part-time in these jobs. 

\11 Occupational categorizations are based on OPM definitions.  At
New Mexico military installations, examples of occupation titles in
the blue-collar category include general aircraft worker, ammunition
repair worker, general construction worker, electrician, and missile
mechanic, among others.  Examples of occupation titles in the
technical occupational category include air traffic control, medical
care and treatment worker, operators/analysts, flight operations,
general nuclear weapons equipment repair, and auditing and
accounting, among others.  Examples of civilian occupation titles in
the professional category include general engineering, contracting,
operations research, physics, accounting, computer engineering,
social science, and aerospace engineering, among others. 

\12 Examples of occupation titles in the professional category for
DOE employees include general engineering and general physical
science, among others.  Examples of occupation titles in the
administrative category include security administration and
management program analyst, among others. 


   FEDERAL EXPENDITURE AND
   EMPLOYMENT DATA ARE INCOMPLETE
------------------------------------------------------------ Letter :5

Official federal data sources are useful for gaining a preliminary
understanding of the composition of federal expenditures in states. 
However, fundamental characteristics of the federal data make it
difficult to determine the direct economic impact of federal
activities on states.  For example, our analysis of defense-related
expenditures and employment did not include information on DOD
contractor employment because there is no official DOD or other
federal source of such information.  Federal government data sources
provide insufficient evidence for determining where federal dollars
are actually spent, how much is actually spent, and the number or
type of jobs that the federal dollars directly generate because of
numerous limitations in scope and coverage and in reporting
requirements or procedures.  Our related findings that pertain to the
data sources used and reviewed in our work are summarized in tables 1
and 2. 




                                     Table 1
                     
                       Qualities of Data on Direct Federal
                              Expenditures in States

            Type of
            direct      Tracking
            expenditur  system/
Source      e           office\a    Key limitations known or reported
----------  ----------  ----------  --------------------------------------------
Federal     DOD         DD350 to    DD350
expenditur  procuremen  DOD/DIOR/   ï¿½Dollar values are obligations, not
es by       t           WHS to      expenditures.
state                   FPDS        ï¿½Reporting threshold includes only
report                              obligations over $25,000.
(Census                             ï¿½Classified data is masked or unreported.
Bureau)                             ï¿½DOD/IG study found data integrity and
                                    reliability problems with DD350.
                                    ï¿½DD350 records differed from GAO survey-
                                    based results of total DOD contract awards
                                    to New Mexico contractors during 1988-96.
                                    ï¿½Does not include data on subcontracts.


            Non-DOD     Multiple    FPDS
            procuremen  agencies    ï¿½Most dollar values are obligations, not
            t           to FPDS     expenditures.
                                    ï¿½Reporting threshold includes only
                                    obligations over $25,000.
                                    ï¿½It is unclear whether data represent actual
                                    in-state expenditures.

            DOD grants  DD2566 to   FAADS
                        DOD/DIOR/   ï¿½FAADS dollar amounts are obligations, not
                        WHS to      actual expenditures.
                        FAADS

            Non-DOD     Multiple    FAADS
            grants      agencies    ï¿½FAADS dollar amounts are obligations, not
                        to FAADS    actual expenditures.

            DOD salary  DFAS to     DFAS:
            and wages   DOD/DIOR/   ï¿½Defense finance and accounting systems have
                        WHS         reliability problems.\b

                                    DOD/WHS/DIOR
                                    ï¿½Does not include salaries for undistributed
                                    personnel.\c
                                    ï¿½Estimates made for other personnel whose
                                    location cannot be exactly determined;
                                    unknown how many or what areas may be most
                                    impacted.
                                    ï¿½No information on nonappropriated fund
                                    staff.
                                    ï¿½Only includes federal employee data; no
                                    information on private contractor
                                    employees.


            Non-DOD     Multiple    OPM/CPDF
            salary and  agencies    ï¿½Wages for federal workers are estimated for
            wages       to OPM/     state reports.
                        CPDF        ï¿½Includes some federal workers. Not included
                                    are CIA, DIA, NSA, judicial and legislative
                                    branches, TVA, and Postal Service. FBI does
                                    not report duty stations for employees.
                                    ï¿½Part-time and seasonal worker salaries are
                                    annualized.
                                    ï¿½Wages for workers who get piece-rate,
                                    hourly rate, or other nonsalary rates
                                    include only unit-rate compensation, not
                                    total compensation.


Office of   Retired     Multiple    DMDC
the         military    services    ï¿½Obtained different figures from Actuary and
Actuary     pay         to DMDC     DMDC for New Mexico sample. Data from
(DOD)                               Actuary and DMDC were different from a third
                                    DOD source, raising questions about
                                    reliability of information.

Annuities   Retired     Multiple    None reported.
roll (OPM)  DOD         agencies
            civilian
            pay and
            all other
            non-DOD
            retired
            pay
--------------------------------------------------------------------------------
Legend:
CIA--Central Intelligence Agency
DD350--Department of Defense [Form] 350
DFAS--Defense Finance and Accounting Service
DIA--Defense Intelligence Agency
DIOR/WHS--Directorate for Information Operations and Reports,
Washington
Headquarters Service
DMDC--Defense Manpower Data Center
DOD/IG-Department of Defense, Inspector General
FAADS--Federal Assistance Awards Data System
FBI--Federal Bureau Of Investigation
NSA-- National Security Agency
TVA--Tennessee Valley Authority

\a This column identifies the database or office that provides the
original data, which is "fed" to the source listed in the first
column.  For example, DOD's procurement expenditures reported in the
Federal Expenditures by State report are generated through the
tracking system that starts with the collection of information on the
DD350, which is centrally reported through DOD's Directorate for
Information Operations and Reports, Washington Headquarters Services,
to the Federal Procurement Data System and then finally reported to
the Census Bureau.  The DD350 is the official DOD form used to record
individual DOD contract actions (obligations or deobligations) in
excess of $25,000.  Contract offices complete DD350s according to
instructions detailed in the Defense Federal Acquisition Regulations
supplement.  The dollar value of contracts is recorded along with
information on the name and location of contractors, the location
where the work will be performed, products or services purchased,
contract solicitation procedures used (competitive or other than
competitive), and the type of business contracted with (large, small,
small disadvantaged), among other things. 

\b We previously reported on these problems in our report entitled
DOD Infrastructure:  DOD's Planned Finance and Accounting Structure
Is Not Well Justified (GAO/NSIAD-95-127, Sept.  1995). 

\c DOD's definition of undistributed personnel includes military
personnel in a transient status, for example, personnel on travel,
proceed, leave en route or in a temporary duty status while on
permanent change of station (PCS) orders, or for which an exact
location is classified.  In 1996, undistributed personnel were 2.6
percent of all military personnel. 




                                     Table 2
                     
                       Qualities of Data on Direct Federal
                               Employment in States

            Type of     Tracking
            direct      system/
Source      employment  office      Known limitations
----------  ----------  ----------  --------------------------------------------
Atlas/      Number of   DFAS/       DFAS:
data        DOD         other DOD   ï¿½Defense finance and accounting systems have
abstract    active,     offices to  reliability problems.\a
for the     inactive,   DOD/DIOR/
U.S. and    and         WHS         DOD/WHS/DIOR
selected    civilian                ï¿½Does not include undistributed personnel.
areas       employees               ï¿½Estimates conducted for other personnel
(DOD)                               whose location cannot be exactly determined;
                                    unknown how many or what areas may be most
                                    impacted.
                                    ï¿½No information on nonappropriated fund
                                    staff.
                                    ï¿½Only includes federal employee data; no
                                    information on private contractor
                                    employees.
                                    ï¿½No program/system-level contractor
                                    employment.


Office of   Number of   Multiple    DMDC
the         retired     services/   ï¿½Obtained different figures for total
Actuary     military    agencies    retirees from Actuary and DMDC for New
(DOD)       personnel   to DMDC     Mexico sample, raising questions about
            receiving               reliability of information.
            pay

Annuities   Number of   Multiple    None reported.
Roll (OPM)  retired     agencies
            DOD
            civilians
            and all
            other non-
            DOD
            retired
            personnel
            receiving
            pay

OPM/CPDF    Number of   Multiple    OPM/CPDF
            all non-    agencies    ï¿½Includes only the number of federal
            DOD                     employees; no data on private contractor
            civilian                employees.
            federal                 ï¿½Includes only some federal workers. Does
            employees               not include CIA, DIA, NSA, judicial and
                                    legislative branches, TVA, and Postal
                                    Service. FBI does not report duty stations
                                    for employees.


Office of   Number of   Multiple    Office of Worker and Community Transition
Worker and  DOE prime   DOE field   ï¿½Employee counts are not provided by
Community   contractor  offices or  specific location but by major employer, so
Transition  employees   facilities  actual number at specific locations must be
(DOE)                               assumed.
                                    ï¿½Some contractor employment is "work for
                                    others" and "other work," not part of funded
                                    DOE major program activities.

--------------------------------------------------------------------------------
\a We previously reported on these problems in our report entitled
DOD Infrastructure:  DOD's Planned Finance and Accounting Structure
Is Not Well Justified (GAO/NSIAD-95-127, Sept.  1995). 

To gain further insights into the reliability of the federal
government's data we focused on characteristics of existing DOD data. 
Although DOD's procurement expenditure data (DD350) is used in broad
policy contexts and used to evaluate the status of programs that are
believed to be important to economic security, the form is not
designed to provide information on all DOD expenditures in a single
state or at the national level.  Procurement contracts under $25,000
are not included, no information on DOD subcontracts of any value are
included, and financial data related to classified programs may or
may not be reported or be accurate. 

DOD acknowledges that the DD350 does not completely account for all
procurement expenditures, and although this limitation is generally
understood and acknowledged by informed users, the possible
implications are not.  We surveyed the top five DOD contractors in
New Mexico to determine how much money they received in DOD prime
contracts and subcontracts and compared their responses to DOD's
records (the DD350 data) of their total contracts.\13 The comparisons
revealed that in no case were the DOD records of the dollar value of
contracts awarded to these companies the same as the contractors'
records.  Differences between DOD and contractors' records ranged
from $20 million for prime contracts to $80 million for total
contracts.  In some cases, the DOD records appeared to overstate the
amount the contractors received, while in other cases the DOD records
appeared to understate the amount. 

Our research suggests several possible reasons for the
inconsistencies between contractor records and DOD records.  For
example, expenditures associated with procurement contracts can leak
from a state's economy if a company subcontracts part of the work
elsewhere.  One study reported that of $5.2 billion in DOD prime
contracts received by McDonnell Douglas in St.  Louis, Missouri, less
than 3 percent, or $156 million, stayed in Missouri due to
out-of-state subcontracting.\14 However, from our survey of
contractors in New Mexico we determined that leakages were more
prevalent for certain types of procurement contracts.  While our
survey showed overall that more than 80 percent of the total DOD
prime contract dollars remained in the state, for every year between
1988 and 1996, it also showed that the businesses that predominantly
received service contracts, rather than supply and equipment
contracts (i.e., major hard goods/weapons), kept nearly all of the
DOD contract money they received in the state.  This is particularly
relevant because other DOD data indicate that in every year between
1988 and 1996, DOD procurement contracts for services account for the
largest dollar volume of contracts to New Mexico.\15

Also, service contracts may more likely be under DOD's $25,000
reporting threshold\16 and therefore excluded from total expenditures
as officially reported by DOD.  Furthermore, injections of dollars
from subcontracts with out-of-state firms or with other in-state
firms are not tracked by DOD, yet would have been included in the
contractors' records. 

Finally, the DOD Inspector General reported in 1989 that the DD350
data had reliability problems due to instances of unreported contract
obligations and other errors in reported data.  The Inspector General
made no recommendations and has not assessed the reliability and
validity of the DD350 contract tracking system since then. 

The existing data that track defense-related employment are limited
in their scope, coverage, and reliability.  Among the most notable
limitation in the data is the lack of a central or official source of
data on private-sector employment associated with DOD contracts. 
Information on the number of jobs associated with particular defense
contracts or weapon programs are repeatedly discussed in the media
and in Congress.  Further, DOD has stated that defense procurement
dollars promote the creation of jobs.  However, DOD officials have
also indicated that they do not collect information on the job
impacts of particular DOD budget decisions. 

To obtain information on the employment associated with defense
contracts or the employment linked to particular defense programs, it
is necessary to contact individual defense contractors and/or DOD
system program offices directly.  The contractor employment data we
obtained from our survey of defense contractors in New Mexico is
summarized in appendix III, along with other survey findings.  The
responses from the top four contractors who provided us data
indicated that the total number of direct jobs associated with DOD
contracts was approximately 19,200 during 1988-96.\17 The total DOD
federal employment (active duty, inactive, and civilians) in the
state for the same period (1989 data included) was approximately
328,000.\18 A comparison of employment data from three top DOE prime
contractors to the data from the top four DOD prime contractors
indicates that, over the period 1994-96, DOE had about eight prime
contractor employees to every one DOD prime contractor employee in
New Mexico.\19 We also obtained employment and expenditure data for a
sample of specific defense programs that were known to have some
involvement with New Mexico contractors (see table 3).\20




                                Table 3
                
                 New Mexico Expenditures and Employment
                    for a Sample of Defense Programs

                                                  FY 96
                             Total FY 96   expenditures  FY 96
                                 program         in New  employment in
Program                        funding\a       Mexico\b  New Mexico\c
-------------------------  -------------  -------------  -------------
Airborne laser (ABL) (Air    $19,000,000     $4,409,000  24
 Force research and
 development)
C-17 (Air Force            $2,270,000,00     $8,818,000  28
 procurement)                          0
Echelon Above Corps (EAC)    $44,500,000     $5,283,825  63
 Communication Program
 (Army procurement)
Kiowa Warrior helicopter     $64,300,000    $20,000,000  115
 (Army procurement)
Sense and destroy armor      $41,100,000       $383,542  3
 (SADARM) submunition
 (Army procurement)
Stinger missile (Army      $16,900,000\d     $157,876\e  Not available
 procurement)
TOW2B missile (Army          $12,000,000   $5,500,000\f  20
 procurement)
----------------------------------------------------------------------
\a Source of funding for the airborne laser is DOD Comptroller 1997
R-1 budget tables.  For all procurement programs, the 1997 P-1 is the
source. 

\b Data was provided by the DOD system program offices and/or the
prime contractors for the systems. 

\c Data was provided by the DOD system program offices and/or the
prime contractors for the systems. 

\d This is the amount of the fiscal year 1997 Stinger program
funding. 

\e This is the amount of the fiscal year 1997 Stinger-related
expenditures made in New Mexico. 

\f Estimate. 


--------------------
\13 In order, the top five contractors in New Mexico in 1996 were
Lockheed Martin, Honeywell Defense Avionics, DynCorp, EG&G, and Kit
Pack Company.  We obtained complete survey data from four of the top
contractors.  Lockheed Martin was unable to provide reliable
information on the value of DOD contracts they received for work
performed in New Mexico.  See appendix III for a complete discussion
of the contractor survey methods and results. 

\14 R.  Atkinson, "Defense Spending Cuts and Regional Economic
Impact:  An Overview.  " Economic Geography, 69(2), (1993), pp. 
107-122. 

\15 DOD categories of service contracts include the operation of
government-owned facilities; automatic data processing and
telecommunications services; architect and engineering services;
social services; medical services; and lease or rental of facilities,
among others.  Other broad categories of DOD contracts include supply
and equipment contracts (e.g., weapons) and research, development,
testing, and evaluation contracts. 

\16 J.R.  Crump, "Sectoral Composition and Spatial Distribution of
Department of Defense Services Procurement," Professional Geographer,
45(2), (1993), pp.  286-296. 

\17 This is approximately 2,100 defense contractor employees on a
yearly basis.  In its 1996 review of the New Mexico economy, Sunwest
Bank lists New Mexico's 132 largest employers.  Of the top five New
Mexico defense contractors we contacted for our survey, only
Honeywell appears on the list with a ranking as the state's 45th
largest employer. 

\18 This is approximately 36,000 DOD federal employees on a yearly
basis. 

\19 This includes DOE prime contractor employment at the Los Alamos
and Sandia National laboratories and at the Waste Isolation Pilot
Plant in Carlsbad. 

\20 DOD's public affairs office publishes summaries of each contract
it awards that exceed $5 million.  These DOD summaries, referred to
as Blue Tops, are distributed to Congress, the public, the media and
industry.  The contract summaries include information on who the
contract was awarded to, the total amount of the contract, what the
contract is for, and where the work will be done.  These summaries
can be useful for current event information; however, the information
reported can also be misleading if one is attempting to determine the
DOD contract dollars directed to specific states.  For example, a
summary from September 1996 indicated that the Navajo Refining
Company in Artesia, New Mexico was awarded an $86-million contract. 
Although the summary indicated that the work would be performed in
Roswell, New Mexico and El Paso, Texas, data we obtained from other
DOD offices indicated that $36,000 of the work was done in New
Mexico, whereas $85 million was done in Texas.




   CONCLUSIONS
------------------------------------------------------------ Letter :6

The available data indicate that the state of New Mexico receives
relatively large amounts of federal dollars.  Defense-related federal
activities in the state have contributed to the development of the
economy, and recent efforts to diversify the economic base appear
linked to continued growth.  The best available data indicate that in
New Mexico DOE and DOD account for about 90 percent of all federal
procurement spending (1993-96), 54 percent of expenditures for
federal worker salary and wages (1988-96), 72 percent of all federal
jobs in the state (1988-96), and 68 percent of all retired federal
workers living in the state (1990-96).  Specifically, DOE accounts
for 80 percent of the defense-related procurement expenditures, and
DOD accounts for 93 percent of the defense-related salary and wage
expenditures, 97 percent of the defense-related federal jobs, and 99
percent of the federal workers retired from defense-related agencies
and living in New Mexico.  The largest component of DOE employment is
private contractor employment, while the largest component of DOD
employment is federal employment, namely active duty military
members. 

On one hand, determining the full and complete economic magnitude of
federal expenditures in states, whether defense or nondefense, and
the related employment is not possible with existing data.  Trying to
reconcile differences among data sources and account for gaps or
questionable data is very resource-intensive and does not necessarily
yield benefits in precision or accuracy.  On the other hand, the
existing data are not without value, nor should the government
necessarily strive for increased data collection that could actually
entail more costs than benefits.  The limitations in federal data
may, in part, reflect the fact that data collection trails behind
changes in federal policy or shifts in policy relevance.  Those who
rely on federal data need to be alert to their drawbacks and exercise
discretion when using them. 


   AGENCY COMMENTS
------------------------------------------------------------ Letter :7

In oral comments on a draft of this report, DOD concurred with our
findings and conclusions.  It also provided several technical
comments, which we incorporated in the text where appropriate. 


   SCOPE AND METHODOLOGY
------------------------------------------------------------ Letter :8

In conducting our work, we contacted and interviewed officials and
experts from federal and state government offices and the private
sector.  Because the scope of the work covered all federal
expenditures and related employment in New Mexico over an 8-year
period, there was a large range and number of contacts and outreach
efforts we made in completing our work.  We made over 50 contacts
throughout federal and state governments and the private sector.  Our
final results were produced from databases from four separate federal
agencies; our survey of New Mexico defense contractors encompassing 8
years of financial and business information; information obtained
from a review of more than
30 publications; and information we obtained from numerous documented
interviews with key officials.  A list of the offices we contacted is
in appendix I. 

To determine the characteristics of the New Mexico economy and recent
changes in the economy, we reviewed and analyzed economic data and
information we obtained from interviews with New Mexico state
officials, federal government officials, and available federal and
state data sources, including the Bureau of Economic Analysis and the
Bureau of Business and Economic Research at the University of New
Mexico. 

To determine the direct defense-related and nondefense-related
federal expenditures and employment in New Mexico over the period
1988-1996\21 we contacted multiple federal offices and obtained
official data from DOD and DOE.  We obtained data on all other
nondefense-related federal expenditures from the Census Bureau.  All
available data on DOD and DOE expenditures were categorized as
defense-related.\22 We obtained total nondefense-related employment
data from OPM's Central Personnel Data File.  All expenditure figures
were adjusted for inflation and are presented in constant 1996
dollars.  Appendix II contains the complete overview and figures
depicting our findings related to direct federal expenditures and
employment in New Mexico. 

To determine the extent to which available government data provides
reliable information on defense spending and employment, we evaluated
the qualities of the existing federal data.  We reviewed technical
documentation for the sources used, interviewed agency officials
about the data sources, conducted crosschecks of data that appeared
in multiple sources but had been derived from the same source, and in
the case of DOD procurement expenditures, compared the results of DOD
data to our survey results.  Survey results are discussed in appendix
III.  Given the outcome of our review, federal data limitations and
data reliability concerns are discussed in our findings and reflected
in the report's conclusions. 

Our work was conducted between November 1996 and October 1997 in
accordance with generally accepted government standards. 


--------------------
\21 We did not report on indirect or induced effects of federal
expenditures.  Direct expenditures include, for example, the actual
amount of a contract awarded to a business, and direct employment
includes the actual number of jobs created by that contract.  Direct
expenditures and employment produce some indirect and induced
effects.  Indirect expenditures include, for example, purchases of
supplies or services that support the contract or the initial direct
expenditure.  Indirect employment includes the number of jobs that
are created from the indirect expenditures.  Induced expenditures and
employment include the effects created by spending from wages earned
on retail purchases, housing, and restaurants, among others. 

\22 Some DOD expenditures may be for civil functions not directly
related to national defense.  DOE programs include defense,
environmental management, and energy research, among others.  The
predominant mission of the national laboratories in New Mexico focus
on nuclear weapons research and engineering.  Since the predominant
mission of the DOD programs and the DOE programs in New Mexico are
defense-related we categorized all the expenditures and employment as
defense-related. 


---------------------------------------------------------- Letter :8.1

As agreed with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 14 days
from its issue date.  At that time, we will send copies of this
report to other interested congressional committees and members. 
Copies will also be made available to others upon request. 

Please contact me at (202) 512-3092 if you or your staff have any
questions concerning this report.  Major contributors to this report
were Carolyn Copper, John Oppenheim, and David Bernet. 

Sincerely yours,

Kwai-Cheung Chan
Director, Special Studies and Evaluations


OFFICES WE CONTACTED
=========================================================== Appendix I


   DEPARTMENT OF DEFENSE,
   WASHINGTON, D.C. 
--------------------------------------------------------- Appendix I:1

Directorate for Information Operations and Reports, Washington
 Headquarters Services
Office of Program Analysis and Evaluation
Office of the Comptroller
Office of Economic Adjustment
Defense Manpower Data Center
Office of the Actuary
Secretary of the Air Force
Industrial Affairs and Installations
Office of Nonappropriated Fund Personnel
Defense Finance and Accounting Service
Public Affairs
Inspector General


   DEPARTMENT OF ENERGY,
   WASHINGTON, D.C. 
--------------------------------------------------------- Appendix I:2

Office of Worker and Community Transition
Office of Human Resources and Administration
Office of the Budget


   DEPARTMENT OF COMMERCE,
   WASHINGTON, D.C. 
--------------------------------------------------------- Appendix I:3

Bureau of the Census
Bureau of Economic Analysis


   CONGRESSIONAL AGENCIES,
   WASHINGTON, D.C. 
--------------------------------------------------------- Appendix I:4

Congressional Budget Office
Congressional Research Service


   FEDERAL FACILITIES, NEW MEXICO
--------------------------------------------------------- Appendix I:5

Cannon Air Force Base, Clovis, New Mexico
Holloman Air Force Base, Alamogordo, New Mexico
Kirtland Air Force Base, Albuquerque, New Mexico
White Sands Missile Range, White Sands, New Mexico
Albuquerque Operations Office, Department of Energy,
 Albuquerque, New Mexico
Bureau of Land Management, Department of the Interior, Las Cruces and
 Albuquerque, New Mexico


   STATE GOVERNMENT OFFICES
--------------------------------------------------------- Appendix I:6

Santa Fe, New Mexico
Department of Economic Development
Department of Taxation and Revenue
Albuquerque, New Mexico
Department of Labor
Richmond, Virginia
Economic Development Partnership, Department of Business Assistance
Annapolis, Maryland
Federal Response and Technology Commercialization


   UNIVERSITIES
--------------------------------------------------------- Appendix I:7

Bureau of Business and Economic Research, University of New Mexico,
 Albuquerque, New Mexico
Department of Economics, University of Maryland, College Park,
Maryland


   DEFENSE CONTRACTORS
--------------------------------------------------------- Appendix I:8

Honeywell Defense Avionics, Albuquerque, New Mexico
Laguna Industries, Laguna, New Mexico
McDonnell Douglas, Long Beach, California
Aeroparts Manufacturing and Repair, Rio Rancho, New Mexico
SBS Technologies, Albuquerque, New Mexico
Physics, Mathematics, and Computers, Socorro, New Mexico
Kit Pack Company, Las Cruces, New Mexico
EG&G Management Systems, Albuquerque, New Mexico
Lockheed Martin, Bethesda, Maryland
DynCorp, Reston, Virginia


   OTHERS
--------------------------------------------------------- Appendix I:9

Professional Aerospace Contractors Association of New Mexico,
 Albuquerque, New Mexico
Intel Corporation, Albuquerque, New Mexico
American Electronics Association, Santa Clara, California
Logistics Management Institute, McClean, Virginia
Academy for State and Local Governments, Washington, D.C.
National Council of State Governments, Washington, D.C.
National Legislative Council, Washington, D.C.
National Governors Association,Washington, D.C.
RAND, Washington, D.C. 


DIRECT FEDERAL EXPENDITURES AND
EMPLOYMENT IN NEW MEXICO
========================================================== Appendix II

This appendix presents 1988-96 (1) trends in total direct federal
expenditures and employment in New Mexico and within specific
spending categories, (2) defense-related and nondefense-related
expenditures and employment, and (3) the Department of Energy's (DOE)
and the Department of Defense's (DOD) share of the defense-related
expenditures and employment. 

We used existing databases and a survey on how much money is directly
spent and how many people are directly employed to determine
expenditures and employment.  We did not assess the indirect or
induced effects of federal expenditures and employment.  All
expenditure data were adjusted for inflation and are presented in
constant 1996 dollars.  Data for all years were not always available. 


   FEDERAL EXPENDITURES IN NEW
   MEXICO
-------------------------------------------------------- Appendix II:1

Federal expenditures in New Mexico fluctuated between about $10
billion and $12 billion, 1988 through 1996.  The highest level of
spending occurred in 1996 (see fig.  II.1). 

   Figure II.1:  Federal
   Expenditures in New Mexico
   (1988-96)

   (See figure in printed
   edition.)

Source:  Our analysis of Census data. 

This increase in federal expenditures for New Mexico is consistent
with nationwide trends.\1

Total federal employment in New Mexico generally increased between
1988 and 1994, then declined to 1996.  Total employment in 1996 is
the lowest level of any year in the period (see fig.  II.2).  The
decline in federal employment in New Mexico in the last several years
is consistent with trends in declining nationwide federal employment. 

   Figure II.2:  New Mexico
   Federal Employment (1988-96)

   (See figure in printed
   edition.)

Source:  Our analysis of OPM and DOD data.  Postal Service employment
is not included. 

Figure II.3 shows the specific expenditure trends in procurement,
grants, salaries and wages for federal workers, and direct payments
to individuals. 

   Figure II.3:  Total Federal
   Spending on Procurement,
   Grants, Federal Employee
   Salaries and Wages, and Direct
   Payments in New Mexico
   (1988-96)

   (See figure in printed
   edition.)

Procurement expenditures in New Mexico have generally declined over
time but did increase between 1989 and 1992.  In the 1988-96 time
frame, procurement expenditures were at their lowest in 1996. 
Expenditures on grants and direct payments have increased over time
and have not shown periods of decline.  This is consistent with
national trends.  Federal salary and wage trends are marked by small
increases over time with periods of stability following an increase. 


--------------------
\1 Federal Fiscal Trends:  Fiscal Years 1971-1995 (GAO/AIMD-97-3,
Nov.  1996). 


   DEFENSE-RELATED AND
   NONDEFENSE-RELATED FEDERAL
   EXPENDITURES IN NEW MEXICO
-------------------------------------------------------- Appendix II:2

Defense-related procurement expenditures far exceeded
nondefense-related procurement expenditures during 1993-96.\2 But
both types of expenditures have been declining (see fig.  II.4).  The
decline in defense-related expenditures is consistent with overall
trends in declining DOD and DOE budgets.\3

   Figure II.4:  Defense-Related
   and Nondefense-Related Federal
   Procurement Expenditures in New
   Mexico (1993-96)

   (See figure in printed
   edition.)

Nondefense-related agencies accounted for more of the expenditures
for federal grants to New Mexico (see fig.  II.5).  The top five
agencies in terms of expenditures on federal grants to New Mexico
were the Departments of Health and Human Services (HHS),
Transportation, Interior, Agriculture, and Education.  Expenditures
on nondefense-related grants were 99 percent of the total grant
expenditures in each year between 1988 and 1996. 

   Figure II.5:  Defense-Related
   and Nondefense-Related Federal
   Grant Expenditures in New
   Mexico (1988-96)

   (See figure in printed
   edition.)

Defense-related agencies accounted for more of the total salaries and
wages for federal workers than nondefense-related agencies between
1988 and 1996 (see fig.  II.6). 

   Figure II.6:  Salaries and
   Wages to Defense-Related and
   Nondefense-Related Federal
   Workers in New Mexico (1988-96)

   (See figure in printed
   edition.)

Between 1988 and 1993 total expenditures on salaries and wages for
nondefense-related workers increased steadily, slowly declining in
the last 4 years.  On the other hand, salary and wage expenditures
for defense-related workers generally declined between 1988 and 1993
but increased slightly between 1995 and 1996.  Salaries and wages
were at their highest in 1996 for defense-related workers were and at
their highest in 1993 for nondefense-related federal workers. 

It is not possible to make clear federal agency distinctions in
direct payment expenditures.  These expenditures are commonly
reported by federal program, not by federal agency.  Given the
reporting criterion used, we determined which federal program
accounted for most of the direct payments in New Mexico.  In some but
not all cases, this information is sufficient to determine which
federal agency accounted for most of the expenditures. 

Programs administered by HHS accounted for over 50 percent of the
total direct payment expenditures in New Mexico in each year between
1988 and 1996:  the average was 63 percent (see fig.  II.7).  The
programs included in the HHS roll-up include Social Security,
Medicare, and Supplemental Security Income. 

   Figure II.7:  Distribution of
   Federal Direct Payments in New
   Mexico, by Federal Program
   (1988-96)

   (See figure in printed
   edition.)

Note:  Totals may not always add to 100 percent due to rounding. 

Payments for federal retirement and disability made up the second
largest category of direct payments in New Mexico in each year
between 1988 and 1996.  On average, these payments accounted for 18
percent of all direct payments made in New Mexico during 1988-96. 
The Food Stamp Program, administered by the Department of
Agriculture, on average, accounted for 5 percent, and direct payments
to individuals associated with all other programs, on average,
accounted for 14 percent of the total direct payments over the same
time period.\4

We could not determine the breakdown between the defense-related and
nondefense-related distribution of federal retirement payments
directly from the Census data.  Therefore, we obtained additional
data from DOD and the Office of Personnel Management (OPM).  Figure
II.8 shows that payments to workers retired from the defense-related
agencies account for the majority--on average 68 percent--of the
total annuities for retired federal workers in New Mexico, between
1988 and 1996.  Total annuities for defense and nondefense-related
retired federal workers have increased over time. 

   Figure II.8:  Total Annuities
   for Federal Workers Living in
   New Mexico and Retired From
   Defense-Related and
   Nondefense-Related Agencies
   (1988-96)

   (See figure in printed
   edition.)


--------------------
\2 Procurement expenditures identified by all specific federal
agencies were not reported by Census until 1993. 

\3 See Defense Industry:  Trends in DOD Spending, Industrial
Productivity and Competition (GAO/PEMD-97-3), Jan.  1997) and
Department of Energy:  Funding and Workforce Reduced, but Spending
Remains Stable (GAO/RCED-97-96, Apr.  1997). 

\4 All other programs include unemployment compensation, veteran's
benefit programs, payments to railroad workers, housing assistance
programs, Pell Grants, earned income tax credits, National Guaranteed
Student Loan interest subsidies, federal workers' compensation, black
lung disease payments, and other programs. 


   DEFENSE-RELATED AND
   NONDEFENSE-RELATED FEDERAL
   EMPLOYMENT IN NEW MEXICO
-------------------------------------------------------- Appendix II:3

Federal workers from the defense-related agencies accounted for the
majority of the total federal employment in New Mexico during 1988-96
(see fig.  II.9).  Federal jobs in the defense-related agencies, on
average, accounted for 72 percent of the total federal jobs in New
Mexico.  Total federal employment declined by approximately 4,000
jobs between 1992 and 1996; about 84 percent of these jobs were in
defense-related agencies. 

   Figure II.9:  Defense-Related
   and Nondefense-Related Federal
   Employment in New Mexico
   (1988-96)

   (See figure in printed
   edition.)

Source:  Our analysis of DOD and OPM data.  Postal Service employment
not included.  OPM sources do not include 1989 data. 

Defense-related agencies in New Mexico account for about 68 percent
of the federal retirees, on average, between 1990 and 1996.  The
number of federal workers retired from defense and nondefense-related
agencies and living in New Mexico has increased over time. 

   Figure II.10:  Federal Retired
   Workers from Defense and
   Nondefense-Related Agencies
   Living in New Mexico (1988-96)

   (See figure in printed
   edition.)


   DOD AND DOE SHARE OF
   DEFENSE-RELATED EXPENDITURES IN
   NEW MEXICO
-------------------------------------------------------- Appendix II:4

The defense-related agencies in New Mexico accounted for the majority
of procurement expenditures, total annuities for retired federal
workers, and salaries and wages for federal employees.  In figures
II.11, II.12, and II.14, we show the trends in the DOD and DOE share
of the expenditures in each of these categories.  We also show the
number of DOD and DOE federal retirees in New Mexico (see fig. 
II.13). 

Between 1993 and 1996, DOE accounted for more of the defense
procurement dollars that went to New Mexico than DOD (see fig. 
II.11).  Consistent with overall declining DOE and DOD budgets, DOE
and DOD procurement expenditures in New Mexico have declined in the
last several years. 

   Figure II.11:  DOD and DOE
   Procurement Expenditures in New
   Mexico (1993-96)

   (See figure in printed
   edition.)

Figure II.12 shows that payments to DOD retired federal workers
living in New Mexico account for most of the total annuities to
federal workers retired from defense-related agencies between 1990
and 1996.  On average, annuities to retired DOD workers accounted for
98 percent of total annuities between 1990 and 1996. 

   Figure II.12:  Annuities to
   Workers Retired From DOD and
   DOE and Living in New Mexico
   (1990-96)

   (See figure in printed
   edition.)

Also, more former DOD than DOE federal employees were living in New
Mexico between 1990 and 1996 (see fig.  II.13). 

   Figure II.13:  DOD and DOE
   Retired Federal Workers in New
   Mexico (1990-96)

   (See figure in printed
   edition.)

The increase in retired DOD workers in New Mexico is consistent with
an overall increase in the number of retired active duty military
members and DOD civilians. 

Figure II.14 shows that DOD also accounts for nearly all of the
salary and wage expenditures for federal employees of defense-related
agencies. 

   Figure II.14:  DOD and DOE
   Federal Employee Salary and
   Wage Expenditures in New Mexico
   (1988-96)

   (See figure in printed
   edition.)

Source:  Our analysis of Census and OPM data. 

On average, DOD accounted for 93 percent of the defense-related
salaries and wages for federal employees.  The total amount of DOD
and DOE salary and wage expenditures has fluctuated some over the
years, but no sharp increases or decreases have occurred. 

DOE mostly employs prime contractor employees, who are not counted as
federal employees, thus, their numbers are not included in federal
data.  DOE data we obtained indicates that the salaries and wages for
DOE prime contractor employees in New Mexico are greater than those
of DOD federal employees in the state.  For example, between 1990 and
1994 the total salaries and wages for DOD federal employees were
about $4 billion and, for DOE prime contractors were about $6
billion.\5 Comparable figures on the total compensation to DOD prime
contractor employees in New Mexico were not available.  However, the
data we obtained from our survey of the top New Mexico contractors
shows that the total compensation to their employees was $332 million
between 1990 and 1994, or about $6.6 million per year. 


--------------------
\5 On a yearly basis, this is approximately $800 million in DOD
federal employee salaries and wages and $1.2 billion in DOE prime
contractor salaries. 


   DOD AND DOE SHARE OF
   DEFENSE-RELATED EMPLOYMENT IN
   NEW MEXICO
-------------------------------------------------------- Appendix II:5

Defense-related federal employment in New Mexico is higher than
nondefense-related employment.  In this section, we show the DOD and
DOE portions of defense-related employment over time, including DOD's
and DOE's numbers and types of occupations. 

On average, DOD accounted for 97 percent of the total defense-related
federal employment in New Mexico between 1988 and 1996 (see fig. 
II.15). 

   Figure II.15:  DOE and DOD
   Employment in New Mexico
   (1988-96)

   (See figure in printed
   edition.)

Source:  Our analysis of DOD and OPM data. 

In each year between 1988 and 1996, active duty military members were
the single largest group of DOD federal employees in New Mexico. 
Inactive duty military and DOD civilian employees, respectively,
accounted for the second and third largest component of DOD federal
employment (see
fig.  II.16). 

   Figure II.16:  DOD Active,
   Inactive, and Civilian
   Employment in New Mexico
   (1988-96)

   (See figure in printed
   edition.)

Active duty and inactive duty military members, and DOD civilians
ranked first, third, and second, respectively, in accounting for the
largest share of salary and wages for DOD federal employees in New
Mexico from 1988 to 1996 (see fig.  II.17). 

   Figure II.17:  Salary and Wages
   for DOD Active and Inactive
   Duty Members and DOD Civilians
   in New Mexico (1988-96)

   (See figure in printed
   edition.)

Between 1988 and 1996 more of the DOD active duty military jobs in
New Mexico were blue collar and technical compared to administrative,
clerical, white collar, or professional job occupations (see fig. 
II.18). 

   Figure II.18:  Job Occupations
   of DOD Active Duty Military in
   New Mexico (1988-96)

   (See figure in printed
   edition.)

Note:  Reserves and National Guard not included. 

Source:  Defense Manpower Data Center. 

The job occupations of DOD civilians were more evenly dispersed
across categories than DOD military jobs.  Professional job
occupations accounted for the most DOD civilian jobs in New Mexico
between 1988 and 1996 (see fig.  II.19). 

   Figure II.19:  Job Occupations
   of DOD Civilians in New Mexico
   (1988-96)

   (See figure in printed
   edition.)

Source:  Defense Manpower Data Center. 

The majority of DOE federal jobs in New Mexico between 1988 and 1996
were professional and administrative (see fig.  II.20). 

   Figure II.20:  Job Occupations
   of DOE Federal Employees in New
   Mexico (1988-96)

   (See figure in printed
   edition.)


SURVEY OF TOP DEFENSE CONTRACTORS
IN NEW MEXICO
========================================================= Appendix III

The principal purpose of our survey was to determine and characterize
the flow of defense dollars to contractors and to illuminate and
quantify the limitations of existing data sources that document
defense spending in states. 


   SURVEY METHODS
------------------------------------------------------- Appendix III:1

For our survey sample, we selected contractors who were among the top
five in terms of the total dollar amount of DOD prime contracts
awarded in fiscal year 1996.  Time and resource constraints prevented
us from surveying every business that was awarded a defense contract
and performed work in New Mexico.  For example, in 1996 alone, 471
businesses were awarded DOD contracts exceeding $25,000 for work
principally done in New Mexico. 

We obtained DOD's DD350 data to determine the total value of DOD
prime contracts awarded to all businesses in 1996 with the principal
place of work in New Mexico.  From this population we selected five
contractors:  Honeywell, DynCorp, EG&G, Kit Pack Company, and
Lockheed Martin.  In 1996, prime contracts to these businesses
accounted for 26 percent of the total value of all DOD prime
contracts awarded to businesses in New Mexico.  In the period covered
by our survey, that is, 1988-96, the percentage of total DOD prime
contract awards accounted for by the top five New Mexico contractors
ranged from 26 to 46 percent.  Different companies have been in the
list of the top five over the years.  However, over the survey
period, Honeywell and DynCorp were consistently among the top five. 

Contractors were asked to complete several questions about DOD
contracts they were awarded as a prime and subcontractor between
1988-96.  We asked them to indicate the total value of all DOD
contracts received, the dollar amount of contract work that was
subcontracted or was interdivisional work, the amounts subcontracted
in-state and out-of-state, the amount of salary and wages for all
contracts completed by the contractor and by subcontractors, and the
number of full-time equivalent (FTE) positions for work completed by
the contractor and for subcontractors. 


   CONTRACTOR BACKGROUND
------------------------------------------------------- Appendix III:2

As a group Honeywell, Lockheed Martin, DynCorp, and EG&G are large,
diversified corporations with business establishments physically
located in New Mexico but actual corporate headquarters located
elsewhere in the country.  Kit Pack is a relatively smaller company,
with its business headquarters and all operations located in New
Mexico. 

During the period of time covered by our survey, Honeywell's
principal DOD work in New Mexico was research, development, and
testing and evaluation services for military aircraft and the
manufacturing of aircraft avionics components.  In 1996, DOD awarded
prime contracts to Honeywell to provide automatic pilot mechanisms;
flight instruments; and research, development, and testing and
evaluation services related to aircraft engine manufacturing, among
other things.  Its survey data was completed by staff at Honeywell's
business establishment in Albuquerque. 

DynCorp is a large professional and technical services firm. 
DynCorp's principal work in New Mexico is providing business
services, which include aircraft maintenance and repair at military
bases, and operations services provided at government-owned
facilities.  In 1996, DOD awarded prime contracts to DynCorp to
provide maintenance and repair services to equipment and laboratory
instruments, telecommunications services, and other services
associated with operating a government-owned facility at White Sands
Missile Range, among other things.  DynCorp's survey data was
completed by staff at the corporate headquarters in Reston, Virginia. 
DynCorp's responses were based on financial data for DynCorp and its
subsidiaries that also operate in New Mexico (e.g., Aerotherm). 

EG&G's principal DOD work in New Mexico is providing communications
equipment; operating radar and navigation facilities at Holloman Air
Force Base; and doing advanced research, development, testing and
evaluation work.  In 1996, DOD awarded prime contracts to EG&G to
provide advanced development and exploratory research and development
(including medical) services at Kirtland Air Force Base and to
operate radar and navigation facilities at Holloman Air Force Base,
among other things.  EG&G's survey data was completed by staff at the
Albuquerque office and includes data only for EG&G Management
Systems. 

Kit Pack Company is located in Las Cruces, south of Holloman Air
Force Base near White Sands Missile Range.  Kit Pack's principal DOD
work in New Mexico is providing aircraft spare parts and modification
kits.  In 1996, DOD awarded prime contracts to Kit Pack to provide
aircraft hydraulics, vacuum and deicing system components, airframe
structural components, and torque converters and speed changers,
among other things.  After it completed and returned the survey to
us, Kit Pack officials informed us that it was currently operating
under Chapter 11 bankruptcy due to the termination for default of an
Army contract.  Kit Pack had filed an appeal of the termination,
which was pending when we completed our work.  The company indicated
that it has seen a severe reduction in the number of DOD contracts
awarded since it filed for bankruptcy.  Kit Pack staff in Las Cruces
completed our survey. 

We were unable to obtain survey information from Lockheed Martin. 
Company officials indicated that they did not have the type of
information we requested broken out by states or geographical
locations.  In a follow-up meeting, company officials provided us
with information on their total expenditures to New Mexico suppliers,
annual payroll for their employees in New Mexico and the number of
employees in the state between 1992 and 1996.  The information was
developed by staff in Lockheed Martin's Washington operations office. 

We could not use Lockheed Martin's information because it was not
broken out by specific federal agencies, nor could we determine
whether the total expenditures, payroll, or employment were
associated with government-funded work or whether they were part of
the company's commercial business.  Over the course of several
meetings and conversations with Lockheed Martin officials, we
obtained detailed supplier expenditure information from the Lockheed
Martin Consolidated Procurement Program which was broken out by
specific Lockheed Martin business units.  Company officials said that
this would provide an indication of the type of business activity
(e.g., DOD, DOE, NASA, and commercial) that the expenditures were
made for.  In addition, we were given information on corporate sales
and payroll by staff in Lockheed Martin's tax department. 

We discovered several discrepancies in the company's financial
information.  When we discussed these with company officials, they
indicated that the data provided by the Washington operations office
were "less reliable" than other data.  Company officials also
indicated that their record-keeping had been challenged by the recent
merger/acquisition activities (i.e., Lockheed and Martin Marietta in
1995 and the Loral acquisition in 1997).  Lockheed Martin officials
said that different companies had different information systems and
that some information may have been lost during the recent merger. 


   KEY LIMITATIONS
------------------------------------------------------- Appendix III:3

Our survey was not designed to specify or measure the exact amount of
all DOD contract dollars that flow into New Mexico.  Rather, its
purpose was to reflect the nature of the flow of DOD prime and
subcontract dollars to a sample of top New Mexico contractors and to
compare these results to existing DOD data. 

Among the four contractors that completed the survey, none indicated
that they could not provide reliable responses to the survey items. 
The most common limitation was the lack of information on FTEs and
wages for subcontracted work.  Specifically, contractors indicated
the following limitations in their responses to us. 

  -- Honeywell provided information on the dollar amount of the
     orders it received during the calendar year and estimates of
     subcontracted work and employees and wages associated with
     subcontracted work. 

  -- Kit Pack did not have FTE or wage information on its
     subcontractors and indicated that it no longer had payroll
     records for its own staff for 1988, 1989, or 1991. 

  -- EG&G did not have records for FTEs and wages associated with
     subcontracted work. 

  -- DynCorp did not have information on its subcontractors prior to
     1993.  To report fiscal year information, DynCorp had to convert
     some company financial data that was not identified by fiscal
     years. 


   SURVEY FINDINGS
------------------------------------------------------- Appendix III:4

We treated all survey data received from contractors as proprietary. 
Therefore, in discussing survey findings, contractor names are not
used and data is aggregated to protect business-sensitive
information.  All dollars were adjusted for inflation and are
constant 1996 dollars.  All of the contractors surveyed were DOD
prime contractors.  Two of the four contractors we surveyed indicated
that they were also DOD subcontractors. 

The total amount of DOD prime and contract subcontract awards has
declined over the 9-year period.\6 The totals reported for 1996 were
the lowest of all the years.  For the 9-year period of our survey,
expenditures for DOD prime contracts ($1.5 billion) were roughly the
same as for subcontracts ($1.4 billion).  However, in 5 of the 9
years, the contractors received more subcontract than prime contract
dollars (see fig.  III.1). 

   Figure III.1:  DOD Contracts
   Awarded to the Top Four New
   Mexico Defense Contractors
   (1988-96)

   (See figure in printed
   edition.)

Between 1988 and 1996, the percent of prime contract dollars that
remained in-state was consistently greater than 80 percent (see fig. 
III.2).  The 9-year average was 83 percent. 

   Figure III.2:  Contract Dollars
   Received by the Top Four New
   Mexico Defense Contractors That
   Stayed in-State (1988-96)

   (See figure in printed
   edition.)

Although the average percent of prime contract dollars that remained
in New Mexico was high, examination of specific contractor data
indicates important exceptions.  For two of the contractors, the
survey results indicated that nearly 100 percent of the prime
contract dollars they received remained in-state between 1988 and
1996.  However, one contractor's data shows that less than 50 percent
of prime contract dollars received remained in-state each year
between 1988 and 1996.  Approximately 70 percent of the total prime
contract awards received by another contractor remained in-state for
all years (see fig.  III.3). 

   Figure III.3:  Differences in
   Percent of Prime Contract
   Dollars That Remained in-State
   (1988-96)

   (See figure in printed
   edition.)

For the two contractors that were also DOD subcontractors, a slightly
smaller percentage of their subcontract dollars remained in-state
compared to the percentage of their prime contract dollars (see fig. 
III.4).  On average, 75 percent of subcontract dollars remained
in-state between 1988 and 1996. 

   Figure III.4:  Subcontract
   Dollars That Stayed in-State
   (1988-96)

   (See figure in printed
   edition.)

The contractors indicated that the majority of jobs supported by
their DOD prime contracts remained in-state.  On average, 73 percent
of the jobs remained in-state during 1988-96.  The lowest yearly
percentage was 66 percent in 1989 and 1990, and the highest was 83
percent in 1996 (see fig.  III.5). 

   Figure III.5:  DOD Prime
   Contract and Subcontract Jobs
   That Stayed in-State (1988-96)

   (See figure in printed
   edition.)

On average, 73 percent of the total wages for employees working on
DOD prime contracts and subcontracts remained in-state between 1988
and 1996 (see fig.  III.6).  From 1988 to 1996 the percent of wages
that remained in-state generally increased. 

   Figure III.6:  Wages for DOD
   Prime Contract and Subcontract
   Work That Stayed in-State
   (1988-96)

   (See figure in printed
   edition.)


--------------------
\6 This is consistent with the general decline in total DOD
procurement budgets.  See Defense Industry:  Trends in DOD Spending,
Industrial Productivity and Competition (GAO/PEMD-97-3, Jan.  1997). 


   SURVEY RESULTS COMPARED TO DOD
   RECORDS
------------------------------------------------------- Appendix III:5

We compared our survey results to DOD's records of the total amount
of contract awards received by the contractors between 1994 and 1996. 
DOD sources collect and report information only on prime contracts
while our survey collected information on DOD prime contracts and
subcontracts.  Thus, we expected that DOD's records and the
contractors' would be different as was revealed in the survey. 
Therefore, we compare DOD's records of total prime contracts to our
survey results on the amount of prime contracts received by the
contractors in New Mexico and that remained in the state.  However,
to shed further light on and quantify, where possible, the
limitations in existing DOD data, we also compared the amount of
total contracts, defined as in-state prime contracts and
subcontracts, to the DOD totals, defined as prime contracts (see fig. 
III.7).\7

   Figure III.7:  DOD Records of
   Total Contract Awards Received
   Compared with Our Survey
   Results of Contractor Records

   (See figure in printed
   edition.)

Note:  Dollars in millions, except where indicated.  Our prime
contract data includes total prime contract awards that remained in
the state and our total data includes prime contracts and
subcontracts that remained in-state.  All DOD data is from the DD350
database. 

The overall comparison between the contractors' records and DOD's
records of total prime contract amounts shows that DOD records can
both overstate and understate the total amount of prime contracts
that actually end up in a state's economy.  In 1994, the contractors'
records show that $93.6 million in DOD prime contract work was done
in New Mexico.  On the other hand, DOD's records indicate that the
contractors received $144.9 million in prime contracts, representing
a possible $51 million, or about a 54-percent overstatement. 
However, in 1995, the contractors' records showed that $143.3 million
in DOD prime contract work was done in the state, whereas DOD's
records show that the businesses received $117.2 million,
representing a possible $26-million, or about an 18 percent
understatement. 

As expected, a comparison of the contractors' records of the total
contracts (in-state prime contracts and in-state subcontracts) to the
existing DOD records of total prime contracts shows that the totals
reported by the contractors were consistently greater than the totals
reported in DOD's records. 


--------------------
\7 When reporting the total value of prime contract awards for
specific companies, DOD reports information according to the
"ultimate" owner of the company.  For example, DynCorp has a
subsidiary named Aerotherm that also performs work in New Mexico. 
Therefore, DOD's records of contract totals for DynCorp includes
awards to Aerotherm.  We accounted for this DOD reporting procedure
when conducting the comparisons between DOD records and the
contractors' records. 


   SUMMARY
------------------------------------------------------- Appendix III:6

Between 1988 and 1996, more than 70 percent of the DOD prime contract
and subcontract dollars and associated jobs and wages for the top
four contractors in New Mexico remained in the state.  However, a
notable exception in one contractor's data, combined with the fact
that the principal type of contract awarded to firms in New Mexico
generally is for services, suggests that the survey findings may be
explained, in part, by the nature of the contracts that are awarded
to the top New Mexico defense contractors. 

The responses of one contractor showed that about one third of the
DOD prime contract dollars received remained in state between 1988
and 1996.  This pattern is an outlier relative to the data for the
other three contractors.  Further, the DOD prime contracts this
company received between 1994 and 1996 were exclusively for hardware
and equipment.  On the other hand, nearly 100 percent of two
contractors' prime contract dollars remained in the state and all of
the contracts were for services (e.g., operation of military
facilities, technical services).  Data for the single remaining
contractor indicate that 70 to 80 percent of its DOD prime contract
work remained in the state between 1988 and 1996.  This contractor
received DOD contracts for services and for some hardware procurement
between 1994 to 1996. 

The presence of four military installations in New Mexico (Cannon,
Holloman, and Kirtland Air Force Bases and White Sands Missile Range)
drives the need for services to support the operation and maintenance
of the installations.  DOD's data show that in each year during
1988-96, the largest dollar amount of DOD prime contracts awarded to
all businesses in New Mexico was for services.  Further, DOD's data
shows that between 1994 and 1996 the four military installations
consistently received the largest dollar amount of DOD prime
contracts compared to all other cities or locations in the state. 
Given that these installations generate service contracts, it is
reasonable to expect that the dollars received by these top four
contractors for performing the service contracts would remain in the
state.  However, the survey data indicate it would be inappropriate
to generalize that expectation to all types of DOD contracts (e.g.,
major hardware, equipment, and supplies). 

DOD's official data on prime contracts for the top four contractors
overstated as well understated amounts reported by the contractors. 
Our research suggests at least three reasons for this.  First, DOD's
data does not account for leakages of DOD prime contract dollars from
the state's economy that may occur through the subcontracting
process.  All of the contractors we surveyed indicated that they
subcontracted out of state at one time or another or did it
consistently.  Second, DOD is required to report only on prime
contracts greater than $25,000.  This reporting threshold may be very
inclusive for certain types of DOD purchases (e.g., major weapon
systems).  However, for other types of DOD purchases
(e.g., service contracts), this reporting threshold may mask a large
proportion of actual DOD expenditures.  And third, the existing DOD
data sources do not account for injections into a state's economy
from subcontracts that companies receive from other DOD contractors. 

*** End of document. ***