C €f- Z : ff s Price Changes of Defense Purchases of the United States Prepared for Department of Defense Assistant Secretary of Defense Comptroller i1 PREFACE This study is part of continuing research on the national income and product accounts prepared by the Bureau of Economic Analysis, U.S. Department of Commerce. This report was prepared for, and with the financial support of, the U.S. Department of Defense. The judgments expressed here do not necessarily reflect those of the Defense Department, Within the Bureau, general direction was provided by Charles A. Waite, Chief Economist. The research was implemented in the Government Division, Joseph C. Wakefield, Chief. Technical direction and super- vision were provided by Richard C. Ziemer, Chief, Price Measurement Branch. Administrative management was provided by Walter H. Bennett, Special Assistant to the Chief, Government Division. Supervisory Economists Robert J. Shue and Vincent J. Kamenicky provided leadership to: John S. Aldrich, Debra M. Crowe, Karl D. Galbraith, Michael G. Kassack, Jean A. Orme, Robert S. Luke, Richard D. McCarthy, Robert T. Mangan, Abner Sachs, Leon M. Taksel , and Stephen A. Whatley. Secre- tarial support and statistical assistance were provided by Claudia R. Littlefield, Abigail L. Cummings, Nancy J. Nevin, and Darlene V. Johnson. Allan D. Searle, formerly Chief of the Division of Industrial Price Indexes, Bureau of Labor Statistics, U.S. Department of Labor, served as consultant. This study benefited from the strong assistance and support of Clifford J. Miller, Deputy Comptroller, (Plans and Systems) in the Office of the Assistant Secretary of Defense (Comptroller). Leonard G. Campbell served as his representative. Defense Department participants were: Commander Rolf H. Clark, Navy; Captain Bobby Jackson, Air Force; Charles M. Magrum, Army; and Brandon B. Shea, Defense Logistics Agency. m Digitized by the Internet Archive in 2013 http://archive.org/details/pricechangesofdeOOunit TABLE OF CONTENTS Page Part 1 : Background 1 Project organi zation 2 Report structure 5 Part 2: Results 6 Overal 1 trends 6 Durabl e goods 6 Nondurabl e goods 15 Services 16 Structures 16 Compensation 19 Outlays 19 Part 3: Framework 21 Current-dollar purchases 21 Constant-dollar purchases 23 Quality change 24 Implicit price deflator 25 Comparison with gross national product 25 Ou tl ays def 1 a tors 26 Part 4: Development 27 Data constraints 27 Procedures 28 Categories 31 Part 5: Evaluation 47 Input data 47 Reliability of results 48 General pricing problems 48 Major worki ng groups 49 Future plans 50 Historical data 51 Appendix A: Detailed Methodology 53 Introductory note 53 Ai rcraf t 53 1 . Definition 53 2 . Concepts ,. 53 3 . Sampl e 55 TABLE OF CONTENTS (continued) Page 4. Weights 55 5. Prices 57 6. Quality 59 7. Sources 60 Ammunition 61 1 . Definition 61 2 . Concepts 61 3. Sample 62 4 . Weights 64 5 . Pri ces 64 6 . Qual i ty 64 7. Sources 65 Communications 65 1 . Definition 65 2. Concepts 65 3. Sample 66 4. Weights 66 5 . Pri ces 67 6. Qual i ty * 67 7. Sources 67 Compensation 68 1 . Definition 68 2. Concepts 68 3. Sample 69 4. Weights 69 5. Prices 69 6. Qual i ty 71 7. Sources 71 Construction 71 1 . Definition 71 2. Concepts 72 3. Sample 73 4. Weights 74 5. Prices 75 6. Quality 76 7. Sources 77 vi TABLE OF CONTENTS (continued) Page Depot Maintenance 77 1 . Definition 77 2. Concepts 78 3. Sample 78 4. Weights 80 5. Prices 80 6. Qual i ty 81 7 . Sources 82 Electronic Equipment 82 1 . Definition 82 2 . Concepts 82 3. Sample 83 4. Weights 85 5. Prices. 85 6. Qual ity. 85 7. Sources 86 Installation Support Services 87 1 . Definition 87 2 . Concepts 87 3. Sample 88 4. Weights 88 5. Prices 89 6. Quality 89 7 . Sources 90 Medical Services 90 1 . Def i ni ti on 90 2. Concepts 90 3. Sample 91 4. Weights 91 5. Prices 91 6. Quality 92 7. Sources 92 Missiles 92 1. Definition 92 2. Concepts 92 3. Sample 93 vii TABLE OF CONTENTS (continued) Page 4. Weights 94 5 . Pri ces 95 6. Qua! i ty 96 7. Sources 96 Other Equipment 97 1. Definition 97 2. Concepts 97 3 . Sampl e 97 4. Weights 98 5. Prices 98 6. Quality 99 7. Sources 99 Other Services 100 1 . Definition 100 2 . Concepts 1 00 3. Sample 100 4. Weights 101 5. Prices 101 6. Qual i ty 102 7. Sources 102 Research and Devel opment 1 03 1 . Definition 103 2. Concepts 1 03 3. Sample 104 4. Weights 104 5. Prices 104 6 . Qual i ty 1 05 7 . Sources 105 Retired Military Pay 106 1 . Definition 106 2. Concepts 106 3. Sample 107 4. Weights ." 108 5. Prices 108 6. Qual ity ' 108 7 . Sources 1 08 viii TABLE OF CONTENTS (continued) Page Ship Construction 109 1 . Def i ni tion 1 09 2. Concepts 1 09 3 . Sampl e 110 4. Weights Ill 5. Prices Ill 6 . Qual i ty 113 7 . Sources 113 I Stock Funds: Defense Stock Fund 113 1 . Def ini tion 113 2. Concepts 114 3. Sample 116 4. Weights 118 5. Pri ces 119 6. Quality 120 7. Sources 120 Stock Funds: Military Services 121 1 . Definition 1 21 2. Concepts 121 3. Sample 122 4. Weights 1 22 5. Prices 1 23 6. Qual ity. . .' 1 23 7. Sources 124 Transportation of Things 124 1. Definition 124 2 . Concepts 125 3. Sampl e 125 4. Weights 1 26 5. Prices 127 6. Qual ity 1 27 7. Sources , 127 Travel and Transportation of Persons 1 28 1 . Def i ni tion 128 2. Concepts 128 3. Sample 129 IX TABLE OF CONTENTS (continued) Page 4. Weights 130 5 . Pr i ces 130 6 . Qual i ty 1 31 7. Sources 131 Utilities 1 31 1. Definition 131 2. Concepts 1 32 3. Sampl e 1 33 4. Weights 133 5. Prices 134 6. Qual i ty 1 34 7. Sources 135 Vehicles 135 1 . Def i ni ti on 135 2. Concepts 135 3. Sample 136 4. Weights 137 5. Prices 1 38 6 . Qual i ty 138 7 . Sources 1 38 Weapons 1 38 1 . Def i ni ti on 1 38 2. Concepts 1 39 3 . Sampl e 1 39 4. Weights 140 5. Prices 140 6. Qual i ty 1 40 7. Sources 141 Appendix B: Detailed Results 143 Table B-l. -- Implicit Price Deflators for D0D Purchases of Goods and Services, by Major Category 143 Table B-2. -- Implicit Price Deflators for D0D Purchases of Goods and Services, by Major Category 144 Table B-3. -- Implicit Price Deflators for D0D Purchases of New Aircraft Components, by Military Service 146 x TABLE OF CONTENTS (continued) Page Table B-4. -- Implicit Price Deflators for DOD Purchases of Ammunition Components, by Major Category 148 Table B-5. -- Implicit Price Deflators for DOD Purchases of Communications and Electronic Equipment, by Major Subcategory 150 Table B-6. -- Implicit Price Deflators for Compensation of DOD Military and Civilian Personnel 152 Table B-7. -- Implicit Price Deflators for DOD Purchases of Construction, by Major Subcategory 154 Table B-8. -- Implicit Price Deflators for Selected Subcategories of the Defense Stock Fund............ 156 Table B-9. -- Implicit Price Deflators for DOD Purchases of New Missiles Systems Components, by Major Subcategory 158 Table B-10. -- Implicit Price Deflators for DOD Purchases of New Ships 160 Table B-ll. — Implicit Price Deflators for DOD Purchases of Transportation of Things 162 Table B-12. — Implicit Price Deflators for DOD Purchases of Travel and Transportation of Persons 164 Table B-13. — Implicit Price Deflators for DOD Purchases of Utilities, by Major Type 166 Appendix C: Observation Group 169 Commi ttee Members 1 69 Agenda - October 19, 1976 169 Agenda - May 17, 1977 169 Agenda - June 27, 1978 170 Appendix D: Appropriation Categories 171 Mi 1 i tary personnel 171 Reti red mi 1 i tary personnel 171 Operation and maintenance 172 Procurement 172 Research, development, test, and evaluation 173 Military construction . 173 Family housing, defense 173 Civil defense 174 Glossary of abbreviations 175 xi Part 1 : Background Measurement of changes in the amount of national security being purchased by the Government has presented serious problems for economic analysts. National security varies not only directly with a nation's own defense expenditures and its perception of the threat, but inversely with increased effectiveness of an adversary's arms. If a value could be assigned to national security, then, at least conceptually, the price of national security could be defined in terms of units of security per dollar expended. However, this measure is impractical because of the difficulty in objectively defining and measuring a unit of national security. Most often, as a practical matter, the value of national security is thought of as the total value of goods and ser- vices purchased by the Government in support of the national defense. Changes in the quantity of national security purchased are then measured by removing the effect of price changes over time from these goods and services. However, there has been no objective way to assess accurately the impact of price changes on defense purchases, as all past measures have been based on price changes of goods and services purchased in the private sector of the economy for nondefense purchases. The Joint Economic Committee, congressional budget committees, Congressional Budget Office, Department of Commerce, and Department of Defense (DOD) expressed concern over the absence of a defense deflator in the national income and product accounts (NIPA's) and the inade- quacy of the existing Federal purchases deflator to reflect price changes accurately for DOD purchases. The Bureau of Economic Analysis (BEA), in its continuing research effort to improve the gross national product (GNP) estimates, also recognized the need to overcome the major statistical shortcomings of the existing Federal purchases deflator, namely, the use of private sector price indexes and the inadequate detail on the type of DOD purchases. Substantive work on improving constant-dollar estimates of defense purchases began in February 1973, when BEA began to study both the ideal requirements and the feasibility of measuring the effect of price changes on military expenditures. The results of this study, sponsored by the U.S. Arms Control and Disarmament Agency, were -, reported in June 1975 in Measuring Price Changes of Military Expenditures . The report (1) concluded that it was feasible and essential to construct price indexes for DOD expenditures, largely based on data available in DOD; (2) developed specific proposals for pricing various types of U.S. defense expenditures; and (3) recommended that DOD support the preparation of a Department-wide price index. U.S. Department of Commerce, Bureau of Economic Analysis Measuring Price Changes of Military Expenditures , June 1975. ; With the urging of the Joint Economic Committee, U.S. Congress, and the presentation of a study proposal prepared by BEA in August 1975, DOD entered into an agreement with BEA for BEA to research and develop measures for estimating defense purchases in constant prices within the framework of the NIPA's, and to publish an official defense deflator. Also, BEA was to provide DOD with the basis for measuring and fore- casting inflation rates, particularly rates for budget appropriation categories and weapon systems. Finally, BEA's work was to serve as a pilot effort for the development of deflators for other Federal agencies, State and local governments, and other nations. The results of this project are measures of the real volume of inputs used to provide national defense (national security), not the amount of national security that Government expenditures provide. Project organization The BEA staff consisted of approximately 20 professionals organized along functional lines relating to the types of military procurement of goods and services. In addition, each of the military services and the Defense Logistics Agency (DLA) provided a representative to parti- cipate in the project. A detailed study plan was developed that defined work units, listed and described tasks, and established schedules and deadlines. Documentation of the research began early by means of the preparation of position and technical papers by the BEA staff. DOD purchases were categorized into three working groups that covered 52 work units (figure 1). The structure of the units was based on an anticipated conceptual approach, identification of similar products and services, military service, budget appropriation category, and ease of data collection. The project was conducted in three phases: the framework of the research was constructed; the deflators were developed; and, finally, maintenance procedures were established to continue estimating the constant-dollar series in the NIPA's (figure 2). Emphasis was placed on obtaining actual prices paid by DOD, col- lecting detailed data on types of purchases, and minimizing any undue reporting burden on DOD. Because data collection from the private sector is extremely difficult to initiate and expensive to maintain, the use of existing DOD data improved the speed of collection and maximized the amount of data that could be handled within the given time frame. Only new ship construction and contract research and development (R. & D.) data were obtained from private industry. Over a 2-year period, the BEA staff and DOD representatives visited and discussed data requirements with the representatives of many military commands and private industrial firms. In some cases, data files containing millions of entries had to be processed to develop the basic price indexes. Approximately 15,000 basic series were required to develop the results presented in this study. Many DOD officials were briefed on the progress of the research, as were various congressional committee staffs and offices including the Joint Economic Committee, Senate and House armed services committees, to CD O •r— > S- CD to O) e uo to UJ to o CD 1 — 1 u > o cc s- LlJ Q. oo >> e S- ro CO •!- -M r— rO to +-> c re o a ro •■- o +-> "r- rO e E 3 O E +-> E =5 O 4-> +J re fO to to e e a> cu CL Q. E E o o o o CD u i~ o u_ >>>> S- E > •r- S_ f0 < c c c IQ (O ID E E E CD CD CD +->+-> +J c c c: ro ro ro e O CD •i- o +-> e s- CD to 4-> +•> to e E T- O fO a E I I O) en e e to to 3 3 o o -E sz o Ql Cl to rO >> >,.— fO 4-> +->+-> i — i — ro U O O O T-T- +J •!- Q. Q. Q. E E to "a CD CD CD re ro E CD O Q Q U_ Lx. i— i s; c o •r— +J cj 3 S- CD -M O to S- e o o u. >>T- S- s- «=c < re I I 4-> S- i~ •i- CD CD O O E O to s- CD a. 4- o e o •r— +J o -a o S- Q. CD S- -o e re C7> CD E OVr- ro +-> to +J E +■> l/> '.i— c o s- ai D_ Q_ CC +J to e ct> CD e E •<- Q..E O -M CD 4- > O CD -O E O -a -r- E 4J re re +-> re 4-> S- o Q. to e re s- +J E to ro CD re to CD E to ro CD S- CC I— CD t- > r— ro t- S- +-> to to i to >> •r- o o o o E •r— 'r~ •r™ E ZD E O E O E O CD ^ S- s- S- 4-» o u O O to s 00 CD CD CD o #* O O C3 CD o s- CD 'a; X) to •r— CD CD o +J E E E CJ) o s- a: u_ X fO ro fO o u_ 3 LU >> >> CD CD _J >^ >» CD re S- E > E +-> to to to O S- E > •r- I— •1 — S- ro 3 E E E s- CD •1 — S- ro U_ o < +J >, CD >>"C o O O s- ro s- ro ro s- E 4- > E C C C •r - •i— •r- tO 4- CD CJ o O O •r" s- CD ra 3 3 3 CD to -t-> -«-> +-> u cu E •r— u_ •r- •i— ca;ea;Q2:i4_i4_u_4-> a» to •r— ■i— •!— •r— Q- •i — E E >, E >■ i 1 1 1 to i — E E E E E SZ rj s- 3 E 3 > s- S- S- s- ^: -^ -^ «i- o o 3 3 3 O ^>i +-> E '1 — E s- E ro CD cu CD CD O O O tO ••- Q. E £ £ •r— f— o E «a: E 3 I — o o O +-> +j +-> -M •)-> +J +-> 3 CD CD <<<3 CD O S- to LO o E CD ^~ to U- o to CJ UJ CD >, >> •I"" CD •I— 1— CD i- E > S- -M XJ i — > C/) U •i— i_ ro *r— O CD "r— S- >- S- 1 — to <=C ^ •i— to to to +-> o E 'i— z: >> >> E E CD CD CD to •1 — 3 E E o S- E > CD r— r— t— E to ro CD Q_ •r- S- ro E O t- -r- •!- o S- i — o +-> +-> to to to > CU +-> >5 "^g ■M +-> +-> E tO •!— T- •!— Q. E E to to 4-4-4- O •r- E E E •r— o •r- •1 — ro ro ro U JE u S- 1 — E s_ s- s- s- r— r^ f— ^~ to ro J — O u o a CD ro n— i— r— . Q. E ra Gi- u s- s- -M -Q ro ro ro 2 •i— JD j2 ro •i— "r - *r— E E E E CD x: 3 CD 3 ■»-> to OJ to c u CO O) o c c ^~ e 3 0) »o 3 fO CD ■o s- r^ S- > +-> S- •i — o +J ^~ CO Q. rt3 s- Q -a CL) •a Q. a> O Q. ^~ O a> ^— > CD CD > x> T3 T3 0) TO cd +■> ■o CD CD CO to to > to -a -o a> -^ CD •r- CD CD cd s- $- 3 CD U +-> +-> 3 o CT O O 03 i- S- Q. 3 U E +-> a +-> o 4-> +* 4-> > cu ^™ ^~ ^~ .c -C .c CO o > Q- a. Q- en CD zr> ^~ S- •r— E E E •r— •1— ■r- M- Q. C fO «3 fO CD CD CD CD Q. ^> oo C/) tn 3 s s Q < ^ c CD •1— S- o O +J o •r— i— c fO i — s 4-> rtJ •I"" > E C O a. ^~ o CO •r— Q. CD rtJ i- S- 4- O o c o. u. CD S- o •r— Q. Senate and House budget committees, Senate and House defense appropria- tion subcommittees, and Congressional Budget Office. An observation group composed of experts in the fields of pricing and defense, periodically reviewed and commented on the conceptual approach and statistical status of the research. Report structure A summary of the results is presented in part 2; the framework of the study in terms of current- and constant-dollar purchases in the NIPA's in part 3; a brief discussion of the methodology in part 4; and an evaluation of the research and development activity, application to related areas, and future research plans are presented in part 5. The appendixes provide significantly more detail concerning the results and methodology. Part 2: Results Preliminary implicit price deflators (IPD) for DOD purchases for each calendar year (CY) 1972-77 are shown in table 1, and current- and constant-dollar estimates of purchases for the same periods are shown in tables 2 and 3. Although the estimates have been prepared on a NIPA basis, they differ from the estimates currently published by BEA, because of minor statistical problems. Overall trends During the period CY 1972-77, the defense purchases IPD increased 41.9 percent, compared with the 40.4 percent increase for the total private GNP IPD (table 4). From 1972 to 1974, the IPD for defense purchases increased 15.1 percent while the private GNP IPD increased at a slightly faster rate, 15.6 percent. In CY 1974-77 this pattern was reversed as the IPD for defense purchases rose 23.3 percent com- pared with a 21.4 percent increase for the private GNP IPD. However, the IPD for defense purchases excluding compensation, increased at a faster rate than that for the total private GNP for every year in the period CY 1972-77. 2 The differences in these annual rates were small for the period 1972-75; however, in CY 1975-77, the IPD for defense purchases less compensation increased 16.7 percent, 53.8 percent higher than the increase for the private GNP IPD. Federal purchases in the NIPA ' s are classified into five major product groups: durable goods, nondurable goods, services, structures, and compensation. 3 DOD purchases have been similarly classified; results in each classification are discussed below. Durable goods Durable goods accounted for approximately 23 percent of total DOD purchases in CY 1977, up from 22 percent in CY's 1976 and 1975 and approximately 20 percent for the period CY 1972-74. Included in this group are all weapon systems (e.g., aircraft, missiles, ships), combat and nonconbat vehicles, communications and electronics equipment, troop support equipment, as well as other types of equipment. The IPD for DOD purchases of durable goods increased between 8 and 10 percent a year for the period CY 1974-77, up significantly from the 2-3 percent annual increases in CY's 1973 and 1974. The small increases in the DOD IPD in CY's 1973-74 were the net result of lower prices of new aircraft (20 percent of durables) and new missile components (6 percent of durables) largely offsetting higher prices for other durables, The IPD for new aircraft components (not including modification kits or spare and repair parts), increased 1 percent in CY 1973 but decreased 2. Exclusion of compensation from DOD purchases makes a more consistent comparison with private GNP (see section 3, pages 35-36). 3. Definitions of these groups appear in section 3, page 36. CO -t-> U $- 3 (O "O CL O a) s- Q D- O o S- <+- i— O O II 4- CSJ CO Q-O^ s- >»— O I— +-> >- i— o O) ro q s: , z. o •I-" Q. a> fC E o •r— -»-> OM co QJ S- o i — cr> en r-» lo lo >3- cn r-* LO LO LO LO 3 +J OWOJOi — en r-- co i— cs «vl- co lo cn O O i— CM oo <=d- <=!- CfldOC O O i— r— 3 r— p— i — n— f— p— S- +-> CO CO a> o cnro NOi i — r-rs^fs >=i- cn -vf csj u •i— OLO^Or-'sl- COOlOt- ro <3- r«» co > ooi-m— CSJ CO cn o o c o o o o ! zs D1 f— f— f— i~ ~ n~ i—~ r*~ Q CO QJ o lo r-. lo lo csj ro <^- i— o r— CO O LO O LO CO en csj csj cn r- csj ro cn ^j- f0 (0 +J JC o lo en >^- csj i — co cn cn cs. ^- lo r-^ CTi o u oor-wn^- CTi CTi CTi C o o o o 1— S- Q. 1 — 1 1 — 1 i — i i — i > • i— i i — i > csj oo «3- lo lo r^ CSJ CO r-~ r-- i--. r-^ r--. r-. l^« r-^ Cn CT) Ol Cn C7"> Cn en CTi >->->->->- >- >- >- uuuuuu c_ ) c_ > s- ra - i- fO -o c CD to CU lO 03 sz o S- 3 D_ E C -»-> o S- u 03 Q. 1 a) Q +-> o a o S- 3 i — o -a n 4- O CM S- r~- to D_ CT> s- , — o <+- +-> O >- (0 C_> 1 — , o I— CD S_ o o •1— •<—> i. m D_ -i_ +-> >. "i — .a a •r— ^— a. E cu -a 03 +-> (0 V) c CU Q. E O o to CD S- +-> o s- +-> to cu u > cu I/O cu .a 03 I/O s- -a 3 o ■o o C C7> O CU i — oo -Q T3 fO O S- O =S CD Q 03 o to cu in 03 o r- >v|- i— CO 00 C\J CM 00 roujoii — 'tNWO i— r— CO 00 oo cr> o cr> co LD <3" i — CT> C\J CM tO to CM p— O CTi 1-^ ID CM CM CO CO NOOfOi — CO «3- <=3" <^- ■3" ID ID ID co ld id id «=d" tO 00 CO CO O Otnootn OOOkDN CTi CM i — CO O ■ — CM CM ^COr-rv CM CM CO CO a-> cr. <3- "3- CO CO «vl- "vt" O "3" ID LD LD LD ID ID 1^- O O i— CO ■sITv LO LO CT> ID CM 1-^ CM ID CO ID ^■LfilVOO CO ID CM CM 00 r-«. C~> CTi NOLD<^ CTi CT> CT> C3~i CO CM CT) CM O t— r- i— CM CM CM CM -3- 1— cr> i— «* r-^ <* cr> LON^d- N LD .— O «3" CO CO ID tO o o o o i — co r^ to CM CO «3" *3" CM CM CM CM anDmo CM CO CO «vt" CT> LO tO CO OLONCTi CO O CM i— o r-» i — id CO CTi CTi CTi ^" CO tD CO CONr-O CO ■— CO cn i — cm to >- LD cr. >- to at >- o CT> >- s- cu 5- S- 03 -a c cu 03 >- o n to CD to ns -E 4-> o o J- =J 3 X> CO* £- i— CD D_ i— 00 O e 4- -o cu o 4- 4- a> cu o Q CL >> to E o i- -I- O i— 00 4- o 0) (O'r E 2:2: +-> s- >> fO .£} Q. CD Q I CM CU A3 E O +-> c\j co ro cm co cr> ■vl-ror-^ CD ID IX CD fO romro'd-coo co co cr> ix ID LD CM CD to cm rx IX ix id en CO CO CO 1— E #t *t «t r\ n rt CD ID ID |x CD CD CM CO CO CO CO 00 CO 00 CD Q. ro ro ro ro ■* ■* E O O to CD S- 1— rx 1— cd st 00 CM 00 CD CM CM ID IX CM 1 3 rx co 00 id st oc CM CM CO CO st CO CO CM 4-> s- l^iDNNCOO CO St CO CO CO st CO CO cu CO cu ^— _Q ID CO ID st r— CO CO O 1— IX r— CD CO O CO to O ID CO ID CO st CO St •— ^- in 1 — 1 — 1 — 1 — f— r— r— E en O 2: CD r-onO'tro CD CD st CM 1— CO CO CO r— - to CM LD CM ID ID Ix st CD CD CO 1— st CT) CD -Q -a criroNmooo ID CO ID rx CO CO O ID fO j- ^«tlT)NCOO CO CO CO CO CO CO CO CO 3 en Q to cu CO CO CM CO >— O CO CM O CO 1— 1— .— O 1— ■ i/i ifl'JtNOOCO CD st ID ID st st ID O ta ns ONOMIDN ID CD ID 1 — CO 00 1— st +-> -E U CM CM CO 1— CO O IX CO IX CO CO CO IX CO 1— Q. n ix r^oo co cn • ••••• 1 — 1 1— 1 1 — 1 > 1 — 1 t— 1 t— 1 > CM CO st ID ID IX CM CO rx rx rx ix rx ix IX IX CD CD CD CD CD CD cd CD 1 — 1 — >->->->->->- >- >- UUUUCJCJ <_> CD S- f0 cu >- S- (0 -a E CU fO CD >- o CO fC CO. <£ Q_ I — i 2: -O «N CD CO n CD c: 00 •1 — rO +j .c E O s- CJ CO 3 S- a. 1 1 — 00 O O c 13 "O cd -a M- M- cu S- O Q Q_ 00 ^- M- E O O •r- 4-> ai 1 — c Cl 1 — O) >> •1 — E K s~ +J i. S- fO Q. •"-J O) ro a 2: 1 >^ 1 JD CM CD r— _Q ra c "r— ' +-> to cn ■=*■ >3- CO 00 >=J- CM CM LO r^. CT> cm «3- r-^ to ra LO 1— - CT> to LO 1 — r~- cn CX> CO CO 1 — <=3- to to CO 00 e CD 1— 1— 1— LO to r-- to cn 00 CT> CT> CO CO CO CO o~> CX> OI CX> CT» cti cn CTi CT> CT> CTi CT> O 000 O Q. H— [— 1— — r— - ^^ E CJ) co CD S- LO CD 00 C\J CM LO r^. cm ^f- CM CO CM O LO . 00 CT> +-> LO LO LO LO LO LO LO LD LO >=d- "=d- LO mLO-d- «^- u 3 s- -l-> CO CO a> LO 00 CM ^t CM CM O CM CTi CO <3" 00 cn «3- r-- r^ a O O CO LO r-- cri LO <—, *d" LO CTi CTi r-«- CO > 1 — CO CO LO 00 ^f CM CM CM 1^ 00 CO CO O 1— «— ■?r «^- «* ^ vf <3" «sl- ^3" LO LO LO CD co a> p— _o 00 00 co r-« ro co j > C0 1— CM O ^ CO "=i- r>- ra CO CM CO LO CO r^- lo 1 — 1 — O CM LO LO r-- cn 00 00 s- -0 =J co to co lo "3" CO CM «d" CM 1— CM CO roro CO tO 1— CM LO 1— CO . -Q -O WONO ^J- CM CO <^f IO «3" LO O "vt- CO rO u s_ O 00 ^t- ro -^j- «vf- «>|- ">f «d- >* «=3" -^j- *3" LO LO LO LO 3 CD Q CO CD tO CTi CM LO r^ <*f CM LO 1— co r^ LO CO LO CM 1 — CO nco nn r-> 1 — r-. cm CM ^d" CO LO CO CO CM •3" +-> ra -C CO CD 1— CM O LO CO CO ■St CO CTi CO CTi 00 00 O CO CTi CTi O O O cn 1— 1 — 1 — CM CO I— Z5 Q. CM CM r— CM CM CM CM CM CM CM CM CM 1— 1 1 — 1 1 — 1 t— i > ^1- LO to r--. 1-^ r-» r-. r-- CT) CT> CT> CTi 1 — 1 — 1 — 1 5- >- >- >- C_> CJ) CJ C_3 S- - ra XJ c CD ra C_3 >- o 10 to 03 CO <=C CD to <+- <1J Q to S- +J fO O i— 3 I— -a o o Q S_ D_ CM ^ cr> o i— cu >- >> 4- h- M- O O S- +-> O to E -i-3 C CD 03 O E S ••- +-> r— cm en nctii — lo r-. lo en co 03 co co ^t co r^ co CM O P>» CM O CD <=j- CO to ojrovo ro inn CTi CTi r~» IX) ^1" CO CM CM c n «■> r> »\ rv #\ r\ e\ r> tfi #1 #\ W\ #\ ID LO CO C\J CM i — i — CO CO 00 CO 00 00 00 CO Q. co co co co co co E O C_> to CD N S- i — CO CO LO CO VO CO LO i— CM co co lo r-- 3 r^ r--» a~> lo co co co r-^ CM ^1" LO CO +-> LO r-~ LO LO >^- CO -3- *d- -vj- CO «vt- ■=*■ ■=*■ ^d- o ^ r. *t #\ n «n 3 S- +J to to r-- co CM CO «d- IX) u CsJ i— (-» i— CM LO wocncTi CO CM LO O •r— OJr-i-NOO r-» cm *d- r~- CO ■— LO LO > S- LO LO LO CO CO CO CO <3" CO CO CO -^ CO CO CM r— LO 00 CM 03 V) ooo^ LO LO CO CO i — i — i — r— r-~ 3 o n *» #s *\ *\ rv 9\ 9\ 9\ r» 9\ n "D O LO «vj- CO CM CM CM 1 — 1 — 1— 1— 1 — 1 — C en o z: r-v. i — lo ^3- LO CO LO l-x i^ i^. cr> lo 03 o S- o ^t CO *3" LO «=j" LO CO CO CO CO CO CO CM CO 3 CD Q to >^- ix) lo cri lo i — cm t— to ID >D ^J" LT) IX) CM i— CO LO 00 CO 00 CO i— (0 fO O CM CO CM i— en o r^ lo lo LO CO CD CO +-> -C o u cm oo rs lo co co COCOrvN NMr)yD h- 3 l"~- LO LO LO LO LO a. .... .... 1 — 1 1 — 1 1 — II — 1 > 1 — 1 1 — 1 > 1— 1 1 — 1 1 — 1 1— 1 1 — 1 M 1 — 1 1 — 1 CM CO «=t ID IX) r~- CM CO r-- r-- r^ r-» r-» r-. r-» !->> CTi 0~i 0~i CTi Ol CT> CTl CTi >->->->->->- >- 5- O <_) O C_) C_> <_) o o s- 03 a; s_ 03 "O c CU I— 03 O >- o 11 co S- .e E rC a O i — s- u i — 3 O D_ 1 o (L) +-> CM CO u r-- C 3 cr> cu -a i — M- o O) S- >- Q D_ c_> M- M- H- O O O +J CU GO e a. E cu >> o £ h- •i — -!-> i — S- s- i — ta o •r — a. ••"-J :>. a> re Q 1 -Q CO a> ^— -Q fO E o •r™ 4-> co <3- co co OOON «3" i — CO co co cr> fO «3" CO CM CH co en «d- o cri i — airv CO CO ^fr O CO E CM i — i — O i— r— O O CO CTi CO CO 00 00 00 CO CO CO CO CO 00 00 CO 00 r-^ r-* r-- r-^ r-^ r«^ r^ r^ o_ E o C_) CO CU S- CO CO CT> O UDi — r-- i — OlLOCOr- «* O CM o 3 i — CO O CO i — CTi CM CM O CO CO CD CTi r*» i— CM 4-> <3- ^f- >^- <=*• "St CO <* ■=*■ «d- CO CO CO CO CO CO CO U 3 S_ -l-> CO co cu [^ CM O CO NOUiCO CO CM»00 CO r- r-- co cr> u CO r— i— CO ^- O CT> CO CD 00 LD CM i — O CM CT> > Nr-yD^O ^•Lfir-O O CO CO CM i — CO CO CM CO ^" CO CO CO 'co CO CO CO CO CO CO CO CO CO CO cu 00 O) ^— _Q ld co r-~ i^ ir> i — oin fAJ00N<^ 00 |--~ CTi CM O CO o o i r VDN IXJID CD C£) LO CD CD LD LD LD LD CD CO LD 1 — i — i 1— 1 i— i i — i «3" LO CD r-» r~~ r-» r^ r^ CTi CX> CTi ' — 1 — 1 — 1 >- >- >- >- C_) (_> C_3 o s- cu >- S- ro "O E CU +— fO o >- o 12 GO CD to rO -E o i- 3 D_ CD OO E CD 4- QJ Q 4- O +-> a +j 3 E T3 CD O E S- -!-> D_ S_ ro i— D- rO a> e Q O O •r- o S- +-> r— O ro II 4- z: (XI r~^ to CO en s- to i — o o +-> s- >- rO C5 o p— M- a> CD -»-> Q ro > QJ •■- O l- •r- O- S- o_ -a e +-> ro Q. E ro to U tO Z3 o -a i- o en s- Q. CD 4-> i— ro ro > E ■r- O S- T- a. +-> ro e '4- o CD to c CD to 4- to E CD CD O Q i— •■- +-> OO ro CD 00 to E 4-> ro CD E -E Q. CD O E E S- O +-> 3 U S- Q. ro O- CD Q CD to E CD 4- CD Q 4- to O CD to +-> ro E -E CD O E S- +-> 3 5- Q- 3- o io lo to cm o o o i— co co *t OiinrM i — lo ro Nlfi tr> «d- ■?*■ r-» CO cr> o >— CTi CTi O O C\J ^" tO CO o o o o O co o cm r-» i — o cr> to co r-. r--. omiacoNCTi OCi-MCT* io«t roo orocrioo ro cm r^ co CO CTi o i— cri en o o COLO NN o o o o o to r>. to to cm omounnoi O <0 LO «vf- CM i— O O i — CM CO «d" CO ^" i— O cm cm cn r^ CO "3" CO LO CM CO CO >^- co cr> cr> cm CTi CTi CTi O ^d- LO 1^ CTi o o o o cm co «3- lo to r-. r--» r-. r-. r-- r- r^ CTI C> 0~> CX> C?> CTi CM CTi 1 — oououo >- C_> CO CT> >- o o e o (j 4- o 3 ro CD S- ca CD o S- CD E E O o 4- O CD E +-> S- rO Q. CD Q oo to • to =3 CD E CD •i-" sz to +J 3 CO £+3 E -a cd CD S- -E S- to 3 •i- o -Q 4- =3 O Q. >; to CD CD > 4-> $_ fO 3 E oo •r— -M CD to -E CD +-> S- ro E E CD O -r- >- S_ 4- - S- to ro "O T~ -a CD to E > >> CD • |— j^_ r~" S- ro ro CD E O Q - i — C_> 13 to CD to ro -E o 5- 3 Q- CD to E CD "O 4- CD 01 3 Q E •r- 4- +-> O E O +-> u E CD 1 E +-> +-> s- o (0 3 CL-O CD o Q S- o D_ O S- r o i — II 4- fO CNJ E r-~ to o en i- •r- i — o +-> 4J rd >- fD s: o ^— M- I/) Ol to Q o CD cr O •i — 0) S- +J D_ fO > +-> • i — •i— S- O CL •j— r— "O a. E E nD CD fC Private gross national product U3OC0>* «3-COCOLO C\J CT> LO CO kDNi — i — ooolo t^ i — co ■ — ■ — i — c\j cNJc\jc\JCNj co en en co nm^t'sr Department of Defense purchases less compensation oocolo ooco<* >^-ror^cNj cr> oo lo cti i — rv ro Lf> cmOi — i — i^r— «=d-r^ > — ooud i— OOOi — LO tO O CM LO to CT> CT> >=d-oooo i— i— CM CM CM CM CO CO CO CO CO CO <* IT) LO If) Department of Defense purchases cr> lo to co oooi— i— cocncncr) co r- i— o 0<*Or^CT> ONi-lD >* 00 tO CO CO r— CO CT> i— CM tO CXi CM CM LO CO O O CM LO 1 — ■ — i — CO i — i — i — i — c\jc\jc\jc\j co co co co co -=d- >vj- <=*■ CY 1974: I II III IV CY 1975: I II III IV CY 1976: I II III IV CY 1977: I II III IV E o E o o UJ 4- o 3 m CD S- 13 CD Cl CD o S- CD E E o c_> 4- o +J E CD E +J S- (a Q. CD o . . t/0 I/) • to r> cd E CD •!- .e to 4-> 3 CQ >> .Q 4J E -O CD CD S- -E S- IA 3 •i- o p"*" -Q 4- 3 O Cl >1 to CD CD > +J s_ ra 3 E o~i •r— +-> CD tO -E CD 4-> S- «3 E E CD o t- 5^ s- 4- " s- to fO -a -r- XJ CD to E > >, CD •r™ r~ ^— S- fO fO CD E o O et II . >- 1 — c_> 14 approximately 9 percent in CY 1974. Large deliveries of Navy new aircraft components with price decreases associated with second and third production contracts of new weapon systems more than offset price increases of Air Force new aircraft components. The IPD for new missile systems components declined 4 percent annually during the period CY 1972-74. For this study, DOD purchases of missile components were divided into those for "large" (inter- continental and submarine-launched ballistic) and "small" (all other) missiles; roughly half the value is in each category. Increases in the IPD for Navy small missiles during the period were not large enough to offset the decreases in Army (10 percent) and Air Force (30 percent) small missiles due to increased production runs and competi- tive procurements. Prices for large missile components decreased 4 percent in CY 1973 but increased 3 percent in CY 1974. The IPD for other durables increased 5 percent in CY 1973 and 7 percent in CY 1974. Among the major components, the IPD for new ship construction (approximately 13 percent of durables) increased 13 per- cent and 17 percent respectively whereas the IPD for aircraft spare parts and modification kits (12 percent of durables) increased by a more moderate 3 percent annually. All other major components recorded price increases that approximated the average for other durables. In the period CY 1974-77, the IPD for DOD purchases of durable goods increased at a much higher rate than in the CY 1972-74 period. Unlike the CY 1972-74 period, annual increases were recorded in eyery major category. The IPD for aircraft components rose even though there were some price decreases associated with second and third production contracts of new weapon systems. Further price decreases for Army small missiles components were more than offset by increases in the other military services' small and large missile programs. The ship construction IPD increased over 11 percent in CY 1975 and almost 7 percent in CY's 1976 and 1977. The combat and noncombat vehicles IPD increased 12 percent in CY 1975 and almost 22 percent in CY 1976 because of higher priced deliveries under new contracts for several types of vehicles. Nondurable goods Purchases of nondurable goods account for approximately 7 percent of annual DOD purchases. Over 90 percent of nondurable purchases are for bulk petroleum products, ammunition, subsistence, and clothing and textiles. The IPD for DOD purchases of nondurables increased over 63 percent in the CY 1972-74 period (over 44 percent in CY 1974) but only 30 percent in the CY 1974-77 period. The large increase in the defense IPD for nondurables in the CY 1972-74 period was the direct result of significantly higher prices paid for bulk petroleum products (50 percent of nondurables), which rose 169 percent (127 percent in CY 1974). The IPD for food items (11 percent of nondurables) rose 12 percent. due to higher prices for meat, poultry, and dairy products. Prices for clothing and textile items 15 (10 percent of nondurables) were 25 percent higher in CY 1974 than in CY 1972 as price increases were recorded in every major subcategory, and prices of ammunition components (20 percent of nondurables) rose about 19 percent in the CY 1972-74 period led by a 31 percent increase in components produced or assembled in Government-owned, contractor- operated (GOCO) plants. A significant decrease in the rate of increase in the IPD for non- durable goods in the CY 1974-77 period occurred as prices for every major subcategory of nondurables except ammunition showed smaller increases. The IPD for petroleum products rose only 25 percent and the increase in the IPD for clothing and textiles and food items was 12 percent and 14 percent respectively. Although ammunition prices (especially for fuzes and explosives) increased significantly (33 percent), these increases were coupled with an almost 50 percent decline in purchases; thus the overall impact on the nondurables IPD was relatively small . Services Services accounted for between 21 percent and 23 percent of annual DOD purchases during the period CY 1972-77. The services category consists primarily of contract R. & D., travel, transportation, utilities, contract maintenance, communications, and medical services. Contract R. & D. -- about 35 percent of total services -- is the largest single subcategory. Prices for DOD purchases of services increased 16 percent in the period CY 1972-74 and almost 33 percent in the period CY '1974-77. During this time the IPD for services in the private GNP increased 15 percent and 24 percent respectively (table 5). The increase in the DOD IPD in the period CY 1972-74 for services was led by utilities (42 percent increase), transportation of things (27 percent), and other services (25 percent) such as engineering and technical services, indirect hire, and computer services. Most of these increases took place in CY 1974 as energy costs rose due to higher petroleum prices. Annual increases in the IPD for services remained high in the period CY 1974-77 as higher energy costs continued to drive up prices for all services. During this period, prices of utilities again paced the total increase. These prices rose over 51 percent largely due to higher electricity prices and natural gas rates. R. & D. prices rose over 35 percent during this period due to higher contractor costs for labor, materials, energy, fringe benefits, and other overhead charges. Although many R. & D. contracts allow for escalation which permits contractors to pass on increases in labor and material costs, many other types of contract services do not; therefore, price increases in these categories lag because price changes can be reflected only when contracts are renegotiated. Price increases for all other subcate- gories approximated those for total services except for medical ser- vices which recorded an increase of over 64 percent and communications services which increased only 17 percent during the period. • 16 ■»-> o 13 T3 o s- D_ , — 03 c o •r* +J 03 z to to o s- cc (U 4J 4-> rO O > Z3 •r- "O S- O Q_ S- O- © S- o O <4- r— 4- o II CM to a> I s *. s- Q-CTi O >> i — -»-» 1— 03 >- j— i- c_ 4- o 0) ■•■j Q 03 CO o >> •I-" XI s- D_ -P •r - O ID Qi .a 03 to CO S- O I s *. CsJ i— oo cn CM CT> CM I s *. CM ID \D I s *. zs 4-> O CTt U"> CT> LO cr> CO CO O CM LO CO i— CO O OOPOfTl-lfl CT> CD O © O O r— r— 3 1 — 1— r— i— r^" i — S- -(-> oo to CO o ONNr-OkD CO to CO CO «3" CO CO CO •p— > O *3" *d" O i— CM CO LD I s *. s- O O i — CM CO «-J- CftOlOO O O O O Ct) i— r— i— r— r— i — 1/0 CO jQ V) o i — i — un I s *. O CO CM ** CM CO O CO 03 XJ • • • • • • • • S- O o r** s o cm to cncnor- CO tO CT) CM 3 O O O i— CO CO CO CTi CD O O O O o «— T3 CXi r-~ r— r— i— t— r— E o z CD < — to o to co lo to i — CO CM CM CD ID CD CM CT> -Q -O 03 O Or-00r-00*l- cn o o cx> O O CM CM S- O O O O CM CM CO cr> o o s . O O O O 3 CD r— r— i"~ f~ r—" i~~ Q • • • • • . • • • • • • • • - • • . . . . . • • • • • * • • • ■ • 1 — 1 1— 1 I — 1 I — I > 1 1 1— 1 > • ••••• CM CO CT> CTi CT> CT) s . cr> 1 — 1 — >->->->->->- 2- >- CJ> (_> CJ) O O O C_> c_> CO (J S- E 4- •<- O V) 3 +J CO e Co -t-> E E +J O) S- $- 03 S- D. 3 CO C_> Q 4- • o oo • >> Z3 CO > CO s- .E 3 •<-> oo >> CO jd _e +-> "O CD E _E •«- to •i— « i — to -Q T- 3 tO Q. >. r— tO 03 CO E +-> «=C 03 E o •i— *r~ +J E to O CO E o E <-> O UJ i- i- 03 4- 4- Qi O >- "O CO 3 S- > 03 03 •i- CO -a s- s- E CO 3 CO Q 0Q r— 03 1 (_> CO II +J o >- "ZZ O 17 4-> O 3 o S- CL , — CO c o •r™ +j ro ■z. oo V) o -o J- cd o 3 a) •I" +J +-> fO c > o ■r- u S- D_ 1 O S- +-> O o o i — M- 3 II "O CNI l/l O I""- &_ S- CD O o_ i — +-> R3 <+- >■ , — o CJ <*- CD O) o D. QJ r— u ■r- S- S- O O. re +J S. •^ u >^ •p- -Q CD 00 O) s- 3 u 3 S- +-> 00 iBOiaro CD >=3- X) CNJ O O CO CD CO CO O VD LT> LO CO V£) co CD o > s- CD 00 CNJ CD CNJ CNJ CNJ i— O i— r-- co >— <3- o id o r^ O CNJ U3 CD r— co un r-- CNJ CNJ CNJ CNJ CD i— sf CNJ CO CO CO CD i— «3" Lf> CO ^ «3" *3" Nondurable goods OlOr-N CO O LD O ro^D^N CD CO LD «^- ^coon i— r- CNJ CNJ r** co i— co CNJ CNI CO CO CNJ CNJ CNJ CNI CO CO CO CO CO CO r-- CO CO CO CO CO Durable goods CNJ CNJ r^, LD LD CNJ LD CO co co r-~ co O CO O X) «d" CD CO O O O i— r^ i— cnj <3- i— CNI CNJ CNJ ID r++ CD i — CNI CNJ CNI CO CNJ CO ID IT) CO CO CO CO 1—1 1— 1 i— I t— 1 en 5- C_> ID CD >- o CD >- o CD >- CJ> QJ U S- QJ E • E oo O 00 C_3 CL> c <+- -r- O oo 3 +-> CO c OJ +-> E c +-> 0J i- s- ) =3 QJ > QJ S- .c 3 +J oo >> 0) -Q .C , 4-> "U QJ C x: •■- 00 •i— « i — oo -O «r- 3 00 n. >- ^- 00 - -o QJ 3 i- > ra f0 •i- QJ -a S- '*- c QJ 3 QJ o ca ^~ O >• sz O Structures Structures accounted for between 2 and 3 percent of annual DOD purchases of goods and services during the period CY 1972-77. This category consists of new construction and purchases of existing structures, including various types of military construction and family housing. The IPD for DOD purchases of structures increased almost 50 percent during the period CY 1972-77, whereas the private GNP IPD increased about 60 percent. In general, military construction and family housing are purchased on long-term, firm fixed-price contracts; therefore, prices paid by DOD in any time period may not reflect the current market for con- struction labor and material costs. This was a factor in the more moderate rise in the DOD IPD for structures in CY 1975 -- 7 percent increase compared with that of 11 percent for the private GNP IPD for structures. Most new construction put-in-place in CY 1975 was under contracts let before the energy crisis. In the CY 1976-77 period price increases experienced by DOD were closer to these for the entire economy when more construction was put-in-place under new contracts. Compensation Compensation accounted for 46 percent of total DOD purchases in CY 1977 down from 48 percent in CY's 1975-76, 47 percent in CY 1974 and over 49 percent in CY's 1972-73. Compensation of employees -- about 60 percent for uniformed military personnel and 40 percent for civil- ian counterparts -- consists of wages and salaries and supplements to wages and salaries, such as DOD contributions for Federal old age, survivors, disability, and hospital insurance; Federal civilian retire- ment; and workmen's compensation. The IPD for compensation increased between 5 and 7 percent annually between CY 1972 and 1977. These increases generally reflected the salary increases of military and civilian general schedule (GS) employees although other civilians (e.g., wage grade) received much higher increases. Outlays IPD's have also been prepared on an outlays basis for each major appropriation category of the DOD budget (table 6). The differences between outlays and purchases are basically differences in coverage and timing (see Part 3). The IPD's for each category include only transactions between DOD and non-DOD sources; they exclude intra-DOD transactions, such as with revolving funds. Similarly, the IPD's for revolving funds include purchases by these funds from non-DOD sources. Therefore, caution should be used in applying these deflators to outlays in the various appropriations categories that include a large number of intra-DOD transactions. However, the total IPD is appro- priate to deflate total outlays. 19 >> ro a> ai a c 14- > w— +-> > c a> a; a: f= +-> t- s- m 4- a. O) -0 c ro O s- O >>-— 4- S- II O CM to cnr~- C- 1— +-> ro ro >- ■ — C_J M- c ai a •1 — +J ai ro < i •1— S- S- D- Q O S- +-> D- •1— CL U T3 r— C 00 Cn 00 "* co O 3 CM ro ^~ CO > 4_ 0) q: o3 Ol c C -r- O CO >>•!- 3 S_ +-> cn CM CT> en ro <_> x: -(-> 3 CM 1 — ro CM *i- •r- S- >, 1 — CM ro "* co 1 — +-* r— 1 — 1 — r— ■ — 1 — H- ••— CO •!— s: c e ro 4- LU CO CO CM Ol O e>3 (— "3- CM 1^ t"-. CO Q r— CM ro "3" a: +j c ai E ro CM CD r— 1 — 01 s- CO O 00 O 3 r— 1— CM *d" U O 1- CL. 00 ai e c ro r-. 00 CM O cn •■- c +> ai r- CD CTl O ro +J O r— CM ro en s- c CD -i- Q_ ro O E >> ro Q. ro O r— CM r-» XI i QJ i- E cn CM CM CM ro c +-> O CO ro 01 > ro 4-> «* en CD CM CO 3 8 CD CD CO <* ro O •— CM ro "3- ro +-> O \— CM ro ** LT) CO i~~ r- r-. 1^ r^. r^. r-^ 0-1 Cn en Ol en cn >- s- >- >- >- >- C_> Q. i- o CD ai +J ro S- a. • S- CO Q- c Q. ro • r— +J S- ro O 'I- CU , — S- -0 CJ ro c ■0 S- ■ 1— c a; 20 Part 3: Framework The principal objective of this study was to provide quarterly measures of DOD purchases in constant prices within the framework of the NIPA's. In the NIPA's, Government purchases of goods and services consist of the compensation of Government employees, purchases from business and abroad, and net purchases of used goods. Transfer pay- ments, grants-in-aid, net interest paid, subsidies, and sales and purchases of land and financial assets are excluded. For this study, the definition was restricted to DOD purchases from all non-DOD sources. Current- dollar purchases and outlays DOD purchases on a NIPA basis will differ from other published measures of DOD spending, such as budget authority or outlays, because of coverage and timing. Coverage differences consist of DOD outlays that do not appear as NIPA purchases. These differences amount to $8.1 billion in fiscal year (FY) 1977, of which $8 billion was military retired pay. Timing differences occur because purchases are largely recorded on a delivery basis, while outlays are recorded on a cash basis. Except for purchases of new construction and ship building, which are recorded in the NIPA's on a work-put-in-place basis, goods on which progress payments are made are not recorded as defense purchases until the goods are delivered to DOD. For example, progress payments on a particular aircraft are recorded as DOD outlays. While the payments are being made, the NIPA's will show an increase in private sector inventories equal to the amount of work actually done on the aircraft. When the aircraft is delivered to DOD, Government purchases are increased by the amount paid for the aircraft, and private sector inventories are reduced by a corresponding amount. Timing differences amounted to about $60 million in FY 1977. In periods when procurement activity is relatively stable, these differ- ences will be small; however, in periods of rapidly increasing or decreasing procurement activity, these differences can amount to several billions of dollars. Revolving funds. -- DOD outlays from appropriation categories may not flow directly to the private sector of the economy, because in many cases transactions are made with revolving funds and then with the private sector. DOD revolving funds consist of stock funds, industrial funds, management funds and trust funds. Management funds are principally concerned with financial trans- actions and make no purchases themselves. These funds were generally "created to simplify the financing and accounting for operations supported by two or more appropriations. "4 Since these funds operate as financial clearinghouses, they do not affect DOD purchases. Many trust funds deal with gifts, bequests, etc., from private persons to 4. The Budget of the United States Government , Fiscal Year 1978 , Appendix, pp. 278-79. 21 the various military services and with the surcharge collections from sales by commissary stores. Expenditures by these trust funds are classified as Government purchases and are included in the appropriate categories. Other trust funds, such as Foreign Military Sales and Military Assistance Program are defined as outside the scope of this study. Stock funds were created to finance the acquisition of inventories of consumable materials and supplies for resale to the military ser- vices and other authorized customers, such as other Federal Government agencies. In addition, the stock funds finance inventories of con- sumable materials for use in case of mobilization. Stock funds do not receive annual appropriations but are authorized to incur obligations in anticipation of future sales as purchases are financed by the sale of goods. The value of a Government purchase is recorded in the NIPA's when the good is bought from the private sector of the economy. This is not necessarily the same value (price) at which the good is subse- quently sold to a particular military service. The good is sold at a "standard price" that includes certain charges in addition to the purchase price. The stock fund standard price is calculated from the most recent average cost of the good, plus estimated first and second destination transportation costs, plus a percentage for loss and damage. The costs of warehouse maintenance, utilities, or personnel are not included in the price, but are included in the operation and maintenance appropriations of DLA and the military services. DOD stock fund purchases on a NIPA basis are recorded at the time the good is delivered to DOD. A subsequent purchase of the good from the stock fund by a military service is not treated as a Government purchase as it is an intra-DOD transaction. The difference in timing between delivery to DOD and subsequent resale can be considerable. Industrial funds finance activities of an industrial or commercial nature. These activities are performed on a reimbursable basis, vary in nature, and range from operations of the Military Airlift Command (MAC) and Military Sealift Command (MSC) through depot maintenance services, printing, scientific research, and operations of arsenals. While many of these activities are performed in Government-owned, Government-operated (GOGO) facilities some are subcontracted to the private sector. Activities financed by industrial funds receive orders from the military services for goods or services and then are reimbursed by these customers by means of progress payments for the full production costs of the goods and services rendered. These transactions are internal to DOD and are therefore not treated as Government purchases. However, transactions between these activities and the private sector are treated as Government purchases. For example: An Army purchase of a 155mm howitzer from a GOGO arsenal is recorded in the NIPA's as the purchase of employee services (wages and salaries, benefits, etc.), materials, utilities, etc., required to produce the howitzer. Thus, the inputs to industrially funded activities are recorded as Government purchases. 22 Contractor- operated faa-ili ties. -- In contrast to GOGO facilities, GOCO facilities are classified in the NIPA's as part of the private sector; therefore, a DOD purchase from a GOCO plant is a Government purchase. The unit price of an item purchased from a GOCO plant represents only the production costs incurred by the contractor, e.g., labor costs, material costs, incentive fee; expenditures for Govern- ment-furnished equipment CGFE) and Government- furnished materials (GFM) are not included because they are separately stated costs to the Government and not to the contractor. This definition of price is in contrast to that for items purchased in the private sector: For the latter, the unit price represents a "market price" that includes all costs. This is the reason that "like" items purchased from both GOCO and private sector plants are treated as different products and not as the same product at different prices. Summary. -- In summary, DOD purchases on a NIPA basis in current- dollars will differ from DOD outlays because of certain timing and coverage differences. In addition, expenditures of the stock funds and industrially funded activities are treated as DOD purchases but the sales from these activities to the military services are not. Also, GOCO facilities are treated as being part of the private sector; therefore, purchases from them are treated as DOD purchases. Constant- dollar purchases The physical volume of DOD purchases over time is measured by valuing the purchases in constant prices; that is, the quantity of goods and services purchased in any given period is valued at its price in a base period. Thus, the effects of price change are eliminated. To prepare measures of constant-dollar DOD purchases, price series for the components constituting the total must be constructed. The technique for measuring price changes over time is "specification pricing." This technique measures the "pure" price change in an individual good or service and is the same procedure as that used in the Consumer Price Index (CPI) and the Producer Price Index (PPI). 5 Essentially, specification pricing involves defining the price- determining characteristics of an individual item and holding them constant over time. Price-determining characteristics are those that influence the price of an item. For example, for a combat boot, the type and quality of sole and upper material may be price-determining characteristics, while boot size and color may not. For each period, the price of an item with the specified price-determining character- istics is divided by the base price for that item. This results in a price relative. The price relative for a given period is divided into the current-dollar value of purchases of that period to y,ield constant- dollar purchases. Thus, the constant-dollar series for an item reflects the actual quantities purchased over time valued at base period prices. 5. The PPI was formerly known as the Wholesale Price Index. 23 Unfortunately, price-determining characteristics of a purchased item do not remain constant over time, because of changes in military needs, changes in technology, or other reasons. Thus it is often impossible to reprice an item with a given specification. In these cases, the price of the changed or replacement item is adjusted for any difference in quality. Quality change The measurement of quality change is the most difficult aspect of developing price indexes. In order to develop series consistent with other price indexes used in deflating the NIPA's (e.g., PPI), the "performance/cost-of-production" technique 6 was used to measure quality change. This technique assumes that a change in performance is a quality change and that the production cost associated with the change is the appropriate measure of the value of the quality change. In other words, if a physical change affects the performance of a given item, the production cost of that change is defined as the value of the quality change. In the example of combat boots, if a change in the sole from leather to rubber lengthens the life of the boot, the cost difference between the leather and rubber sole is the value of the increased quality and is not treated as a price change. This is the same technique as currently used in the PPI. In general, for goods, quality changes are all physical changes that affect cost and performance. For services, quality change includes all work statement changes that affect cost and performance. A quality change to a good or service is assessed by its effect on DOD's mission: If the change helps the mission, it is a quality increase; if it detracts it is a quality decrease. Any change not affecting the mission is a pure price change and does not involve quality. For example, a physical change that increases the maneu- verability of a fighter aircraft is a quality increase, and the price would be adjusted. However, more elaborate decorative paint on that same aircraft would not affect the mission and, therefore, would not be a quality increase. It is important to remember that the valuation of quality change is based on the cost and not the usefulness of the change. Even though a particular change might make the good twice as useful, the quality change would be valued as the cost of making the good twice as useful, rather than doubling the value of the good. Two special cases of quality change are: (1) "costless quality change" and (2) a quality increase at a decrease in cost. The concept used in this study, that of valuing quality change on the basis of cost, renders a costless quality change impossible. If the quality of a good changed with no associated change in cost, the valuation of the change would have to be zero - no quality change. In the second case, a good that has a quality increase associated with a cost decrease, is 6. For a discussion of various quality adjustment techniques, see Measuring Price Changes of Military Expenditures , p. 41ff. 24 assumed to have a price decrease (ceteris paribus) equal to the cost decrease. This assumption results in zero quality change. If this were not done, quality would be assumed to be decreasing because the real value of the good would be reduced. Implicit price deflator The IPD is the ratio of purchases in current-dollars and purchases in constant-dollars; it is often used as an indicator of inflation. It must be remembered, however, that the IPD is a price index of a different formulation than the PPI or CPI, which measure only price change between points in time. These indexes are based on the Laspeyres formula and use fixed weights/ i.e., the same market basket of goods and services is repriced. Thus, any change in the index number repre- sents the price change of a fixed set of items. The IPD is based on the Paasche formula and uses changing weights, which reflect the current market basket. Thus, between any two time periods other than the base period, a change in the IPD can arise from a shift in weights as well as from a change in price. The inclusion of the current DOD purchasing pattern as weights in the defense IPD means that it can be interpreted as measuring the purchasing power of the DOD dollar. Thus, the defense IPD reflects more than just price change, it provides a measure of how the DOD dollar has changed from the base period with regard to the amount of goods and services it can purchase. Comparison with gross national product Comparisons between IPD's for DOD purchases and for total private GNP provide some insight into the differing impact of price change on DOD purchases .and on the private economy. In these comparisons, however, the treatment of Government compensation - Government pur- chases of employee services - requires special considerations. Because the NIPA concept of real compensation (see part 4) does not allow for changes in productivity of Government employees, these purchases are different from other purchased services where productivity can change over time. For this reason, the exclusion of the compensation of Government employees makes the subtotal "DOD purchases less compensa- tion" a more consistent comparison with private GNP. Estimates of current- and constant-dollar GNP are presented by major type of product in four groups: (1) durable goods, (2) nondurable goods, (3) services, and (4) structures. DOD purchases have been presented in this manner also, with compensation of Government employees shown separately. Following are brief definitions of the product groups in this study. Durable goods. -- This group consists of items with a normal life expectancy of 1 year or more. Included are such items as weapon systems, capital equipment, and radio receivers. 7. A discussion of the Laspeyres, Paasche, and other index formulae is contained in Measuring Price Changes of Military Expenditures , p. 14ff. 25 Nondurable goods. -- This group consists of items with a normal life expectancy of less than 1 year. Included are such items as food, clothing, and drug preparations. Services. — This group consists of intangible commodities. Included are such services as medical care, transportation, repairs, and utilities such as electricity and natural gas. Structures. -- This group consists of new construction and purchases of existing structures. Compensation. -- This group consists of wages and salaries of Govern- ment employees and supplements to wages and salaries, such as DOD contributions for Federal old age, survivors, disability, and hospital insurance; Federal civilian retirements, and workmen's compensation. Outlays deflators IPD's have also been prepared on an outlays basis. Differences between total outlays and total purchases IPD's arise only from the previously mentioned coverage and timing differences. Although the price of an individual item is the same in both series, the time periods in which there were transactions at that price may differ. However, the measure of price change over time for that item will be the same in both series. IPD's were prepared for outlays classified by major appropriation categories. These categories are aggregations of the 60 that are presented to Congress in the DOD budget. The aggregations were based on similarity of composition. The categories presented in this study are: Military personnel Retired military personnel Operation and maintenance Procurement Research, development, test, and evaluation Military construction and family housing Definitions of these categories appear in appendix C. The IPD's are statistically reliable for these categories only. Future expansion of sample size and coverage (discussed in part 5) will result in deflators for more detailed categories. Pricing was undertaken only for DOD purchases from the private sector; intra-DOD transactions were excluded. Thus, the IPD for the military personnel category, for example, consists only of travel and compensation of military personnel. Outlays to stock funds for clothing, food, etc., are intra-DOD transactions and are not deflated; however, stock fund purchases of clothing, food, etc. are deflated as part of the revolving funds category. Therefore, the IPD's measure price changes of outlays by major appropriation categories and by the revolv- ing funds to the private sector. The aggregation of these series, when properly weighted, will equal the IPD for total outlays by DOD to the private sector. 26 Part 4: Development DOD purchases a wide variety of goods and services in addition to the commonly perceived items such as aircraft, ships, tanks, and guns. For example, DLA alone is responsible for more than 1.7 million separate products, none of which is a weapon. The large number of items made it impossible to price e^ery good or service purchased by DOD, therefore, samples were selected for pricing. Selection of the samples first required the exact determination of what DOD purchased and in what quantities. Purchases in a given period were divided into broad categories, and then into classes of similar products or services. The prices of selected items from each class were used to estimate the price change for the entire class. Before the procedures and data collection methods are discussed, some of the constraints on the project will be considered. Data constraints First, whenever possible, data were to be obtained from within DOD in order to alleviate the paperwork burden on private industry. In addition, the BEA staff was insufficient to collect large amounts of data from the private sector. Furthermore, the use of DOD records would make it possible to determine the exact price at which trans- actions were actually made; a s/ery important factor in developing accurate deflators for DOD purchases. For ship construction and contract research and development, appro- priate data were unavailable within DOD; therefore, private contractors were requested to provide this information. Additional data sources would have been helpful in several other categories; however, neither the resources nor the time were available to implement direct collection Second, DOD's reporting burden was not to be excessive. There were some cases where the extraction of the desired data would have placed an excessive burden on DOD personnel. Because project staff were unavailable to spend the time necessary to extract the data, the number of items priced and the type of pricing were reduced. These two constraints limited the length of the time series which could be constructed. Moreover, because detailed DOD records are not kept indefinitely, reasonably complete data were available only for FY 1974 and more recent periods. In order to put DOD purchases on the same constant-dollar base as the NIPA's (CY 1972), available data were supplemented by selected series from the WPI and CPI. When permanent funding is available for this project, summary techniques will be used to prepare historical estimates (quarterly from CY 1946 through CY 1971) consistent with the NIPA's. Third, data with a security classification were not to be used. The project staff adopted this constraint so that detailed results could be published. As it turned out, the dollar value of purchases covered by this constraint was not large enough to seriously affect the total or any intermediate level of data intended for publication. 27 Procedures The steps used in constructing the constant-dollar defense purchases were: • Define work units • Specify concepts • Select sample • Develop weights • Develop prices • Adjust for quality change These steps and their general application to the product or service categories are briefly examined here and discussed in more detail in appendix A. Define work units. -- Work units were defined by grouping individual products and services into categories, thereby reducing the number of price series necessary to provide reasonably accurate measures of price change. The overriding criterion for this grouping should have been similarity of price movement; however, there was no historical information on price change. Therefore, categories were based on similarity of factor inputs, similarity of production process, simi- larity of end use, or expected similarity of price movement. As a result, defense purchases were grouped into 22 categories: Aircraft Ammunition Communications services Compensation Construction Depot maintenance Electronic equipment Installation support services Medical services Missiles Other equipment Other services Research and development Retired military pay Ship construction Stock funds: defense stock fund Stock funds: military services Transportation of things Travel and transportation of persons Utilities Vehicles Weapons 28 These categories, or work units, provided manageable groupings for the development of the data. Purchases by industrially funded activities (e.g., GOGO facilities) were not treated as a separate group. Pur- chases by these activities were included in other categories, e.g., compensation of employees, utilities, rents, services. DOD data for these categories were generally obtained from: (1) financial analysts - expenditure data, (2) program or item managers and contract specialists - price and delivery data, and (3) technical staff - quality and specification change evaluations. Specify concepts . -- Concepts of price, quantity, and quality had to be specified for each category. A variety of contractual techniques are used by DOD for purchasing goods and services and it is not always obvious what is the unit of purchase, the price paid, or the criteria for quality. Thus it was necessary to define these in each sector in a manner consistent with the overall concepts of the NIPA's. Select sample. -- Within each category, a sampling frame was constructed, The sampling frame is a list of the items structured in a format appropriate to the sampling technique to be used. Expenditures for each item (goods and services) in each category were obtained and subclassified by similarity of item (and in many cases, by military service). The sample of items to be priced was selected from within these strata (subcategories), which varied in number from 5 to over 100, depending on the nature of the category and the availability of data. The sampling technique that was used most often is "selection on the basis of probability proportional to size (PPS)," as it leads to statistically unbiased estimates. In addition, a PPS sample produces a more statistically accurate result than a simple random sample with the same number of observations. Thus, for categories broken into > many strata, it was possible to select and price only a few items within each stratum. The probability of selection of a given item was directly proportional to the dollar expenditures for that item within a stratum. The number of sampled items necessary for a statistically reliable estimate of price change depends on the variance in price movement among items in the stratum. The division of the categories into strata provided even more similarity among items and helped reduce expected variance in price change. Moreover, since historical infor- mation on possible variances were unavailable, samples were selected so that at least one-half of the base period purchases of the stratum were included. Thus, even if the variances were fairly wide, reason- ably accurate estimates could be prepared. Develop weights. -- As noted previously, in the NIPA's the IPD is a Paasche formula index in which the weights reflect the quantity of items purchased in the given time period. It was necessary, there- fore, to develop quarterly weights for each sampled item. In general, quarterly expenditures for each sampled item were derived from contract control records and used for weights to the stratum level. (This procedure assumed that price changes of nonsampled items moved like 29 the sampled items.) Total expenditures for each stratum, which were generally obtained from financial records, were used to weight the strata to the category level. Expenditures data are inappropriate weights for items which are included in DOD purchases on a delivery basis and for which progress payments are made. Therefore, the value of deliveries was developed for sampled items by multiplying the quantity delivered in a given time period by the price. These weights were then aggregated to the total level. (This procedure, again, assumed that the prices of sampled and nonsampled items moved similarly.) The results, there- fore, properly reflect the changing pattern of DOD purchases by pro- viding changing weights for each price series throughout the time period under study. Develop prices. -- In general, price was measured from the point at which DOD took delivery of the good or service. This was the trans- actions price, the one actually paid by DOD for the particular good or service. The price was usually the final contract price, which reflected all modifications to the contract. Considerable care was exercised to ensure that the latest price was used and that the price was applicable to the delivered quantities or expenditures. In pricing, it is essential that the data collected relate to the same specifications over time. All relevant price-determining charac- teristics must be held constant at the detailed pricing level. For most goods with a military specification, this proved to be relatively easy because all specification changes are documented. Purchased services were more difficult to examine for specification changes, but usually the changes could be determined. Changes in goods or services were evaluated and prices were adjusted for any changes in quality. Adjust for quality change. -- In most cases, quality changes involve physical changes to a good or procedural changes in a service. These changes are usually referred to as engineering changes and are recorded in an engineering change order (ECO). All ECO's on a given item were evaluated to determine if they affected the performance and cost of the item. Most ECO's did not affect the cost or the performance of the item. Costs incurred to remedy a defect in an item were not considered to be quality changes, as it was assumed that DOD contracted for a properly functioning item.° For ECO's that did affect perfor- mance and cost, it was necessary to determine if the change was a quality increase or decrease, and the cost associated with that change. The price was then adjusted accordingly. Quality change in a purchased service usually appears as a change in the statement of work (specification) in the contract. Work state- ments are changed often, but the changes are usually identifiable. In many cases, changes are for additional work to be done at a specified 8. A discussion of quality adjustment criteria for a specific case can be found in appendix A, in the aircraft category. 30 price and can be separated from the original work statement. Changes were evaluated for quality implications, costs determined, and prices adjusted accordingly. Because price changes may accompany specification changes, addi- tional DOD payments were separated into: (1) cost of the quality change and (2) true price change. Quality adjustments were based only on the former costs. Categories Although the preceding procedures were generally applicable to each category of DOD purchases, slight differences were necessary for certain categories, because of concepts employed in the NIPA's or because of the type of data available. Following is a brief overview of each of the 22 categories and a discussion of any procedural differences. Aircraft. -- This category consists of all military purchases of new aircraft, initial and replenishment spares, and equipment for modifi- cations of aircraft already in the DOD inventory. Aircraft, as well as other major weapon systems, are not purchased as a unit by DOD. Rather, engines, some of the avionics, armament, and other equipment are procured separately and then furnished to an airframe contractor as GFM for installation on the airframe. In order to price exactly what DOD purchases, each component is treated separately. This category principally measures price change of airframes and engines; Government- furnished avionics and weapons are priced in other categories, No unusual sampling problems were encountered, and the sample was selected on a PPS basis. Weights for aircraft were developed for both the value of deliveries (NIPA concept) and for outlays. Quarterly data on the value of deliv- ered aircraft were not readily available; however, a value-of-deliveries series was calculated by applying the appropriate unit price to the number of items delivered in a given time period. No serious problems were encountered in developing weights for the outlays deflator. A final unit price was developed, whenever possible, for each con- tract and every procurement option under that contract. When final prices were not available, estimates of the price at completion from contract control documents were used. The price was adjusted for quality change and applied to all relevant deliveries. Many engineering changes occur on aircraft; however, data were available to evaluate the quality implications of these changes and to adjust prices for quality change. The development of an outlays deflator required the assumption that the change in the end item unit price is applicable to each stage of the production process. The price is contracted only for the total unit; therefore, the end item price is applied to all periods where work is in process on the item. The different weights used for the deliveries series and the otitlays series for the same item will not affect the price change of an individ- ual item. In a given quarter, however, the price change of deliveries 31 and outlays may be very different. Aircraft engines, for example, may be delivered under only one contract during a quarter, while at the same time work may be underway on several contract options, each having a very different price for the same engine. Another cause of difference in the level of the IPD's on a deliv- eries or outlays basis is the specific price included in the base period. An example of this occurred in the Navy aircraft subcategory. While initial deliveries of the S-3A were not made until FY 1973, outlays occurred during FY 1972. Thus, in the deliveries series, the S-3A prices did not enter the base period, while in the outlays series they did. In summary, the same unit price series are used in the IPD's on a deliveries basis and outlays basis. Differences arise because, in the outlays series, the prices of items are recorded as the items are being constructed, while in the deliveries series the prices are recorded when items are delivered. Ammunition. -- This category consists of all purchases of military ammunition, including cartridges, bombs, and torpedoes. These items are largely purchased as components also and are assembled on con- tract, primarily in GOCO plants. Thus, there is not a single unit price for a round of ammunition (e.g., Cartridge, 105MM:HEAT-T,M456) , but there are separate price series for primers, projectiles, fuzes, propellants, etc. Although purchases from GOCO plants are treated as end items similar to other purchases, they are not directly comparable to ordinary business purchases because the plant and equipment and, in many cases, maintenance, are furnished__by the Government. The purchase price does not include certain costs--such as depreciation on GFE. Therefore, prices of identical goods purchased from GOCO plants and private business are not directly comparable, and the goods are con- sidered to have different specifications, even if they are physically identical . No unusual sampling problems were encountered, and the sample was selected on a PPS basis. Expenditure data were available on a quarterly basis, and no serious problems were encountered in developing weights. Price data were available, including all contract modifications, and item price series were developed., Data on specification changes were available, and the item prices were adjusted for quality change. Communications services. -- This category consists of interstate and international services purchased by the communications services industrial fund (CSIF) as well as intrastate services, some of which are purchased by the CSIF, but largely purchased by military installa- tions from local telephone companies. Also included are purchases of postage. All interstate and international contracts were directly priced. Expenditures and price data were available for all CSIF contracts and no major problems were encountered, except for the international data. Prices from international contracts did not represent specifications, but included some product mix; better data will be available beginning 32 FY 1977. Intrastate and local expenditures were deflated using the CPI for telephone service. This CPI component was used because of the large number of contracts and the lack of readily available data for stratification and sampling; it should approximate DOD price changes as it includes intrastate tariffs. Quality change presented no major problems; however, it should be noted that a change in the specific type of carrier (e.g., telephone line or satellite) was not considered to be a quality change unless it made a noticeable difference to the user. Most of the recent tech- nological change in communications resulted in the use of satellites rather than long lines or microwave relays. The service purchased by DOD is a communications link between two parties, whether it is a voice link (Autovon) or a data link (Autodin). The gradual change to the use of more advanced technologies has generally kept costs down, rather than changed the quality of service. Data on all DOD postal service expenditures are readily available and indexes easily calculated; therefore, sampling was not attempted and all postal classes were used. Weights were also readily avail- able, and price indexes were combined to the total. No quality adjust- ments were made. Compensation. -- This category consists of wages, salaries, supple- ments, and payments in kind to DOD employees, military and civilian. The concept of Government compensation in the NIPA's is that the unit being purchased is a man-hour worked adjusted for quality change. Quality is measured as a combination of the education and experience of the employee. These two qualities seem to be the major price- determining characteristics of employees in the U.S. economy. The correct base price measure is the base period average compensation for each man-hour worked stratified by education and experience. Current data on education and experience are not available for DOD employees, therefore, a proxy measure was developed. Since data on employment and wages were available for the universe of DOD employees, the universe was used. In addition, pricing was unnecessary because data on the quantities of hours worked could be directly estimated. The GS classification system provides a built-in measure of experience as well as education, because civil service regulations specify the requirements for each position. In addition, promotions to steps within GS-grades reflect additional employee experience. Employment data on GS employees were available by grade and step for each time period. Data on hours worked were not available, but data on "weekly hours paid-for" were. These were applied to the employment data to generate the number of hours paid-for in each quarter. Data were developed on the average hourly compensation by grade and step for the base period and were multiplied by the number of hours in each grade and step to develop the quarterly estimate of hours paid-for valued at base period prices. The wage grade system is very similar to the GS system in that it also is divided into grades and steps. The major series - Wage Grade (WG), Wage Leader (WL) and Wage Supervisor (WS) - in this system were 33 used to represent the entire system. The other series are relatively unimportant, accounting for less than 10 percent of the total wage grade system. The procedures used to develop the quantity of hours paid-for for GS employees were also used for wage grade employees. A similar but somewhat modified procedure was used for military personnel. Employment was stratified by rank and by years of service as a measure of experience. Adjustments were not made for average hours worked since military personnel are considered to be on duty 24 hours a day, except during actual periods of leave. In addition, certain categories of military pay were priced separately. Certain allowances (e.g., basic allowance for quarters, uniform and clothing allowance, family separation allowance) were treated as part of basic pay; changes in rates for these allowances appear as price changes. It was assumed that increases in these allowances do not relate to experience and education and, thus, do not reflect increased employee quality. For this same reason, supplements to wages and salaries and the value of food and clothing furnished to employees were also treated as part of basic pay. However, certain types of pay (e.g., flight pay, jump pay) do reflect additional quality; therefore, expenditures for these were deflated by pricing rate schedules. In summary, the measure of the quantity of compensation purchased is an hour worked adjusted for quality change. For the GS and the wage grade systems, quality is reflected by the grade and step of the employee. For the military, it is reflected by rank and years of service, with adjustments for specialty pay and other such items that reflect increased employee quality. The ratio of actual outlays for compensation to the adjusted hours in base period prices yields the IPD for this category. Construction. -- This category consists of all construction purchased on contract by DOD, including military construction, family housing, and maintenance construction. Acquisitions of existing structures and installation of equipment are included, as is the renovation of Minute- man missile silos. In the NIPA's, construction is measured on a work-put-in-place basis and not on a final delivery (.acceptance) basis. Contract construction, like other long-term contractual DOD purchases, includes provisions for progress payments during the period of construction. These pay- ments will not exactly correspond to the amount of work that has been done at any given point, but they will approximate a large percentage of it. This means that dollar magnitudes and timing of outlays and GNP purchases are yery similar. Construction contracts were grouped into 40 classes; a PPS sample of 10 of these was selected for pricing and then examined for uniformity of construction specifications within the class. Several classes, such as airfields, roads, and troop housing, were further subdivided to make the class being priced more homogeneous. All contracts within a class were not readily available; therefore, expenditures by class were based on a five-quarter moving average of contract awards by class of construction. 34 Construction contracts are written on a firm fixed-price basis; therefore, only changes in contract requirements will change the price per unit. The price per unit (i.e., square foot, linear foot, cubic yard, etc.) was calculated for each contract within a class. The total units for each project were allocated across the construction period from 1 month after the date the contract was let until the date of contract completion. The unit price for each project, weighted by the number of units put-in-place in a given quarter, was summed within the class and an average price determined. Quality change in construction was determined by a change in physical characteristics. Unit price differences in various areas of the country or in different sizes of projects were assumed to be price differences and not quality differences. Rather, the quality of a construction end product was assumed to relate only to its function. In other words, if the structure met the military performance require- ment, it was considered to be acceptable and met the specification. If the performance requirement was changed, the result was a quality change. For example, in troop housing specifications, if the perfor- mance requirements were switched from community bath facilities on each floor to private baths for each room, a quality change in troop housing would result. If, however, lath and plaster walls were used in one facility and gypsum board walls in another, no quality differ- ence would result if both walls met the performance requirements. The program to upgrade the Minuteman missile silos and hardware required a different approach. This program consists of deactivating the missile squadron, removing the missile from the silo, renovating the silo, and installing new electronic equipment in the silo and command post. The construction portion is included in this category and the purchases of new hardware in the electronic equipment category. Since virtually the same operation is being performed on every silo, specification pricing was feasible for the work done on each silo. Any special work required at a particular location comes out of "over and above" funds and can be separated from the price paid for the standard work. This procedure assumes that price changes of "over and above" work and regular work move similarly. In summary, this approach is an attempt to price DOD purchases of the output of the construction industry. It assumes that the final contract price is applicable to all construction stages and that quality change is based on performance characteristics of the partic- ular project within a construction class and not necessarily on the materials that go into the construction of the facility. This pro- cedure is superior to the input method normally used since it picks up changes in capital-labor ratios, labor productivity, and competitive position among contractors and thereby more accurately reflects real prices. Depot maintenance. -- This category consists of contracts with the private sector for periodic maintenance of major weapon systems and their components. Also included are contracts with the private sector for modification of existing aircraft or missiles. Depot maintenance 35 performed in- house by the military services is priced in other cate- gories; e.g., other equipment, stock funds, or compensation of DOD employees. Within depot maintenance, there are two broad subcategories of work: (1) the periodic maintenance and modification of the major weapon systems, and (2) the overhaul or maintenance of components, generally referred to as "exchangeables" maintenance. In general, exchangeables are small items (e.g., gyroscopes, radios, or coding equipment) which, when in need of repair, are removed from the weapon system and replaced with another unit. The item is then repaired for use the next time it is needed. There were no major problems with sampling or weight development for the first subcategory. Pricing was possible even though in some cases it was rather difficult to specify the exact service performed. In many cases, periodic depot maintenance is performed over a long period of time on a given type of aircraft (e.g., the F-4E); thus, it was possible to specify the service performed in a precise manner and trace it over time by using the contract price and specifications. Any additional repairs done at the same time as the periodic mainte- nance are specified in the contract as work done "over and above" the normal work. Prices for "over and above" work were not developed because it was impossible to prepare consistent specifications. This procedure assumes that price changes of "over and above" and regular specification work move similarly. Sampling and pricing for the second subcategory - exchangeables - differed from the normal practice. There are a large number of con- tracts, and each is generally for overhaul or repair of a single small item. In these cases, since the data were automated, the sample consisted of all items with specific stock and contract numbers and which were purchased in 2 consecutive years. For each item meeting these criteria, a price change was calculated. These changes were averaged to develop the aggregate change for the subcategory. While the sampled items were not held constant over time, in all cases only matched items were used to calculate an individual change. The aggre- gate changes were chained together to produce the price index. 9 To keep the matched specifications as consistent as possible, statis- tical tests were applied to the price changes. In most cases, items with a price change more than two standard deviations from the mean were excluded because the changes were assumed to reflect changes in the work statements rather than in the prices. Since there were several hundred matched items in each period, this technique yields a reliable estimate. The chaining procedure consists of multiplying the prior period index by the current period change to derive the current period index. This allows changes calculated from different samples in different time periods to be integrated into a single index. 36 Electronic equipment. -- This category consists of all avionics, communications, and other electronics equipment purchased by the military services. Data were available to construct a sampling frame and a PPS sample was selected. Weight data and price data were also readily available. A problem occurred in deflating purchases of large electronics projects, such as the Air Force's Cobra Dane system, because usually only one unit was produced and could not be repriced. Therefore, the large project was treated as a composite of small electronic compo- nents and the price change was assumed to follow the price change of these components purchased individually. For example, Cobra Dane included several types of amplifiers; the cost of these amplifiers was deflated by using the price change for the other amplifier purchases. The development of data on quality changes for electronic equipment proved to be difficult because, in many cases, specifications for electronic equipment are not changed during its expected service life. For example, specifications for a particular model of a radio pur- chased over a long time will not change, even though technological changes have occurred in the field of electronics. (Maintaining constant specifications makes maintenance easier.) Quality improvements usually occur when a new model radio is being purchased. The statistical problem occurs in linking the new model to the old model, as it is often difficult to tell exactly which model is being replaced. In fact, in some cases, a new model will replace several old models or a larger system of separately purchased compo- nents. It is also difficult to determine the cost of each quality difference. Therefore, in many cases, the new model was treated as a new product and price differences equated to quality differences. The new model was "linked in" with no price change; this procedure is used in many instances in the PPI and CPI. Installation support services. -- This category consists of all con- tractual services, not included elsewhere, which are procured for the operation of military installations. Services ranging from janitorial services and maintenance of equipment other than weapon systems to rental of property and equipment (except automatic data processing equipment) are included. The wide variety of services in this category, the geographic differ- ences in purchasing patterns, and the large number of military installa- tions both within the United States and abroad created considerable problems in selecting a sample. It was difficult to obtain detailed expenditures and to assemble the sampling frame. Data are not avail- able from a centralized location and, as a result, only aggregate financial data have been compiled at this time. Because a sampling frame adequate to the needs of probability sampling has yet to be constructed, a very small judgmental sample of installations was selected to test the methodology. Weights were also wery difficult to obtain, for the same reasons. The detailed expenditures by type of contract service have not-yet been determined. 37 Pricing did not present any unusual conceptual problems. Data from the test sample indicate that adequate prices can be developed from contracts. Data on quality change may prove to be more difficult to develop. Initial findings indicate that contracts change over time and that detailed cost differences may not be readily available. However, one or two test cases indicate that some methods can be used to estimate the costs of quality differences between contracts. In summary, preliminary data indicate that it is possible to develop weight and price measures for installation support services. However, the difficulties of obtaining the data to construct the sampling frame and the weights, as well as the large number of contracts that will be required because of geographic differentials, have delayed the develop- ment of adequate IPD's. As an interim measure, the data from the test sample were combined with data on average earnings in the service industries to develop a proxy measure for price change. Medical services. -- This category consists of all medical services purchased by DOD from the private sector. Purchases for the operation of military hospitals, such as materials, supplies, equipment, and wages and salaries of employees, are included in other categories. The principal element of purchased medical services is the Civilian Health and Medical Plan of the Uniformed Services (CHAMPUS), a health insurance program operated by DOD. The program covers retired military personnel, their dependents, and dependents of uniformed personnel located more than 40 miles from a military hospital. When the need arises, those covered may visit a participating physician or hospital and submit a claim's form to the CHAMPUS office and DOD will pay a prescribed percentage of the fees. Thus DOD is not actually providing medical care, it is acting as an insdrer for the users of the service. In the NIPA's, however, this is treated as a DOD purchase of medical care. The expenses of administering claims are also treated as purchases. Purchases in this category were subcategorized into: (1) expenses of processing claims, and (2) purchases of medical care. Since the processing of claims is contracted to the private sector by DOD, the data were available for weights and prices. The average cost for processing a claim was treated as the price for this subcategory. Purchases of medical care were classified into 31 classes, e.g., pre- scriptions filled, inpatient and outpatient doctors' visits, daily hospital charges. Resource constraints and the lack of easily avail- able data prohibited further subclassification. All 31 classes were included in the sample. Weights were available at this level of detail. Prices were obtained by dividing expenditures in each class by the number of units purchased. This procedure assumes that it does not matter why a person visited a doctor, but that a doctor was visited, and that it does not matter what kind of prescrip- tion was filled, only that a prescription was filled. Since any change in average unit cost was due to a price change, no quality adjustments were made. 38 Missiles. — This category consists of all military purchases of guided missiles, both the small missiles (e.g., air-to-air, surface- to-air) as well as the large intercontinental ballistic missiles. Modification equipment for missiles, spares and repair parts, support equipment and facilities, and equipment for the military space pro- grams are also included in this category. Excluded are the missile components produced or assembled by industrially funded activities and the Minuteman force modernization program. Deflators for missiles, like those for aircraft and other weapon systems, were developed on a delivery basis as well as on an outlays basis. The method for deriving estimates and introducing new products into the series were the same as those used for the aircraft category. All active missile programs and the military space programs were selected for pricing. Weight data were readily available for these areas. Pricing presented few problems for the missile programs as component prices were readily available from contracts. However, the space programs were difficult to price because specifications could not be determined; the amounts involved in any particular program were rela- tively small, and the work was highly experimental. Expenditures, therefore, were deflated by the R. & D. deflator. Data on quality change were available for missile programs. All ECO's were documented and, in most cases, costs of these changes were available. The space programs again presented unique problems, but since they were not priced, they were not adjusted. Other equipment. -- This category consists of all military equipment purchases not included in other categories. It primarily consists of special types of equipment, such as pontoon bridges, shipboard equip- ment, and airfield lighting equipment. Data were readily available to construct the sample frame, and a PPS sample was selected. Weight data were also readily available, and no unusual problems were encountered in deriving weights. Pricing did not prove difficult since specifications are relatively constant over time. Quality adjustments were feasible as all changes are made through documented ECO's. No unusual problems were encountered in developing the quality change estimates. Other services. — This category consists of purchases of contractual services not included elsewhere. It primarily consists of business related services, such as engineering services, printing and repro- duction, automatic data processing (ADP) equipment rental, ADP ser- vices, education services, and technical service purchases. The wide variety of services included in this category and the large number of different DOD installations making these purchases created problems in sample selection. Data on detailed expenditures were not available from a centralized location; and as a result, only aggregate financial data were compiled. An adequate sampling frame could not be assembled; therefore, a very small judgmental sample of contracts was selected. For the same reasons weight data were also difficult to derive. Estimates of expenditures by type of contract service have not been developed. 39 Pricing did not present any conceptual problems, and data from the test sample indicate that adequate prices can be developed from contracts. Data on quality change may be difficult to develop: Specific ser- vice requirements change over time and detailed cost differences may not be readily available. One or two test cases indicate that some methods can be used to estimate the costs of quality differences among contracts. In summary, preliminary data indicate that it is possible to develop weight and price measures for these contractual services. However, the difficulties of obtaining the data to construct the sampling frame and the weights, as well as the large number of contracts that will be required, have delayed the development of adequate IPD's. As an interim measure, the data from the test sample were combined with data on average earnings in the service industries to develop a proxy measure. Research and development. -- This category consists of contractual purchases of research, development, test, and evaluation from the private sector. R. & D. performed at industrially funded activities (DOD laboratories) is priced in other categories, e.g., electronic equipment, other equipment, other services, and employee compensation. The output of an R. & D. activity is virtually impossible to define, much less to develop a specification for pricing. It is impossible to fix either the type of experiment or type of work being done to allow the development of a measure of the output of R. & D. It is therefore necessary to resort to "input pricing" for this category: Prices paid for inputs (e.g., compensation, materials, services) properly weighted by expenditures, are used to represent changes in prices of outputs. This approach is unsatisfactory in that it is very difficult to measure accurately changes in certain overhead items, such as depreciation and profits and changes in the productivity of labor. The composition of direct labor, direct materials, and overhead vary widely among R. & D. contractors. While direct labor and materials presented no conceptual problems, overhead proved less tractable for defining a constant-dollar measure. The assumption was made that the "output quantity" of R. & D. was directly proportional to the quantity of direct labor and materials consumed in any given quarter on R. & D. In other words, since constant-dollar overhead charges are fixed relative to direct labor and material costs, any change in the over- head rate is treated as a price change. In this procedure, changes in the constant-dollar ratio of overhead to materials and labor do not directly contribute to changes in the level of R. & D. accomplished in any period. Data were available on R. & D. expenditures by contractor, and a sample of contractors was selected on a PPS basis. Weights for the direct labor, direct material, and overhead were, solicited from private companies, as were weights for the items. Direct labor was priced on the basis of the hourly rate of pay for a specified type of labor, e.g., aeronautical engineer III, machinist, electronics engineer IV, or a physicist of a certain skill level. 40 Data on hourly wages and salaries were readily available from con- tractors. Specification pricing of various material inputs, however, was generally not available at a detailed level from a contractor. Data were available for groups of products purchased for R. & D. and, in many cases, contractors provided their estimates of the best price indicator for a given class. These material input groupings were used and PPI's applied to develop a deflator for materials. PPI's are appropriate measures of price change in this area since they measure private sector transactions. Quality change was not a problem with labor, because the data reflect a constant skill level for each type of labor. The PPI's used for the materials prices were already adjusted for quality change. Retired military pay. -- This category consists of payments to retired military personnel. These are considered to be transfer payments in the NIPA's and therefore do not directly enter total GNP. To derive a constant-dollar outlays series, however, it is necessary to include these payments. The principal problem with this category was to determine the concept of what real expenditures represent and what constitutes quality change. Retirement pay, or more accurately a retirement system, is generally an incentive for hiring people. It is usually constructed on an actuarially sound basis, with contributions paid by the employer (and employee) while the person is working. Employer contributions are added to the basic cost of the employee's salary and are treated as an expense by the employer at the time they occur. This is the method for treating civilian retirement for DOD. Retired military pay, however, cannot be treated in this fashion. No fund has been established for military retirement, so the funds paid to retirees are appropriated annually to DOD. It is impractical to attempt to calculate a contribution cost at the time the individual was employed and add this cost to his salary at that time. As a result, these data must be treated as current-period expenditures. Retirement pay, and hence the quantity of retirement, differs for people because of different retirement rates, i.e., it differs on the basis of their rank and length of service. Thus, a retired general is paid more than a retired sergeant. Acceptance of these differences as differences in quality is consistent with the treatment of these personnel while they are working; namely, the same quality factors are employed. In light of this, it was assumed that the appropriate measure of price change for retired military pay is the change in the rate schedule of that pay. The procedure equates any change in the rate schedule of retired pay with a price change. Since rates are changed uniformly, based on cost of living adjustments, and the change in any category reflects the price change, sampling was not needed. Weights were, not needed as total payments were available and the price change was uniform. Quality change was not a problem as quality was dependent on active duty criteria. In summary, rank and length of service define quality for retired personnel and variations in the mix of retired personnel constitute 41 quality changes. Price changes occur only when the rate of retired pay changes. Ship construction. -- This category consists of contract construction of new ships and conversion of old ships for the U.S. Navy by private shipyards. Ship construction differs from the purchase of other weapon systems in that it is included in the NIPA's on a work-put-in- place basis. In other words, purchases correspond almost directly to outlays and there is no timing problem. Specification pricing was implemented for new ship construction, but in a form slightly different than used in other categories. The number of ECO's on any given ship and the fact that no two ships are identical, even when they are the same model, make output pricing very difficult. An alternative approach, called the "frozen" ship, has been implemented: At a given point in time, the ship design, includ- ing all engineering changes, is "frozen," or held constant. At this time, the detailed man-hours are specified for the various work break- down structure (WBS) categories, such as hull, propulsion, electric plant, general outfit, furnishings. In addition, the materials used in the various WBS categories and the overhead rates are specified. For each succeeding time period, an estimate is made of the changes in factor inputs needed to construct a ship of the "frozen" design. In other words, for succeeding time periods, estimates of the changes in man-hours, materials, and overhead are made for each of the WBS cate- gories that would be required to construct the "frozen" design in a particular shipyard. The detailed data necessary to construct these estimates were not available from within DOD. Therefore, private shipyards had to be queried for these estimates. To ensure respondent familiarity with the inputs and other shipyard factors, the "frozen" ship design was one that the shipyard had recently constructed. Each succeeding time period required an estimate of the factor input changes caused by productivity gains or losses, Environmental Protection Agency require- ments, Occupational Safety and Health Administration requirements, or any other changes that affected the workload in the shipyard. In addition, the overhead rates changed to reflect changing business conditions of the yard. Even though estimating the changes over a given time span required considerable work on the part of these companies, the approach was well received. All new Navy ship construction was subdivided into 3 classes: surface combatant, surface noncombatant, and submarine. One model was selected from each class: for surface combatant - patrol frigate; for surface noncombatant - auxiliary oiler; and for submarine - nuclear attack submarine. All private shipyards involved in major ship con- struction programs for the Navy during this period were canvassed. If a shipyard did not produce the particular model during the specified time, an alternative model was selected. In all, estimates were obtained from seven shipyards for a total of five models in these classes. Weights for the WBS categories were available from the shipyards. Material prices were not available from shipyards and were to be 42 collected from marine vendors. The Bureau of Labor Statistics (BLS) has been requested to collect material prices and is currently study- ing the matter. In the interim, detailed WPI specifications have been examined, and selected WPI's used to estimate the change in material prices. The labor rates and overhead prices were available from the ship- yards and posed no serious problems. Quality change was not a problem in this category because the "frozen" ship design was used. Stoak funds: defense stock fund. -- This category consists of all purchases by the defense stock fund (DSF) from non-DOD sources for resale to the military services. The DSF is managed by DLA. Other costs incurred by DLA, such as compensation of employees, rental of warehouse space, maintenance of facilities, or other purchased ser- vices, are included in other categories. The DSF category does not include items later resold through post exchanges or commissaries, since those items are not treated as DOD purchases in the NIPA's. The prices used were those of the transaction between DSF and the private sector. They were not the prices that were charged when the good was delivered from the stock fund inventory to the military service. DSF purchases were subdivided into four major classes: clothing and textiles, bulk petroleum products, subsistence, and all other pur- chases. This breakdown was necessitated by the method of record- keeping as well as the similarity of products in these classes. Special treatment was necessary for some items in the clothing and textiles class. In most cases, DOD purchases end items that are distinct and uniquely identifiable; however, DOD may also purchase the same product in different forms. For example, uniform jackets are purchased in two forms. In one, the complete jacket is purchased; in the other, material is purchased and a separate contract is made with a cut, make, and trim establishment to finish the garment. This required the stratification of clothing purchases into: (1) finished items, (2) material, and (3) cut, make and trim services. It should be noted that items made by the DOD clothing factory in Philadelphia, a GOGO facility, are not priced as end items, but its material pur- chases and the wages and salaries paid to its employees are included in other categories. Data on all classes were available to construct a sampling frame, and a PPS sample was selected in each. No unusual problems were encountered in acquiring weight data. Price data were readily obtained for all sampled items. Data on quality change were not as readily accessible as the other data. DLA technical staff reviewed specifica- tion changes on all sampled products and then forwarded data on the nature and cost of the changes to BEA for use in quality adjustment. Stock funds: military services. — This category consists of all purchases by the military services' stock funds from non-DOD sources. The purchases are generally off-the-shelf items that are more unique to each of the military services than the item purchased by the DSF. The price used, again, was the transaction price between the stock fund and the private sector, not the price charged to the particular 43 military service appropriation when the good was delivered from the stock fund inventory. Data were available, a sampling frame was constructed, and a PPS sample selected. Weight data also were available for each of the stock funds. Data on purchase prices were available and presented no unique problems. Specifications were relatively constant over time, and quality change posed only the normal problems of identifying when a change in the specification occurred and then developing the cost of the change. Transportation of things. — This category consists of freight transportation services purchased from the private sector. It does not include transportation provided by MAC and MSC since these are industrially funded activities. MAC and MSC purchases of freight transportation, however, are included. Other purchases by these activities, such as compensation of employees, fuel, etc., are treated in other categories. Because of the variety and types of transportation purchased by DOD, several unique approaches were used in this category. For transporta- tion purchased under legal rate schedules (i.e., Interstate Commerce Commission (ICC) tariffs), the price change is defined by a change in the rate schedule for a specific trip. This is identical to the approach used by BLS for the price index of railroad freight. DOD, however, also purchases some kinds of transportation under yery different arrangements. For example, DOD contracts with private firms to provide regularly scheduled air freight service between certain domestic military bases. In these cases (e.g., LOGAIR and QUICKTRANS), DOD is really purchasing transportation capacity and does not buy a ton-mile of freight moved. The pla/ies fly on schedule whether or not they are fully utilized. The contractor gets paid for flying the planes on the route, not for tons of freight hauled. Thus the con- tract service is the item being purchased, and it is priced over time, adjusted for changes in schedule, aircraft capacity, and other quality factors. To maintain capacity for ocean transportation, DOD may charter vessels on a standby basis. DOD pays a given amount to hold a vessel in reserve and another amount to use it; both amounts were separately priced. Data were available to develop a sampling frame, and a PPS sample of the various types of transportation was selected. Weights were available for the sampled items, and prices were obtainable. In some cases, particularly derivation of the ICC tariff rates, the Military Traffic Management Command (MTMC) expended considerable effort to assemble the appropriate rates from the tariffs. Quality adjustment did not prove a major problem because most speci- fications are fixed over time. Where air routes for QUICKTRANS and LOGAIR were changed, data were available on the changes and estimates of the costs of these additions, or deletions to the schedule were developed. Travel and transportation of persons. — This category consists of all expenditures for the travel and transportation of persons. It 44 includes per diem expenses for temporary travel duty, permanent change of station expenses, and local travel expenses such as taxi fares and bus fares. As in the case of transportation of things, any legally established tariff rates were treated as having a price change only when the rate schedule was changed. Some other travel, such as charters, were priced on capacity basis (e.g., price per plane mile) if the contract stipulated this. Data were available to establish a sampling frame, and a PPS sample of trips was selected. In addition, a PPS sample of expenditures of temporary travel duty and permanent change of station nontransporta- tion expenses were selected. Data were available to weight the price measures to the category total. Data on airfares between specific cities were available, and price measures for air transportation could be easily constructed. Data on per diem allowances and allowances for mileage on privately owned vehicles were available and were used to price these items. Certain miscellaneous expenses, such as taxi fares, were priced using the CPI, since data were not reasonably available; the CPI should be representative because it includes the identical products that DOD purchases. Quality change was not a problem during this period as the specified trips or charters did not change. Utilities. -- This category consists of all electricity, gas, water, sewer services, etc., purchased from non-DOD sources. Electricity generated by a military installation for its own use is excluded, but the wages and salaries, supplies, and purchased services needed to generate the electricity are included in other categories. To main- tain consistency with deflation of utilities in the other areas of the NIPA's, the rate schedule approach was used to price utilities. In this approach, only changes in rate schedules and fuel adjustments are defined as price changes. Any other changes in unit cost are treated as quantity changes. This is the same method used in both the PPI and CPI. Data on expenditures for utilities by military installation were available and were used to select a PPS sample. Weight data were also available by installation. The sample of installations was canvassed, and utility bills together with the appropriate rate schedules were collected to provide the material for pricing. No unusual problems were encountered in developing the prices from these data. Quality adjustment was not needed. Vehicles. -- This category consists of both combat and noncombat vehicles purchased from the private sector. In general, combat vehicles, like other weapon systems, are purchased by component. There are separate contracts, for example, for the M-60 tank hull, for the engine, and for the transmission. These components are later assembled into a finished tank. Commercial vehicles or noncombat vehicles generally are purchased on a per-unit basis, i.e., the complete product is purchased from one vendor. Data were available to construct a sampling frame, and a PPS sample of combat and noncombat vehicles were selected. No unusual problems 45 were encountered in acquiring weight data. Price data were available from contracts for components on the combat vehicles and for the completed units on the other vehicles. There were no unusual problems in determining quality change for combat vehicles because all ECO's were documented. Quality change, however, was sometimes difficult to determine for commercial vehicles because DOD generally uses performance specifications. DOD does not consider producers' changes, as long as they do not make the vehicle fall below the required performance specifications, as changes to the vehicle. Thus, when similar items are purchased in consecutive time periods from different manufacturers, the items may be somewhat different in physical appearance or quality, but these differences are almost impossible to determine and to measure. Weapons. — This category consists of armament and related equipment purchased from non-DOD sources. Among these are small arms, rela- tively small guns, and guns for mounting on aircraft. Large howitzers and cannon are generally manufactured at arsenals, which are GOGO facilities. Inputs to the arsenals (e.g., compensation of employees) are included in other categories. Data were available to construct a sampling frame, and a PPS sample of weapons was selected. Weight data were also available. Pricing data were available from contracts; specifications remained relatively consistent over time and provided no major problems. Quality change again proved to be tractable, as all changes are documented by ECO's and costs of the changes could be developed. 46 Part 5: Evaluation As a general statement, this project verified that it is feasible, within the limitations of conventional pricing techniques, to prepare adequate price deflators for defense purchases. These IPD's and constant-dollar estimates were constructed within the concepts of the NIPA's and yielded reliable measures of real DOD purchases within this structure. The data in this report will not answer all questions on defense inflation, but they do significantly increase the amount of data available for analysis of the defense sector. While it is apparent from the wealth of data available within DOD that measures appropriate for other purposes could be developed, time and resource constraints did not allow the full exploitation of these data. The successful development of price deflation for defense purchases further stresses the need for development of improved deflators for nondefense purchases - Federal as well as State and local government. Price changes of purchases by other Federal agencies could then be contrasted with price changes of defense purchases, and the allocation of Federal as well as State and local resources could be analyzed in detail . The experience gained, and techniques developed on this project could be used in developing (currently, no comparable data exist) comparable deflators among countries. Thus, relative positions of countries could be estimated for changes in defense expenditures, in constant monetary units, and for levels of real expenditures. How- ever, as mentioned earlier, these estimates would not provide actual measures of defense output. Input data Any evaluation of project results must begin with an examination of the input data and techniques. To a large extent, the success of the project was based on the fact that existing DOD data could be utilized. In many cases, readily available records provided the necessary detail to construct useful price and weight data; in others, particularly contract service categories, data were not available at a central location. These data are available at field locations, however, and efforts are being made to collect the data and develop appropriate measures for these categories. The techniques used to select and verify the inputs were also impor- tant. Probability sampling was used wherever possible. For categories where universe data were reasonably available, such as compensation, the entire universe was used. For others, judgment or cutoff samples had to be used; however, these were relatively few. As a whole, the sampling techniques used, yielded statistically unbiased estimates. Input data were also examined for reasonableness over the period. In many cases, this was purely a judgmental examination, but it reduced input errors and assured continuity of the series over time. Also, during the development phase of the project, an independent expert consultant reported on the methodology and data and evaluated their appropriateness for the project goals. 47 Finally, a group of experts from various government agencies, private industry, and not-for-profit organizations reviewed and commented on project goals, conceptual and statistical approaches * and preliminary results. (Group members are listed in appendix C.) These factors, as well as the care and professional judgment that the project staff exercised in processing the data, have resulted in the highest quality product possible, given the available method- ologies of price measurement and the time and budget constraints. Reliability of results Deflators for the five major GNP product groups (durable goods, nondurable goods, services, structures, compensation) are highly reliable, as are deflators for some of the working categories (see appendix B). For a few categories, however, either for reasons of sample coverage or data problems, the resulting series are not con- sidered to be statistically reliable for separate publication at this time. General pricing -problems Specification pricing, which was used in most categories, had limited applicability in some areas, particularly in the physical specification of items. This was most apparent for large electronic computers, because it was difficult to specify the exact characteristics that were purchased. It was also difficult to specify quality for many categories. The same types of problems arose for construction. Pricing on the basis of performance specifications was attempted in construction, but this may not have covered quality changes - some quality changes were not apparent in the performance specified or in the physical structure itself. For example, one type of wall structure may require less maintenance than some other standard type. This kind of quality change was virtually impossible to determine from the characteristics that were priced. Similar difficulties in defining an appropriate specification arose for the contract services. Changes in services over time may not be apparent from the contract work statement. Also, it was often diffi- cult to determine that the service was actually performed satisfactorily or to some accepted standard. For example, what is the measure of satisfactory janitorial services or guard services? This can be carried further to the case of medical services. Should a hospital stay be adjusted for the quality of the physician or the quality of the hospital or the hospital staff? If it should be adjusted, how can it be measured? These types of pricing problems have not been satisfactorily solved in the private sector of the economy, and no attempt has been made to solve them in this project. Instead, a standard approach was taken; for example, for services, a job was considered to be satisfactory if the contract was not terminated. In spite of limitations of this 48 nature, the data generated are believed to be satisfactory as measures of real DOD purchases. Major working groups Following is a brief review of the major working groups of the project. The comments must be interpreted within the context of current pricing concepts and methods discussed above. Weapon systems. — One area of surprisingly successful pricing was weapon systems. Initially, it appeared that the complex nature and the many changes made to weapon systems would overly complicate the development of price change measures. However, it turned out that DOD's purchasing methods actually facilitated the development of the measures, and the types of ECO's encountered posed no major problems. The fact that DOD buys weapon systems as components, rather than complete end products, helped to simplify the method of pricing. Each component could be examined within its own context and a relevant price series developed. Because of the vast difference in timing between delivery of compo- nents and delivery of the complete weapon system, aggregation of components to a total weapon system presented serious problems. Information on component deliveries to DOD was used for this series. Evaluation of quality changes also proved tractable to standard pricing techniques because DOD cost information was readily available on the relatively few ECO's that met the quality adjustment criteria. Although thousands of ECO's amounting to millions of dollars may be associated with a particular weapon systems, many of them occurred in the development stage of the system. In addition, other ECO's involved retrofits, nonrecurring costs, or training equipment and therefore were not relevant to the procurement of an individual item. Many of the remaining ECO's, which did relate to priced components of the weapon system, did not involve physical change that affected the cost and performance of the particular component. The "frozen" ship approach (see appendix A) for new Navy ship con- struction and conversion, a major portion of weapon system purchases, also yielded satisfactory results. In summary, standard pricing techniques were satisfactory for the development of deflators for weapon system purchases. Other goods. -- Other goods purchased by DOD presented no insurmount- able problems for pricing or quality adjustment because the data were reasonably accessible. In many cases, the data were automated and could be used almost directly. Generally, standard techniques were used for quality adjustments, although not all the relevant informa- tion could be fully developed and used. In summary, specification pricing techniques were satisfactory for the development of deflators for other goods. Compensation. — Compensation of DOD employees also presented no insurmountable problems for deflation. Although the formal concept could not be strictly followed because data on experience and educa- tion were not available, suitable proxy measures were used. Data on 49 the proxies - GS grade/step and military rank/years-of-service - were available and presented no major problems in utilization. The data for personnel classified under the wage grade system were incomplete but covered more than 90 percent of the total . As a whole, the use of proxy measures for experience and education was quite satisfactory. Construction. — Satisfactory measures of price change in construction have always been difficult to develop; however, the price per square foot method used for military construction is a significant improve- ment over the input pricing method. The price per square foot method (see appendix A) yields fairly good measures of price change and real quantity of purchases, but there are significant problems in accurately evaluating quality change over time. In general, the unit price measures were satisfactory for developing the real quantity of construction purchased by DOD. Other services. — Development of satisfactory deflators for con- tractual services purchased from the private sector of the economy proved to be the most difficult. For many service categories, data were available and specification pricing techniques yielded good results; however, for others, data were not available centrally, and this resulted in problems of data collection, weight development, and pricing. Many services in the installation support services and other services categories are procured at each military installation; there- fore, the difficulty of assembling sufficient numbers of records of prices, specifications, etc., prevented the development of a full sampling frame for these categories at this time. To test the pro- posed methodology, however, a small sample of specifications for these categories was selected on a judgment basis. Results of the test indicate that no insurmountable problems should occur in developing price measures for these categories. There appear to be no external constraints to completing the ser- vices category other than the availability of time and resources. Specification pricing appears to be possible and the data apparently exist within DOD, even for the areas that have not yet been fully developed. Although the deflators for total services are adequate for the NIPA's, detailed information below this level - particularly for installation support services and other services - will require further work to complete the sample frame and develop detailed price and quantity data. In summary, specification pricing techniques can be used to develop adequate deflators for purchased services. Future plans Although aggregate level deflators are adequate, some of the detailed data series should be improved and some of the categories should be reexamined in light of experience gained to this point. Future efforts will cover: (1) categories where the concept of what is being pur- chased needs refinement, (2) categories where the existing data may not be entirely satisfactory, (3) categories where the existing sample 50 needs strengthening, and (4) categories where the quality adjustment needs further work. Refinement of concept. -- Concepts for several categories need to be reevaluated. These categories are: medical services, R. & D., major electronics systems, and equipment purchased for space programs. Concepts for these categories will be thoroughly reviewed and an attempt will be made to refine them. For the last three, an improved measurable purchase concept may be difficult to develop because of the lack of exact specification of what is being purchased, as well as the nonrepetitive nature of the purchases. An attempt will be made, however, to measure constant-dollar purchases more accurately for these categories than is currently being done. For medical services, it also proved difficult to determine the exact purchase, because the major portion of this category is CHAMPUS, in which DOD functions as a medical insurer rather than as a direct purchaser of medical care. Once the concepts for these categories are refined, data sources and quality adjustment information must be developed before new estimates of constant-dollar purchases can be prepared. Data revaluation. -- For several areas, current data need to be reevaluated and possible alternative data sources need to be examined. These areas include: construction, ship construction, base communica- tions, and the ammunition plant modernization program. For construction, complete data were not available for the sampled items, and quality adjustment data were somewhat inadequate. For ship construction, the material input deflators need to be strengthened by additional pricing from BLS. For the other areas, new sources must be determined that relate more appropriately to actual DOD purchases and that provide enough detail so that specification pricing can be accomplished. There is no conceptual reason why this cannot be done. Sample strengthening. — The existing sample needs to be strengthened for a few categories: depot maintenance, installation support ser- vices, and other services. There appear to be no major problems, other than resources, in obtaining the necessary data or quality adjustment information; the number of sampled contracts in each time period needs to be increased to make the resulting price deflators more representative. Quality adjustment. -- While quality adjustment presented problems for all categories, current techniques have solved most of them. There is still some concern, however, about the data and the pro- cedures to be used for electronic equipment and commercial vehicles. The concern centers on collecting and processing the large quantities of data and ensuring that all adjustments are implemented appropriately. Only a lack of resources prevents this task from being accomplished. In addition to these efforts, all categories will be reviewed for undersampling, oversampling, consistency of the data, and appropriate- ness of the techniques. All categories will be monitored and samples and weights will be updated as required, to reflect changes in DOD purchasing patterns as they occur. 51 Historical data There is considerable interest in an historical series on defense purchases. Because detailed data are not readily accessible, the methodology in this report cannot be fully implemented for time periods earlier than CY 1972; therefore, summary techniques will used to prepare historical estimates of defense purchases consistent with the NIPA's concepts. These measures will provide information for a longer time series: annually from 1929 through 1945 and quarterly from 1946 through 1971. However, this series will reflect serious problems of quality adjustment because detailed data are scarce. Construction of a defense deflator covering a relatively long time span should provide a great deal of additional insight into the nature and timing of price change as it impacts the defense sector relative to other sectors of the economy. 52 Appendix A: Detailed Methodology Introductory note Appendix A contains a detailed description of the methodology used in developing the estimates for each of the 22 categories. The description for each category is divided into seven sections as follows: 1. Definition - description of the category, including relative size to total DOD purchases. 2. Concepts - description of specific concepts and definitions. 3. Sample - description of sample selection procedures. 4. Weights - description of procedures used to develop weights. 5. Prices - description of procedures used to develop price change measures. 6. Quality - description of procedures used to adjust for quality differences. 7. Sources - list of documents and organizations from which data were collected. Aircraft 1. Definition The aircraft category consists of new aircraft, as well as related aircraft components, associated parts and spares, and support equip- ment and facilities purchased by the Air Force, Army, and Navy (including the Marine Corps). Not included in this category are some Government- furnished equipment (e.g., armaments, Air Force avionics) funded by procurement appropriations, purchases from industrially funded activi- ties and purchases by or from the stock funds. These items are included in other appropriate categories. Approximately 7 percent of purchases in this category are classified as services (engineering services, technical aid, etc.) in the NIPA's; the remainder are durable goods. Purchases included in this category accounted for 8.5 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts Price change for purchases of new production aircraft was measured by pricing deliveries of major end items. Pricing specifications were 53 developed for items such as airframes, engines, and contractor-furnished avionics. Military aircraft are rarely purchased entirely from one contractor. Many major end items, such as avionics, engines, and armament, are sometimes purchased separately by DOD and furnished to the airframe contractor for installation; therefore, these end items, plus the airframe and assembly, were priced separately to develop deflators for purchases of aircraft components. The programs for modifying existing aircraft, spare and repair parts, support equipment and facilities were generally not directly priced, but were deflated by a combination of representative aircraft component price series. The price of the total aircraft weapon system at any specific time period is meaningless in the framework of the NIPA's because of the timing of component deliveries. Many of the components delivered in a given time period may not be installed onto the airframe until much later. In addition, the quantity of different components delivered in any time period may vary significantly. For example, many aircraft engines and virtually no airframes may be delivered during a given quarter. As a result, the aggregation of the deflators in any period will not yield an accurate representation of the deflator for a total weapon system. Certain assumptions are inherent in the conceptual approach used for aircraft. First, reasonable adjustments can be made for the price effects of configuration changes that alter performance of an air- craft's mission (section 6). Second, estimated prices at completion, where the specification price is not yet definitized, can be used to estimate the final price and can be revised when the price becomes available. In addition, general and administrative overhead together with profit can be prorated over contract costs to make unit costs good estimates of unit prices. Finally, the learning (progress) curve's effect on unit costs is properly reflected as price change, and the price effect of different size procurements of the same speci- fication is also price change to DOD. When new aircraft models are introduced into the series, it is assumed that the expected price of the 100th new aircraft is a valid overlap price to be used in adjusting for quality differences between the new and old aircraft. This quality difference is then adjusted for inflation and added to the appropriate old aircraft price, thereby linking the new aircraft to the old. Two key determinations in quality adjustment are: Is the new product a quality improvement? Did it cost more than the old product? The F-15A, for example, clearly costs more and outperforms the F-4E. In these cases, the quality differences were valued and the two systems were linked component by component. Quality improvements at lower cost cannot be measured; therefore, in these cases the two weapon systems' price series must be compared directly (see quality adjustment guideline C). For example, the A-10A was directly compared to the A-7D, component by component. Linking the F-15A and the F-4E price series, a case of increased quality and cost, involved the determination of the component costs 54 for both aircraft during an overlapping period. The overlap period selected was that period in which the first production contract on the F-15A was signed. This period was chosen because the decision to sign the first production contract committed the DOD to buy the new air- craft instead of the old aircraft. The expected price of the 100th aircraft was used because it implicitly took into account the learning curve effect on unit price expectations. The learning curve is based on the observation that individuals per- forming repetitive tasks increase the rate of production because of improved manual dexterity. As a result, unit labor requirements decrease during the production run of a given item. If wage increases do not offset the increased productivity, unit costs (and prices) will decline. Initial deliveries of a new model aircraft, therefore, will be made at considerably higher prices than later deliveries. Conse- quently, using the simple difference in overlap prices at the time of the first new model delivery would have overvalued the quality differ- ence. Such an incorrect valuation of the quality difference would have imparted a downward bias to the price index and overstated the real quantity of defense purchases. 3. Sample The aircraft category was stratified into Air Force, Navy, and Army subcategories, and then into new production aircraft and associated equipment purchases. Except for Army modifications, no samples were selected from the aircraft associated equipment purchases; the latter were deflated by other aircraft series. For each of the new production aircraft product classes, a sample was selected with probability proportionate to the estimated value of deliveries for FY 1974-76. Multiyear expenditure or appropriation estimates for the value of deliveries were associated with aircraft model designations for each military service. A sample size of five was arbitrarily set and resulted in product class coverages of more than 80 percent. After the sampling, other aircraft were selected with certainty because of their importance in future procurement activities, e.g., A-10A and F-16A. The following aircraft models were included in the initial new production aircraft product class samples: Air Force Navy Army A-7D A-4M C-12A A-10A A-6E CH-47C C-130H A-7E UTTAS CH-47C E-2C E-3A EA-6B F-4E F-14A F-5E P-3C F-15A S-3A F-16A TAV-8A F-111F 55 Physical end items were judgmentally selected for pricing within each aircraft model. The airframe, engines, and avionics accounted for most of the value of a delivered aircraft; therefore, these components were specification priced. The relative importance of the items priced to total aircraft price for each aircraft model can be seen in table A-1. The percentages express unit prices of sampled items to unit procurement appropriation costs by fiscal year. The difference between the percentages shown and 100 percent represents items not priced; e.g., nonrecurring costs such as tooling, peculiar ground equipment, training simulators, and technical data. Table A-1. -- Fiscal -Year Appropriations for Sampled New Aircraft Models: Percent Directly Priced 1971 1972 1973 1974 1975 1976 Air Force v. A-7D . 66 79 59 78 67 _ A-10A . - - - 63 59 C-130H . - 67 90 - - CH-47C . - - 76 - - E-3A . - - - 74 85 F-4E . 62 76 65 73 - - F-5E . 52 88 79 - - F-15A/B . - 72 73 80 66 F-111F . 67 71 77 95 - - Army C-12A _ _ 66 - _ 99 CH-47C . 97 - - - - 90 Navy A-4M . 44 _ _ 29 _ - A-6E 35 42 43 43 42 39 A-7E . 45 59 62 55 65 61 E-2C . 61 44 53 51 51 EA-6B . 34 37 38 37 36 38 F-14A . 41 47 58 67 67 59 P-3C . 31 38 39 41 44 38 S-3A . 43 62 83 80 82 It was assumed that prices of the unpriced goods and services moved similarly to prices of airframes, engines, and avionics which were specification priced. It would be desirable, however, given enough resources, to price more of the goods and services now imputed to airframes, engines, and avionics. 56 The number of airframes delivered in calendar years 1974, 1975, and 1976, and priced in the sample are shown in table A-2 and compared to universe totals. Table A-2. -- Number of Airframes Delivered 1974 1975 1976 Air Force Sample 101 Universe 315 Navy Sample Universe Sample 223 Universe 265 166 377 262 328 10 11 13 15 185 277 184 219 The Air Force coverage, in terms of the number of airframes, was better than the table indicates because most of the remainder were small value UH-1H helicopters (182 in 1974 and 181 in 1975). The new aircraft samples were selected to obtain high universe coverage and not the most efficient sample size because it was more practical to price as many aircraft models as possible than to develop an efficient sample. New aircraft weapon systems were added with certainty and therefore the probability selections represented less of the universe. To obtain a more efficient sample size, more attention should be placed on the major characteristics determining price change; i.e., age of the item, number of units ordered, and component group. The age of an aircraft system influences price change because of the effect of the learning curve. The learning curve causes initial deliveries to be priced higher than subsequent deliveries, because of expected productivity changes. As the learning curve flattens, the general influences of labor and material cost changes predominate. The number of units ordered in any given year affects the unit price because of constraints imposed by production capacity, long-lead-time items, scheduled work flow, etc. Finally, the component group - i.e. airframe, engines, or avionics - also affects price changes: Price changes of the component groups are quite different. 4. Weights The current-dollar value of new aircraft was estimated for each quarter on the basis of the value of deliveries of priced components. A two-stage blowup method was used to account for unpriced components. 57 In the first stage, a factor was necessary to account for the value of unpriced components within a sampled aircraft system. For each fiscal year, the sum of the unit prices for each priced component within an aircraft system was compared to the corresponding appro- priation unit value. The resulting appropriation factor was used to expand the value of the priced items to represent the total aircraft system. In the second stage, a ratio (max factor) was necessary for estimating the current-dollar value of aircraft that were not in the sample, but that were part of an appropriation. A max factor was derived for each fiscal year; it was the sum of the appropriation values for the sampled aircraft divided into the total aircraft appropriation for each military service. The max factor was then applied to the unit values that resulted from the first-stage blowup. The product was an implicit current- dollar value per unit of delivered components. In any quarter, the sum of the product of delivered units times the derived unit current-dollar values became the estimate of the value of new aircraft deliveries. The assumptions in this two-stage blowup were: all appropriated funds were spent, and items not in the sample were delivered on the same schedule as priced items. The current-dollar weights were adjusted for aircraft procurement included in other categories - e.g., Government- furnished avionics, weapons, and purchases by industrially funded activities. 5. Prices Procurements of components that constitute aircraft systems were frequently made at prices that were not definitized until the rewards, penalties, and ECO's were determined and negotiated when the contract was completed. At any time, a variety of prices may have existed for a physical specification; e.g., billing price, target price, initial contract price, and estimated completion price. Conceptually, the ideal price reflects the ultimate total payment for the same physical specification in different time periods. Estimated completion prices, which are derived from data routinely found in contract control docu- mentation, were used whenever available. Although these prices are only estimates, they are the best ones available because they include profit adjustments resulting from cost overruns or underruns. Where aircraft components were bought under firm fixed-price contracts and there were no configuration changes, prices were taken directly from contracts. A combination of reports was used to ensure that the same physical specification was priced over time. Estimated prices at completion were derived from cost performance reports (CPR) by prorating under- runs, profit, and overhead across contractor costs. Cost information reports were used to identify nonrecurring costs that were not shown in CPR's and that, if included, would distort the recurring specifica- tion's price change. Many reports were difficult to utilize because they contained unique contractor classification systems and interpretations of standard 58 instructions. Also, the reliability of various aspects of particular reports varied. These reports were therefore used with discretion. Some components of the aircraft appropriations were deflated with related price series. For example, Air Force engines were usually bought at the same price as identical engines destined for immediate installation; therefore, the Air Force installed engine price series were used to deflate this component. Navy spare engines and Army aircraft modifications were directly priced. 6. Quality Price series were developed for recurring physical specifications. So that the same physical specification would be priced over time, the price series were adjusted for configuration changes and Government- furnished equipment changes. Contract modification documentation was an invaluable source of information about definitized costs of various configuration changes. The "performance/cost of production" approach was the basis of the quality adjustment guidelines applied to military aircraft. Per- formance changes identified a need for quality adjustment, and the cost of the physical change was used as a measure of the value of the quality change. Physical changes were identified from ECO's and contract modification documentation. Guidelines.— Prices were quality adjusted for all physical changes that have affected cost and performance. Improved performance was defined as the enhancement of the air- craft's mission; e.g., close air support, electronic warfare, anti- submarine warfare. Expected or specified performance characteristics were presumed to be associated with the physical configuration initially developed or introduced into the price index. Unexpected physical defects were considered to be price increases and not quality improve- ments because production aircraft were expected to fit together properly and to work as specified. A. Quality adjustments were made for, but not limited to: 1. Changes in design or materials that changed the aircraft's a. Length of service b. Need for repairs c. Ease of repair d. Weight e. Quality of materials in relation to their function 2. Changes in mechanical features that affected the aircraft's a. Overall operation b. Efficiency c. Ability of a component to perform its function 59 3. Engineering changes that affected the probability of mechanical failure of particular systems. 4. Safety features, e.g., better seat ejection systems. B. Quality adjustments were not made for physical changes associated with: 1. Style 2. Appearance 3. Design solely to make the aircraft seem new or different 4. Comfort 5. Convenience 6. Antipollution, noise abatement equipment, etc., installed for nonmi lita ry objectives 7. Remedy of production compatibility deficiencies Quality change may not be directly proportional to the changed cost of production. For example, new technology may make it possible to achieve recognizably better quality at lower cost. No satisfactory technique has been developed for adjusting for quality change in such situations. Therefore, rather than reflect erroneous quality deteriora- tion no adjustments were made. Prices before and after the change were compared directly, thereby giving at least partial credit for quality improvement. 7. Sources Air Force • Aeronautical Systems Division CASD) , Air Force Systems Command CAFSC), Wright-Patterson Air Force Base, Dayton, Ohio t Electronic Systems Division [ESD) , AFSC, Hanscom Air Force Base, Boston, Massachusetts • "Funds Available, Budget Authorization, Commitments, Obligations, and Expenditures," HAF-ACB(M)7130, Pentagon Army § Army Aviation Systems Command, St. Louis, Missouri • "Appropriation Status by Fiscal Year Program and Subaccounts," DD Comp[M)1002, Pentagon 60 Navy Naval Air Systems Command, Arlington, Virginia t "Appropriation Status by Fiscal Year Program and Subaccounts," NAVCOMPT Form 2158, Pentagon Ammunition 1. Definition The ammunition category consists of bombs, bulk explosives, cartridges, torpedoes, demolition materials, grenades, mines, projectiles, pyrotechnics, rockets without nuclear capability, simulated nuclear ammunition, and warheads in completed round configuration or in separately packaged items of issue for complete round assembly. Also included are the load, assemble, and pack (LAP) services performed on the various purchased components. Excluded are ammunition and components pur- chased from industrially funded activities or stock funds. The items in this category are classified as nondurables in the NIPA's; they accounted for 1.1 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts Ammunition is purchased from three distinct sources: commercial plants, GOCO plants and GOGO plants. Commercial plants are owned and operated by a private contractor. GOCO plants, such as Army ammuni- tion plants are owned by the Government, but operated by a private contractor. The Government furnishes all capital equipment to GOCO plants. GOGO plants, such as Army arsenals and Naval ordnance sta- tions, are owned and operated by the Government. Because GOGO's are industrially funded activities, they are specifically excluded from this category. Purchases by GOGO's (e.g., compensation of employees, materials, supplies) are included in other appropriate categories. A finished round of ammunition consists of many components and a LAP function. For example, a typical round of 105mm ammunition consists of a projectile, explosive, fuze, propellant, primer, and cartridge case. The projectile, fuze, primer, and cartridge case may be pur- chased from a commercial plant or produced at a metal parts GOCO plant. The explosive, propellant, and primer are produced at a GOCO and explosive plant. Each component may be partially assembled at a LAP plant in an intermediate stage. After all the components are acquired and preassembled, they are shipped to another LAP plant for final assembly. At this last stage of production, the Government is not purchasing a final round of ammunition, but rather services required for the final loading, assembling and packing of the round. Expenditures are made at each stage of production for the particular functions performed at that stage, rather than one payment for a finished round of ammunition. Therefore - except for certain small 61 arms ammunition - DOD does not purchase a complete finished round of ammunition, but rather it purchases individual components that are assembled into a finished round. Also, there may be a considerable time lag between the manufacture of some or all components and the LAP function to produce a finished round. For example, a projectile may be produced and stored in 1972 and the explosive produced and stored in 1973. If these two components are used to produce a finished round in 1974, the Army has not bought a finished round of ammunition at that time, but it has bought all the components except the projectile and explosive. If the finished round of ammunition were priced in 1974, rather than pricing the components as they are purchased, the inventory would not be counted in the year it was bought, but rather in the year it was used. For these reasons, the components were priced rather than the finished round of ammunition. Because each component is an item with a detailed specification, the specifications were sampled and items corresponding to the specifications were priced over time. Most commercially produced items are very different from G0C0- produced items; however, a few small arms and metal parts may be similar or identical. Even when the items are identical, the factor costs are different. For GOCO items, the unit price reflects only the factor costs that the contractor incurs; e.g., labor costs, material costs, incentive fee. Amortized capital expenditures and Government- furnished materials are not included in the price since they are a separate cost to the Government and not to the contractor. The unit price of a commercial item, however, represents the market price where the cost of all factors of production are included. For this reason, physically identical items from GOCO and commercial plants were treated as separate products. 3. Sample Ammunition was first divided into two major subcategories - GOCO and commercial. Because all military services purchase items from GOCO plants, one GOCO sample was selected, regardless of customer. Samples were selected for each military service's commercial purchases. GOCO plants perform two functions: (A) production, and (B) LAP. Total purchases from these plants were stratified accordingly. Indi- vidual items (specifications) were then grouped into product classes based on the homogeneity of the items. The number of classes depended upon the nature of the strata: A. Production 1. Cartridges 2. Propellants 3. Explosives 62 B. LAP 1. Projectiles and mortars 2. Grenades and mines 3. Fuzes, primers, delays and detonators 4. Propel lants 5. Bombs 6. Cartridges 7. Rockets Available data contained detailed information on approximately 40 separate components that accounted for about 75 percent of the dollar value of deliveries of all G0C0 plants. The total dollar value of deliveries in the base year was determined for each item and ranked within its respective class. The top three dollar value items in each class were selected. In FY 1974, these 30 items represented 65 percent of the dollar value of deliveries of all G0C0 plants. The Army commercial sample was selected in a different manner. All items purchased in the base year were identified and grouped into classes based on the homogeneity of the items. The classes were: A. Cartridges B. Projectiles C. Explosives D. Fuzes E. Miscellaneous PPS sampling techniques were used; size was defined as expenditures in the base year. The Navy commercial sample was selected by first identifying each item for which there was a contract in the period FY 1974-76. These items were examined and grouped into product classes based on the homogeneity of the items. The classes were: A. Projectiles B. Cartridges C. Bombs D. Fuzes E. Chemicals, charges, explosives and propel lants F. Miscellaneous PPS sampling techniques were used; size was defined as contracted value in FY 1974 and FY 1975. Due to time constraints, a separate Air Force commercial sample was not selected. Air Force commercial purchases of ammunition, which amounted to somewhat more than 20 percent of total ammunition, were deflated by a combination of Army and Navy deflators. 63 4. Weights Total GOCO weight was equal to the value of purchases from GOCO plants by the Air Force, Army, and Navy (including Marine Corps). Weights for the product classes were determined by totaling the value of deliveries of all items within each class. It was assumed that within each class, the relationship of the sampled items to each other equaled the relationship of the sampled items to the product class. Army and Navy commercial ammunition purchases were based on total purchases of ammunition and components less purchases from GOGO's and GOCO's. Sample weights were determined by totaling expenditures for the sampled items. The distribution of the sample weights was used to allocate the total commercial weight. Air Force commercial quarterly weights were based on total expendi- tures for ammunition and associated equipment less purchases from GOGO's and GOCO's. 5. Prices In all cases, the price of each sampled item was determined from contractual data adjusted for any contract modifications that affected price. Total contract value of the sampled item was divided by total contract quantity to determine price. Current price for any sampled item for any given quarter was equal to the weighted average price for all such items purchased in that quarter. Air Force direct purchases from the private sector were not separately priced, but were deflated by a combined series of Army and Navy com- mercial deflators. 6. Quality The major criterion used to determine if there had been a quality change was whether or not there had been a specification change. If there had been a specification change, the resulting change in cost was determined. If increased quality was associated with higher cost, or decreased quality with lower cost, the price was adjusted accordingly. Most quality changes in ammunition resulted from product improve- ments, which did not occur during the production cycle; therefore, most quality adjustments were made when an item was replaced or sub- stituted. A dollar value associated with the cost change was determined, and the price adjusted by the formula: Let P^ = base price P x = unit price in period x P x + 1 = unit price in period x + 1 Y = price change associated with quality change Z = pure price change 64 Then: P / , - P =Y+Z x + 1 x P P + Y x x P b New P b ?. Sources • "Comptroller's GOCO Financial Analysis Report," Army Armament Materiel Readiness Command, Rock Island, Illinois t "Summary of Orders and Cost of Deliveries," Army Armament Materiel Readiness Command, Rock Island, Illinois t "Procurement of Equipment and Missile Management Accounting & Reporting System," Army Armament Materiel Readiness Command, Rock Island, Illinois § "Procurement Production Status Report," Navy Ships Parts and Control Center (SPCC), Mechanicsburg, Pennsylvania Communications 1. Definition The communications category consists of purchases of communications services and postage. Specific types of communications services are: international, interstate, intrastate, and local services, including both the purchase of services and rental of equipment. Major purchases are for the Autovon (voice communication system) and Autodin (auto- matic data transmission system) networks. Also included are payments to the United States Postal Service (USPS) for delivery of various classes of domestic and international mail and related services. Purchases of communications are classified as services in the NIPA's; they accounted for 0.7 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts A large portion of DOD purchases of communications services are made through CSIF. Since CSIF is an industrially funded activity, CSIF purchases were priced because these transactions were between DOD and the private economy. CSIF sales to the military services 'were intra- DOD transactions and were therefore excluded from consideration. CSIF primarily purchases interstate and international communication services for DOD. It also purchases some intrastate or local services, although the bulk of these services are procured by individual military installations. 65 CSIF attempts to provide a given type of service Ce.g., Autovon or Autodin) at the lowest cost; it may, therefore, purchase Cor lease) different kinds of equipment and services from the private economy to provide the given service. In other words, if an equivalent amount of service was provided in two cases, regardless of the particular pieces of equipment that were leased, specifications were assumed to be the same. For example, Telpak service terminals are required for access to Telpak circuits. As long as the appropriate access to the circuits is provided, it does not matter what specific pieces of equipment are involved; the same specification is considered to be purchased over time. In summary, the specific type of service is purchased by the CSIF, and not the piece of equipment. Local and intrastate services, whether purchased by the CSIF or by an individual military installation, are purchased under a standard rate schedule (tariff). The appropriate measure of price change is any change in the rate schedule that is applicable to the particular type of service being purchased. Resource limitations did not allow the development of a sampling frame suitable for probability sampling at this time; therefore, the CPI for telephone service was used to estimate the price change for this subcategory. 3. Sample Communication services purchased by DOD were divided into three subcategories: (1) interstate and international services, (2) intra- state and local services, and (3) postage. CSIF purchases of inter- state and international services were disaggregated into eight product classes: A. Autodin switching centers B. Autodin trunks C. Telpak service terminals D. 60-channel telpak circuits E. 240-channel telpak circuits F. Autovon backbone G. European leasing services H. International leased services There were relatively few specifications within these product classes, and ample data were available to price all specifications directly. Therefore, sampling was not attempted, and all classes were priced. No sample was selected for intrastate and local services because a frame suitable for sampling could not be identified in the time available, Purchases from the USPS were divided into 32 strata, which were equated with specifications. Given the limited number of specifica- tions in this area, all were selected and priced. 4. Weights CSIF expenditures for each class of interstate and international 66 service were available. Aggregate expenditures by military installa- tions for local and intrastate services were also available. Total DOD expenditures for postage were available. Expenditures by class of postage were available from the Air Force, and this distribution was used to derive item weights for the total. 5. Prices Price-per-unit for six of the eight interstate and international communication service classes were obtained. For the European leasing service, only the unit cost of a communication service authorization was available. The erratic quarter to quarter changes in this cost, coupled with the fact that the units were not constant over time, led to the rejection of this as an appropriate deflator. The European leasing class was deflated by the aggregate deflator for the other interstate and international service classes. International leased services were priced using the average monthly cost for the transoceanic trunk portion of this service. The composi- tion of this service remained relatively constant over time, and price change was infrequent; therefore, product mix for this specification was not a problem. Intrastate and local services were not priced. CSIF and local military installations expenditures were deflated using the CPI for telephone service. This procedure was not considered to be totally satisfactory, but the CPI should reflect price change for local and intrastate services since these are the major components of the CPI for telephone service. Postal rates for each class of service were readily available and represented the price for each of these items. 6. Quality A change in quality of communication service was defined as a change in the particular purchased service. In other words, a change in the specific type of carrier or equipment used (e.g., telephone line vs. satellite) was not considered to be a quality change unless it made a noticeable difference in quality to the user. Most recent technological change in communications has resulted in the conversion of long lines to microwave relays and to satellites. DOD purchases a communications link between two parties and not the use of the particular piece of equipment. In many cases, the medium used to transmit the message was unknown and made no noticeable difference to the user; consequently, there was no adjustment for quality change. The CPI used to deflate local services had already been adjusted for quality change by BLS. Measurement of quality change for postal services was not attempted. 7. Sources • CSIF, Defense Communications Agency, Arlington, Virginia 67 • Office of the Comptroller, Defense Commercial Communications Office, Scott Air Force Base, St. Louis, Missouri • Army Commercial Communications Office, Telecommunications Management Division, Fort Huachuca, Arizona t CPI for telephone service, BLS t Postal Policy Division, Headquarters, U.S. Air Force, Pentagon Compensation 1. definition The compensation category consists of the income accruing to military and civilian employees as remuneration for their work. It is the sum of wages and salaries and supplements to wages and salaries, such as DOD contributions for Federal old age, survivors, disability and health insurance, Government retirement (except military retirement), and workmen's compensation. Also included are payments in kind furnished by DOD to its employees, such as food and clothing. The direct and indirect hire of foreign nationals is excluded; this is considered to be the purchase of a service and is included in the other services category. In CY 1977, compensation accounted for 46.5 percent of total DOD purchases of goods and services on a NIPA basis. 2. Concepts In the NIPA's, the quantity being purchased as Government compensa- tion is an hour worked adjusted for quality change; quality is measured as a combination of the education and experience of the employee. In terms of specification pricing compensation, therefore, the same education and experience criteria should be tracked over time. Any changes in compensation of workers, with these criteria held constant, should be construed as price changes. The correct base price is the base period average compensation for each employee-hour worked strati- fied by education and experience. Because these data are not readily available on a recurring basis, a substitute measure was developed. For civilian employees classified under the GS system, the structure provides a built-in measure of experience and education. The Civil Service Commission basic requirements for a given job generally specify the education and experience required. In addition, there are provi- sions for promotions within the GS grade (the step increases), which reflect additional experience related to the particular task that an employee is performing. The GS grade and wi thin-grade step should be held constant over time to reflect fixed education and experience combinations. The same rationale applies to WG personnel, since that system has a similar pay structure with grades reflecting education and experience criteria and within-grade steps for additional experience. 68 The same rationale, with only small modifications, was used to measure military compensation in constant-dollars. Education and experience were assumed to be determined by military pay grade and length of service, with additional adjustments for specialty pays (e.g., submarine pay, flight pay, jump pay) and other items that reflect added quality to the military service. No adjustment was deemed necessary for hours worked, since military personnnel are assumed to be on duty or on call 24 hours a day. Thus, average hours worked were defined as constant. In summary, then, the measure of the quantity of compensation pur- chased was an hour worked adjusted for quality change. In the GS and WG systems, quality was reflected by the grade and step of the employee; in the military it was reflected by rank and years of service. The adjusted hours were then repriced into base period prices. Dividing this aggregate series into actual outlays for compensation yielded the IPD for this category. '3. Sample Compensation of employees was stratified into five subcategories: Army, Air Force, Navy, Marine Corps, and other DOD. Within each sub- category, the classes selected consisted of active duty military, civilian GS, and major components of the civilian WG system. All active duty military service members, classified by pay grade and step within grade, were selected for pricing. Reserve and National Guard compensation was not directly priced, but was deflated by the IPD for the active duty military. DOD civilian employees in the GS and major components of the WG system (WG, WL, and WS), classified by pay grade and step within- grade, were selected. The current-dollar compensation associated with the less-significant WG pay schedules was distributed proportionately to, and deflated along with, WG, WL, and WS compensation. The series excluded represent less than 10 percent of civilian compensation. 4. Weights Weights for the major classes (military, GS, and WG) were readily available. Within each of these classes, the method of constructing the output measures - namely, employee-hours adjusted for quality change valued at base period prices - provided a self-weighting structure to the major class level. The detailed level weights were the employee- hours worked by grade (rank) and wi thin-grade step (length of service) . 5. Prices A. Military compensation: Compensation of active duty military personnel was divided into two subclasses: 69 1. Proficiency, incentive, and specialty pay 2. All other compensation (e.g., basic pay, basic i otner compensation \e.g. illowance for subsistence) The first subclass consists of pay for education or experience in addition to that included in the rank and length of service classifi- cation. Price relatives for this subclass were developed by comparing quarterly rate schedules for these pays. Current-dollars were deflated by the resulting price indexes. The pricing specifications for the second subclass were rank and length of service. Average base period compensation by rank and length of service was not readily available, so basic pay was used as an estimator; the assumption was that other compensation categories were proportional to basic pay. Quarterly employment by rank and length of service was multiplied by base period basic pay by rank and length of service and summed to develop a constant-dollar series for basic pay. The quarterly IPD's for this subclass were derived by dividing current-dollar basic pay by the constant-dollar estimate. The IPD was then applied to the total compensation for this subclass. B. Civilian Compensation 1. GS - The pay rates by grade and step are uniform for each of the military services and for DOD agencies. The Civil Service Commission provided the applicable rate schedules. The weighted (by numbers of personnel) average base year pay rates by military service and DOD agencies were then calculated. 2. WG - The Manpower Research Data Analysis Center provided the quarterly base year personnel distribution by pay schedule (WG, WL, and WS) for each military service and DOD agencies. It also provided quarterly data on dollars paid by pay schedule by pay grade and step within grade. These data were used to calculate average base year pay rates for each grade and step. Deflators for DOD civilian compensation were developed in the follow- ing manner. For GS employees, the number of employees by grade and step for each period were multiplied by the average base period GS salary; this yielded total salaries in each period in base period prices. These salaries were adjusted for changes in overtime by applying data on changes in average weekly hours paid (data on hours worked were not available). This procedure treated 1 hour of overtime as equal to 1 hour of straight time. The adjusted salaries for each quarter were then used to extrapolate total base period GS compensa- tion to derive quarterly estimates of constant-dollar compensation. The current-dollar figure divided by the constant-dollar estimate 70 yielded the IPD for GS compensation. The same procedures were applied to WG personnel, divided into three subclasses: WG, WL, and WS. 6. Quality Changes in average grade were assumed to reflect changes in the average education and experience of the work force; therefore, no quality adjustments were made. 7. Sources • Office of the Secretary of Defense (Comptroller), Pentagon • Department of the Army, Budget Division, Pentagon t DLA, Manpower Research Data Analysis Center, Alexandria, Virginia • U.S. Civil Service Commission, Washington, D.C. Construction 1 . Definition The construction category consists of all construction purchased by the military services. Included are military construction, family housing, and the construction portion of the Minuteman force moderni- zation program. Specifically excluded are the construction performed in- house by DOD employees and the purchase of construction materials. These activities are included in other appropriate categories (e.g., compensation). Military construction [other than family housing) consists of facili- ties built to assist, enhance, house, or perform the activities that the military services need to accomplish their mission. The construc- tion activities range in complexity from manufacturing plants to storage sheds. The family housing portion consists of the construction of new permanent living quarters for DOD military and civilian personnel, preparation of mobile home sites, improvement and rehabilitation of existing accomodations, leasing of rental units, operation and mainte- nance of family housing, payments for housing indebtedness, mortgage insurance premiums, and purchases of family housing in foreign countries The Minuteman force modernization program portion consists of expendi- tures for upgrading the survivability of Minuteman missiles in the silos. Included in this activity are site preparation, silo penetra- tion, increased thickness of silo walls, improved dust and waterproof cover, steel silo liner, and associated landscaping. The construction category is classified as structures and as ser- vices in the NIPA's. Construction amounted to 2.3 percent of total DOD purchases in CY 1977. 71 2. Concepts Purchases of construction in the NIPA's are recorded on a work-put- in-place basis: The assumption is that construction is purchased as it is performed, even though DOD does not take title until the con- tract is fulfilled. During construction, progress payments are made to the contractor on the basis of a fixed percentage of the estimate of work completed. Although these payments will not exactly corre- spond to the total amount of work that has been completed at any given time, they do approximate the amount that has been completed on a particular contract. Outlays data and GNP purchases are, therefore, quite similar in dollar magnitudes and timing. The definition and measurement of an appropriate price for con- struction purchases poses many problems. Price change of a finished product is generally estimated by pricing the inputs to the specific unit being constructed. In other words, prices for the labor and materials that went into the structure are used to estimate the change in the purchase price of the completed structure. This type of pricing is generally unsatisfactory, in that it does not allow for changes in the productivity of labor, for capital or labor substitu- tion, for changes in the composition of materials, or for shifts to the use of prefabricated components for onsite construction. In this project, these problems were bypassed by directly pricing the unit of construction that was purchased. The price per unit (i.e., square foot, linear foot, cubic yard, etc.) for a specific type of construction was defined as the appropriate price. Detailed specifi' cations for each type of construction purchased by DOD exist, but within each type of construction, there are many variations in indi- vidual projects. In many cases, these referred to style or to place- ment of units (e.g., bathrooms) and did not affect the price per unit. Other variations, which affected the price per unit, were considered to be price changes. Thus, the specifications for a given type of construction established the essential value of the project to DOD. In general, the specifications were performance specifications, rather than detailed material of structural specifications. For example, the specifications for troop housing centered around the number of intended occupants for each bathroom, the requirements for soundproofing the walls, size of rooms for single or double occupancy, etc.; they did not necessarily specify the materials from which the walls were to be constructed. Thus, as long as a particular project complied with the performance specification, it met the pricing specifi' cation for that type of construction. The average unit price, there- fore, accurately measured the purchase price of a particular type of construction purchased by DOD. The price index derived from these data reflected any changes in productivity, capital or labor substitu- tion, material substitution, etc., that occurred over time in the construction industry. The price index also reflected changes in construction industry profit margins. Regional differences in price (or cost) for the same specification were considered to be price differences to DOD; however, differences 72 in construction performance requirements because of extreme geographic climatic differences (e.g., a storage facility for JP-4 in Alaska or southern California) were considered to be quality differences. Differences in the cost of labor in two regional markets, however, were considered to be price differences to DOD. 3. Sample Total contract construction purchased by DOD was identified from readily available data. This universe was stratified into 39 classes of military construction. Included in the initial stratification were a family housing class and a miscellaneous class that was subsequently eliminated from the sampling frame because it was too heterogeneous for pricing. The family housing class was selected with certainty because it was funded under a separate appropriation category from the rest of DOD construction. However, because resources were limited, only nine other classes of construction could be selected. The 37 classes were therefore ranked according to size, and PPS techniques were used to select the sample. The sizes were determined by totaling the contract awards by class for 4 fiscal years (FY 1973 through FY 1976) and then ranking the total value of the contract awards by magnitude. The nine classes selected were: A. Airfield pavements B. Training facilities C. Maintenance facilities D. Covered storage facilities E. Administrative buildings F. Troop housing G. Facilities for personnel support and services H. Research, development and test buildings I. Road and streets Because of the diversity of coverage, 4 construction classes were further stratified so that pricing specifications would be more homo- geneous. The strata and substrata for airfield pavements and roads and streets were: A. Rigid pavements B. Flexible pavements 1 . Surface 1 , 2. Base course 2, 3. Subbase course 3. 4. Excavation and 4. grading The strata for troop housing were: A. Bachelor enlisted quarters--Army B. Bachelor enlisted quarters--Navy Surface Base course Subbase course Excavation and grading 73 C. Bachelor enlisted quarters--Air Force D. Bachelor enlisted quarters—Marines E. Bachelor enlisted quarters—Women F. Bachelor enlisted quarters with mess G. Bachelor officers quarters H. Messes I. Service buildings J. Administrative and supply buildings The strata for covered storage buildings were: A. Dry storage B. Hazardous material storage These classes, plus family housing and Minuteman silo modernization, represented direct coverage of over 50 percent of total expenditures for construction during the FY 1974-76 period. 4. Weights Annual expenditures for new military construction and family housing were readily available. Quarterly expenditures were only available from the U.S. Army Corps of Engineers (USACE) and the Naval Facilities Engineering Command (NAVFAC). However, these totals included most of the Air Force construction contracts. As a result, any double counting resulting from lags in reported reimbursement payments had to be avoided. Therefore, total DOD annual data were used as controls and were allocated quarterly according to the USACE and NAVFAC data. The family housing data did not present this problem because, with s/ery few exceptions, each military service let its own contracts. Thus, the reports on the few joint ventures were isolated to prevent double counting. Quarterly expenditures by construction class were not available; therefore, an estimate of expenditures for each sampled class was prepared by using a five-quarter moving average of contract awards data. This distribution was used to assign weights to each sampled class for each period. Implicitly, this procedure assumed that con- struction put in place of the non-sampled classes was proportional to that of the sampled classes. Within each class, the procedure weighted each contract in proportion to the value of square feet of construction put in place that quarter. Quarterly expenditures were available for improvements, minor con- struction and planning; however, Army and Air Force expenditures for family housing operation and maintenance and housing leasing were available only on an annual basis. Therefore, quarterly expenditures for Navy family housing operation and maintenance were used to distrib- ute Army and Air Force expenditures. Total quarterly expenditures for missile silo modernization were obtained directly from the Air Force system project office for the program. 74 5. Prioes The pricing specifications for military construction (except family housing) were the construction classes which rigidly defined the projects in that class. Each project was reported on a prescribed form that contained the following information: A. Contract cost per unit B. Total units (square feet or cubic yards) contracted for C. Date of contract award D. Construction period in days or date of completion of construction E. Total contract costs for the facility As explained previously, the appropriate price for this category was the contract cost per unit, e.g., square foot, square yard, cubic yard, etc., depending on the particular type of construction purchased. Generally construction was purchased on fixed price contracts; therefore, the price was constant for each project through the dura- tion of the construction period. Only changes to the work being performed or legal claims altered the price for a given project. Wherever feasible, the final price paid for a project was used. This price, calculated on per-unit basis, included all changes or modifica- tions during the life of the project. Not all reports within each class were available: For 7 of the classes, between 41 and 72 percent were available; for the remaining 2, less than 25 percent. However, it was assumed that within each class prices for available and unavailable reports moved similarly. Since beginning construction dates were not generally available, it was assumed that construction started in the month after the contract was let. Starting with that month, the total number of quarters for a given project was obtained by dividing the construction period in days by 90. The number of units of construction (square feet or cubic yards) put in place per quarter was determined by dividing total units by the number of calculated quarters. In many cases, construction started or ended during a quarter, rather than at the beginning or end. Therefore, the nearest one-third of a quarter was determined from the starting month, and quarter units were allocated for one- third or two- thirds of the total. Expenditures by project were estimated by multiplying the total units put in place per quarter on the project by the unit price of the project. Total units per quarter and total current-dollar expenditures for that quarter were aggregated by class and divided to arrive at a weighted average price per quarter stated in dollars per unit put in place during that quarter. These average prices by class were used to develop price indexes and then to deflate the current-dollar estimates of the value of construction put in place. Prices per square foot for family housing construction were developed in essentially the same manner. Each military service provided informa- tion on contract price, construction authorization start date, completion 75 date, and number of family units and distribution by military rank. The total number of square feet of living space for each project was determined from regulations prescribing the square feet per housing unit by military rank. Regulations on size of living space were changed starting with construction authorized in FY 1974. The total square feet of living space were divided into the contract cost in order to determine the price per square foot for each project. A weighted average price per square foot per quarter was then developed in the same manner as that for military construction. Cost limitations on family housing during part of the period, as explained later, may have had impacts on the average price that were really quality changes. Examination of the average quarterly prices, as calculated, indicated that some quality changes may not have been readily apparent. Resource limitations prevented correction of this potential deficiency; therefore, the family housing class was deflated by the IPD for military construction. The Minuteman force modernization program required slightly different procedures. The construction portion consists principally in removing the missile from the silo and making structural changes to the silo. The changes include the installation of thicker armor plating, heavier concrete on the door, and new snow and debris removal equipment. Since virtually the same operation was performed on each silo, specifi- cation pricing was feasible on a per-silo basis. In some cases, a particular silo required additional work to bring it to "standard." This additional work was funded out of the "over and above" funds in the contract and were separable from the standard work. The "over and above" work could not be directly priced since it varied by location; however, expenditures were deflated by the price change of standard work on a missile silo. This assumed that the price change of the "over and above" work moved similarly to that of the regular work. 6. Quality Quality change in construction was determined by a change in physi- cal characteristics that related to the intended use of the structure. Unit price differences related to geographic area or to project size were assumed to be price differences and not unit quality differences. Rather, the quality of the construction end product was assumed to relate only to its function. In other words, if a given structure met performance requirements for the class, it was considered acceptable and met the specification for pricing. If the performance requirement was changed, the result was a quality change. For example, in troop housing specifications, if the performance requirements were changed from open dormitory sleeping facilities on each floor to private rooms for each occupant, a quality change occurred. If, however, brick exterior walls were used in one facility and concrete block with stucco walls were used in another facility, there would be no quality difference if both met peformance requirements. Congressional cost limitations imposed on' family housing construction programs in fiscal years 1971, 1972, and 1973 may have resulted in 76 some quality changes that were not apparent from the available data. Specification adjustments from the limitations included such items as elimination of carports and patios, changes in or elimination of landscaping, and changes in construction material. Any of these changes would have affected construction prices and, subsequently, deflators for later years - i.e., FY 1974 and FY 1975. Quality changes can be determined only by studying the plans and specifications of the projects constructed during this period, to determine whether or not the specifications were changed. Examination of the deflators, coupled with discussions with the family housing offices of the various military services, indicated that some changes affected quality. For this reason, the data on family housing were not used for pricing because the results may have been misleading. For missile silos, quality differences in specifications between squadrons of missile silos were removed to adjust each sqaudron con- figuration to the previous squadron. The first squadron renovated was used as a base; then the work statement of the second sqaudron was adjusted to the first by removing the costs of any quality changes. The resulting price difference was used to calculate the deflator. For the third squadron, the costs removed from the second squadron were added back, and the third squadron was quality adjusted to the second squadron in the same manner to arrive at the price difference. Any change in Air Force construction costs per silo, after the adjust- ments, was considered to be a price change. 7. Sources • USACE, Construction Division, Military Construction Directorate, Washington, D.C. • NAVFAC, Criteria Management Division, Alexandria, Virginia • Minuteman System Program Office, Norton Air Force Base, California Depot maintenance 1. Definition The depot maintenance category consists of all DOD depot maintenance activities purchased under contract from the private economy, including periodic maintenance and modification on major weapon systems and their components. Military in-house maintenance is performed at depots that are industrially funded activities; therefore, it is an intra-DOD transaction. Inputs to in-house maintenance are included in other appropriate categories. This category is classified as services in the NIPA's; it accounted for 1.8 percent of total DOD purchases in CY 1977. 77 2. Concepts Three major types of work are performed as depot maintenance: (1) periodic maintenance of major weapon systems, (2) modification of the systems, and (3) overhaul or maintenance of components, generally referred to as "exchangeables" maintenance. The first two - grouped together as one subcategory - could be priced from contracts. The exchangeables maintenance subcategory, however, required another approach; therefore, it was treated separately. For the first subcategory, there was generally a rather elaborately detailed statement of work to be performed on each unit and a price for the work. Thus, a specification for periodic depot maintenance or modification of a weapon system was developed from the contract; it was amenable to specification pricing techniques. Additional work, i.e., that needed to bring a particular aircraft or missile up to "standard," is referred to as "over and above" work and work is generally contracted for as a separate line item and at a specified rate. This type of work does not lend itself to specifica- tion pricing since the amount and cost of the work on any unit are subject to negotiation. Thus, the type or quantity or price of work is not consistent from unit to unit. For the second subcategory - exchangeables maintenance -the work performed on each unit is generally less complex than that performed on the full system. The work ranges from the simple regrooving of tires to the complex periodic maintenance of gyroscopes or coding equipment. Usually, the unit to be repaired is removed from the weapon system and replaced by a spare; the original piece is then overhauled or undergoes periodic maintenance and is returned to the shelf for use when needed. Although contracts and types of mainte- nance vary widely, the work statements are generally relatively simple and consistent over time; for example, maintenance on a gyroscope could consist of merely replacing the bearings. It was assumed that specifications for work on a component were unchanged unless the price change between two periods varied widely from the average for the subcategory. Thus, change in the average unit cost of the maintenance performed on a specified item was con- sidered the price change for that item. A more detailed explanation of the calculation procedures is contained in the prices section. 3. Sample The weapon systems overhaul and modification subcategory was further stratified by branch of military service. Each branch was divided into product classes by major weapon systems; e.g., aircraft, missiles, ships. Product classes for sampling were selected judgmentally from within each military service. Items to be priced from within each product class were selected to include a large portion of contracts that amounted to a substantial total expenditure and that were for maintenance programs that continued for several years. 78 Two product classes were selected for the Army sample: A. Aircraft maintenance 1. CH-47 helicopter overhaul 2. AH-1G helicopter overhaul 3. OV-1 overhaul 4. C-12A logistic support B. Combat vehicle 1. M-88 recovery vehicle conversion and overhaul 2. GOCO maintenance facility--Mainz, Germany Two product classes also were selected for the Air Force sample: A. Aircraft maintenance 1. F-4 overhaul --Casa Getafe, Spain 2. F-4 overhaul --Taiwan 3. KC-135 overhaul 4. B-52G overhaul and modification 5. C-131 overhaul 6. C-9 logistic support 7. T-43 logistic support B. Aircraft engine maintenance 1 . J-69 engine overhaul 2. J-60 engine overhaul 3. J-85 engine overhaul 4. J-57 engine overhaul Only one product class was selected for the Navy sample: A. Ship overhaul 1. SSN overhaul 2. LST overhaul 3. ATF overhaul 4. FF overhaul The exchangeables maintenance subcategory was also stratified by branch of military service. Samples were selected, however, only for Army and Air Force. A sampling frame was created using the "Master File for Maintenance Report" for the Army and "Uniform Depot Maintenance Cost and Production Report" for the Air Force. These reports con- tained information on every exchangeable item overhauled during a given time period. For each time period, all 13-digit stockynumbered items overhauled on contract and having at least one buy of over 79 $25,000, no in-house costs, and nonzero quantities in both the current and the previous time periods were selected for the sample. This resulted in a sample size of over 200 items for the Army and for the Air Force. 4. Weights Expenditures for each subcategory of contract depot maintenance were available from each military service. Within each weapon systems overhaul and modification subcategory, contract expenditures were selected as weights. Within the exchange- ables maintenance subcategory, each sampled item was equally weighted to the total . 5. Prices Periodic maintenance and modification of major weapon systems were priced from contracts. Although each D0D contract was for maintenance on only a single item (e.g., all OV-l's or all CH-47's), within each contract it was necessary to determine whether or not the same mainte- nance tasks had been performed on each unit included under the con- tract. A specification was defined by examining the statement of work, excluding any "over and above" work, and identifying the line item that described the standard recurring work completed on all units under the contract. In many cases, maintenance services were delivered at an old price after a new contract or modification with a new price had taken effect. Thus, the same services often had different prices within the same quarter; this necessitated the development of weighted average prices. The unit prices from contracts and contract modifications, with the quarterly delivery schedules for each unit, were used to calculate the weighted quarterly prices. The sampling procedures for exchangeables (described in section 3) resulted in different samples for each period. In other words, the same items were not sampled over time because only two consecutive time periods were necessary to include an item in the sample. This necessitated the construction of a chain index for this series, rather than an attempt to follow the same pricing specifications over time. The chain index was calculated in the following manner. (1) For each item meeting the sample criteria, a price change was calculated. (2) Individual changes were then averaged to develop the aggregate change for the product class. (3) A simple average of the individual changes was used because it weighted each sampled item equally into the product class. (4) Because, for consecutive changes, the sample was not the same, consecutive estimates were linked together by multi- plying the prior period index by the current period change to derive the current period index. For example, in the base period, the index is 100. A 5-percent change in the second period would yield an index of 105 for the period. A 10-percent change in the third period would 80 yield an index of 115.5 (10 percent times the prior index of 105). This procedure was repeated in each period to generate the index for the given period. Resources were not available to construct price indexes for Navy depot maintenance. Current-dollar expenditures for ship overhaul were deflated using the deflator series for Navy ship conversion in the shipbuilding category. Expenditures for Navy aircraft depot mainte- nance were deflated by equally weighting the Army aircraft depot maintenance deflator with the Air Force aircraft depot maintenance deflator series. 6. Quality Quality adjustment in depot maintenance was tied to work statement changes for a given maintenance or modification program. In general, any change in the statement of work was interpreted as a quality change. For example, quality increase on an aircraft overhaul program was defined as a work statement change that resulted in an increase in the amount of work that was done on an individual aircraft. Quality adjustment of contract prices for depot maintenance or modification was necessary in two cases. The first case, a work statement change caused by modification of the contract, usually contained a dollar value for the modification. The price per unit was then adjusted to reflect only the initial work statement. The price per unit of the original contract was then directly compared with the price per unit of the modified contract. In the second case, in which an entirely new contract was let for the required maintenance tasks, a direct price comparison could not usually be made. Since no dollar value could be tied to work state- ment differences in the two contracts, the price negotiation special- ist was asked to estimate how much of the item price change was caused by quality change and how much was caused by price change. For exchangeables, it was generally assumed that maintenance per- formed under a given 13-digit code did not change significantly in consecutive periods. Examination of the results of the calculation routine, however, indicated some extreme price changes that were probably caused by more than "pure" price change. In these cases, the work statement was probably changed (quality change) or unique repairs (work over and above the standard recurring work) were performed on the item. Initially it was assumed that, at an aggregate level, there would be an approximately equal number of increases and decreases in cost due to these factors, with zero net effect. However, examination of the data indicated that increases outnumbered decreases in unit costs due to quality factors. This was substantiated by statistical tests conducted on the percent changes in unit price for all sampled items from FY 1975 to FY 1976. These tests indicated that the changes were significantly skewed toward high percentage changes in price. Elimination of all sampled items with percentage price changes larger than two standard deviations from the mean resulted in a stable 81 population in which large increases were balanced by large decreases. The items remaining after this adjustment—designed to eliminate bias due to quality change—were used in the calculation of the average percent price change of exchangeables. 7. Sources Air Force • Hill Air Force Base, Ogden, Utah § Kelly Air Force Base, San Antonio, Texas • Tinker Air Force Base, Oklahoma City, Oklahoma • "Uniform Depot Maintenance Cost and Production Report" (H036B), Air Force Logistics Command, (AFLC), Wright- Patterson Air Force Base, Dayton, Ohio Army • U.S. Army Depot Systems Command, Chambersburg, Pennsylvania • U.S. Army Aviation Systems Command, St. Louis, Missouri • U.S. Army Tank and Automotive Readiness Command, Warren, Michigan Electronic equipment 1. Definition The electronic equipment category consists of DOD purchases of ground communications and electronic equipment and airborne communica- tions and electronic equipment (avionics). Communications and elec- tronic equipment includes initial equipment and spare and repair parts purchased by the Other Procurement appropriations of the military services. Avionics includes equipment provided as GFM to the airframe manufacturer to be installed on production aircraft, as well as initial and replenishment spare parts. Specifically excluded are avionic, communications, and electronic equipment purchased from or by the stock funds. This category is classified as durables in the NIPA's; it accounted for 1.5 percent of total DOD purchases in CY 1977. 2. Concepts The DOD contracting procedure determined the unit to be priced. Avionic, communications, and electronic equipment may be purchased as 82 end items or as components of a larger system. In all cases, the unit of purchase, whether end item or component, was the specification priced. Occasionally major complete systems are purchased from one con- tractor. These systems are composed of many different types of communi- cations and electronic and avionic equipment. For example, the Air Force contracted with a single company for delivery of the complete Cobra Dane system. This system consisted of radar, computers, communi- cations equipment, teleprocessing equipment, and construction of a building to house this equipment. This system was designed for a specific function and will not be repurchased; however, it may be replaced by some other system in the future, and spare and repair parts will be purchased for it over time. If a complete system is frequently replaced, it could be priced over time, but these large systems are usually modified instead of replaced. Expenditures for systems that could not be repriced were deflated by a mix of components most closely equaling the makeup of the system. For example, if the total value of a system was made up of 70 percent radar equipment, 20 percent radios and associated equipment, and 10 percent computer, teletype, and associated equipment, expenditures on that system were deflated by a weighted composite deflator developed from deflators for those three product classes. 3. Sample Avionic, communications, and electronic equipment for each of the three military services was stratified into two subcategories: ground communications and electronic equipment, and avionics. These sub- categories were further stratified by the military command that made the purchase. Because of time limitations, data on Navy and Army purchases of avionics for installation in new production aircraft could not be separately identified, and were included with purchases of new production aircraft. Purchases were stratified for sampling as follows: A. Ground communications and electronics equipment Air Force • AFLC t AFSC Army • Communications and Electronics Materiel Readiness Command (CERCOM) Navy • SPCC • Naval Sea Systems Command (NAVSEA) , NAVAIR, and Naval Electronics Systems Command (NAVELEX) 83 B. Avionics Air Force t AFLC To maintain quarterly observations over time, it was necessary to select as large a sample of small electronics items (e.g.* radios, small radar sets) as possible. This was because the military usually purchases an entire requirement of a specific item including spare and repair parts, over a short period of time, for example about 3 years. After this, the item is seldom reordered. Purchases by the AFLC, CERCOM, and SPCC are of this nature. Item sampling frames were developed from computer files containing records of purchases coded by National Stock Number (NSN). After deletion of items that were one-time purchases, the value of deliveries was developed by multiplying the contract unit price by the quantity delivered and summing all records for like items. To ensure a repre- sentative sampling frame, deliveries for the entire period FY 1974-76 were included. Overall sample sizes were determined by the estimated workload relative to the resources available: AFLC - 250 items CERCOM - 150 items SPCC - 150 items Individual items were grouped by four-digit Federal Supply Class (FSC) on the assumption that the price movements for similar items would be similar. Items in each FSC were ranked highest to lowest on the basis of total value of deliveries for FY 1974-76. The sample size for each FSC was assigned judgmental ly within the limits of the overall sample sizes. Items were selected according to PPS from within each FSC. Sampling of AFSC purchases involved a two-stage process. Major system programs were ranked highest to lowest by expenditures for FY 1974-76. PPS sampling techniques were then used to select a sample of 7 projects. Purchases of complete systems from one contractor were not sampled further. Purchases of system components from different contractors were sampled judgmentally. The main selective factors were: repetitive purchase, nonclassified data, or major input into the entire system. NAVSEA, NAVAIR, and NAVELEX programs were ranked by classes--such as ship radars, ship sonars, electronic warfare equipment—highest to lowest on the basis of total expenditures for FY 1974-76. PPS sampling techniques were used to select a sample of programs within each class and a sample of items from these programs was selected judgmentally. The main selective factors were: repetitive purchase, nonclassified data, or major input into the entire system. 84 4 . Weights Quarterly weights for the total electronic equipment and for each of the commands were based upon actual disbursements as recorded in financial data reported by the military services. Weights for the FSC's and the items were based upon the calculated value of deliveries derived from the various procurement history files. The value of deliveries for unsampled items was allocated within the FSC propor- tionally to the sampled items. This procedure assumed that prices of the two moved similarly. Weights for each of the major systems selected were based upon actual disbursements. Weights for systems not selected were allocated proportionally within each command to the systems that were selected. Within each system selected, weights for individual items were based on their relative importance in the system. Again, items not selected were allocated proportionally to the items that were selected. 5. Prices The NSN identifies each item at a detailed level that could be used as the specification. The contract price for each item, adjusted for contract modifications, was used as the price. The average quarterly price was derived by summing the quarterly quantity of items delivered and the quarterly value of deliveries for each NSN and dividing the value of deliveries by the quantity delivered. Data on contract prices were available either from item managers within each of the military commands or from their automated data systems. 6. Quality Determination of quality adjustment procedures for electronic equip- ment presented problems. In many cases, technological changes are not made to a specific item during its useful life. in the field. A given specification for a radio, for example, will be used for years and the configuration of that radio will not change, even though technological changes are occurring. Configuration is kept constant to facilitate maintenance in the field. If the configuration were changed constantly, the logistics problem on repair manuals, spare parts, etc., would outweigh the benefits to be gained by using the very latest technology at eyery point in time. The military, therefore, usually implements quality changes when it purchases a new model; the desired new tech- nological developments are then incorporated into the specification for the new model . Statistical problems occurred at the point of linking the new model back to the old model. In many cases, it was difficult to determine exactly which model was replaced, since the new model may have replaced several older models or a larger system that consisted of several separately purchased components. In addition, it was difficult to 85 cost out each of the quality differences, since they were often quite numerous and not specifically tied to a part of the system or the component in a prior time period. In some cases, therefore, the new model was treated as an entirely new product: It was linked in with no price change. This procedure implicitly treated any difference in price of the new product as a difference in quality. This procedure is similar to that used in the PPI and CPI when an explicit quality adjustment is not made. Some items, however, do not change radically over time. In these cases, quality change is identifiable and could be treated as in other categories. The major criterion used to determine if there had been a quality change was to determine whether or not there had been a specifi- cation change. Military specifications at v the NSN level are so detailed that it is unlikely that there could be a change to the item in question without a change in the specification. If a specification change occurred, a change in cost caused by the specification change was determined. If increased quality was associated with higher cost, or decreased quality with lower cost, the unit price was adjusted accord- ingly. In general, the major source of data for quality adjustment was the ECO. For every physical change affecting a piece of equipment, the ECO stated a dollar (cost) increase or decrease that affected the price of the item. The formula employed for adjusting data for quality change was as follows: Let P^ = base price P = unit price in period x P , = unit price in period x + 1 Y = price change associated with quality change Z = pure price change Then: P x + , - P x - Y ♦ I To adjust for quality change, price relatives were developed so that: P x P + Y X 3 p b New P b 7. Sources Air Force • "J041 Procurement History File" provided by each Air Force Logistics Center 86 • ESD, AFSC, Hanscom Air Force Base, Boston, Massachusetts t "Financial Status Report" HAF-ACB(M)7130, Pentagon Army • "Military Standard Contract Procedure Report," CERCOM, Fort Monmouth, New Jersey • "Appropriation Status by Fiscal Year Program and Subaccounts" DD Comp(M)-1002, Pentagon Navy • "Procurement History File" and "Contract Status File" SPCC, Chambersburg, Pennsylvania • "Appropriation Status by Fiscal Year Program and Subaccounts," CM) 1002, NAVCOMPT Form 2158, Pentagon • NAVSEA, Arlington, Virginia • NAVAIR, Arlington, Virginia t NAVELEX, Arlington, Virginia Installation support services 1. Definition The installation support services category consists of services that are related to the maintenance and operation of military installations and that are not included in a separate category. Rentals of equipment, except ADP equipment, and space are included. This category includes a wide variety of services related to the support of military installations. These services range from janitorial, grounds and building maintenance, minor construction, and repair of equipment to training. Certain services normally associated with a military installation are included in other categories, such as pur- chases of utilities and communications services. Purchases of installation support services are classified as ser- vices in the NIPA's; they accounted for 2.6 percent of total DOD purchases in CY 1977. 2. Concepts Difficulties arose in defining specifications for some of the product classes within this category. Because the services purchased were so varied, and because military procurement offices purchase independently of one another, there were no consistent terms of contract for purchases 87 of a particular service. It was, therefore, impossible, in many cases to define specifications to cover all members of the sample for a given type of service. For example, the terms of contract for custo- dial and grounds maintenance specified (1) a price for the complete job, (2) a price based on dollars per hour worked, or (3) a price based on a rate per square foot of flooring or square mile of grounds maintained. In the first case, it was generally impossible to define the job in enough detail to price it over time. In the second case, a price per hour worked specification was definable and could be priced over time, if the contracts for the following years were stated in similar terms. The third case was similar to the second. As a result of these inconsistencies, specifications for each sampled contract were defined separately. 3. Sample The wide variety of services included in this category, the geo- graphic differences in purchasing patterns, and the large number of military installations within the United States, as well as abroad, created a large number of problems in constructing a sampling frame. Detailed data on expenditures by type of contractual services were not available in a centralized location, and resources limited data collec- tion to aggregate financial totals. Thus, a sampling frame could not be developed and a sample of specification-level items was not selected, Based on aggregate financial data, however, a preliminary set of seven product classes was established: A. Maintenance construction and minor construction B. Equipment repair C. Space rent D. Housekeeping services E. Equipment rent (except ADP) F. Contract operated installations G. Training and education These classes, as well as the specific items included within each, are subject to revision as the full sampling frame becomes established. A preliminary judgmental sample of several maintenance types of contracts was selected to test the proposed methodology. The test showed that a constant-dollar series can be prepared by using data from certain standard sections of contracts for a given service. 4. Weights Weights were also difficult to derive, for the same reasons as discussed for the sampling frame. It does appear, however, that product class level expenditures may be available by military service on an annual basis. Aggregate expenditures are available from each of the military services. 88 5. Prices Pricing was not attempted for the entire category because a sample was not selected. The judgmental sample of maintenance contracts, however, was used to test the methodology for pricing. The method- ology presented no unusual conceptual problems, and, as noted previously, the data indicated that adequate prices can be developed from contracts. Preliminary estimates of price change for this category were made as follows: Expenditures for installation support services, excluding rents, were arbitrarily split into two subcategories containing one- third and two-thirds of the total. Using the data collected from the sample installations, separate price indexes were created for custo- dial services, refuse collection, food services, housing maintenance, electrical plant maintenance, grounds maintenance, and mortuary ser- vices. Each series was then weighted by the current-dollars shown in "DOD Commercial and Industrial Contract Services Report." Using these weights, the price indexes for each series were combined and used to create a constant-dollar series for the subcategory containing one- third of the current-dollar expenditures. A constant-dollar series for the remaining two- thirds was created from a combination of BLS earnings indexes and PPI's for several types of producers' durable equipment. These two constant-dollar series were summed and divided into the current-dollar expenditures to create an implicit deflator series for support services. Space rent and equipment rent (except ADP) were deflated using the CPI for rent. Although CPI and earnings indexes data were not satis- factory deflators for some of the product classes, they were, given the time and resource constraints, the best available for these classes. Development of the full sampling frame and expansion of pricing to all product classes is currently under way and will be implemented as quickly as available resources permit. 6. Quality Quality adjustment for installation support services was related to changes in contract work statements. Any work statement change that affected cost was interpreted as a quality change to the service performed. Since general specifications were not used in this cate- gory, changes were related to contracts in the previous year for a specific type of service at a given installation. Where a work statement change occurred, the modification or adjustment was identified and a cost of the change was determined. The new contract was then adjusted to the same work statement of the previous contract and a price change for the same specification was calculated. 'This contract then became the basis for comparison for the contract in the following year, and the same procedures were used to determine quality change. Thus, an identical specification was not priced over time. The require- ments for this procedure were that the new contract could be adjusted back to the previous contract and that costs could be assigned to the work statement changes. 89 7. Sources Air Force • "Command Operating Budget Status Report" RCS-HAF-ACF(MQ)7148, Air Force Finance Center, Denver, Colorado Army t Department of the Army, Comptroller of the Army - Operation and Maintenance Program, Pentagon Navy t Navy Comptroller's Office, Operations Division, Office of Budgets and Reports, Arlington, Virginia • "Procurement Management Information Report," Navy Material Command, Arlington, Virginia Medical services 1. Definition The medical services category consists of all medical services purchased by DOD. Although the major portion of this category is CHAMPUS, it also includes purchases of medical laboratory services and contract physicians' services. Expenditures made to military hospitals are excluded because they are internal to DOD transactions. Military hospital purchases from the private economy are included in other appropriate categories. Purchases of medical services are classified as services in the NIPA's;' they accounted for 0.7 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts Form to CHAMPUS, and DOD will pay a percentage of the fees. These procedures are directly comparable to privately purchased major media insurance coverage. CHAMPUS coverage ceases at age 65 for individual: entitled to part A of medicare under the Social Security system. Jnder the CHAMPUS program, DOD is not providing medical care, but is icting as an insurer for users of the service. Under the concept of 90 the NIPA's, CHAMPUS expenditures are treated as DOD purchases of medical care and the expense of administering the claims. 3. Sample Purchases of medical services were stratified into three subcate- gories: (1) CHAMPUS purchases of medical care, (2) purchases of the expenses of processing these claims, and (3) purchases of other medical services, such as laboratory tests. CHAMPUS purchases were then subcategorized into 31 product classes, all of which were included in the sample. Also included were the administrative fees for processing claims. Purchases of other medical services were not included because they constituted less than 5 percent of the total and available resources were limited. Sampling was not attempted within the product classes because resources were limited and detailed data were unavailable. 4. Weights Data on payments to contractors for processing CHAMPUS claims were available on an annual basis. Quarterly expenditures, number of units purchased, and claims filed, by product class, were available. 5. Prices The basic CHAMPUS philosophy is that it will "share the cost of any medical procedure or type of medical care which is accepted as being part of good medical practice" within the limits of, and subject to, the eligibility requirements established by Congress and DOD. There- fore, the user does not need to specify the type of treatment for which CHAMPUS shares the cost. In addition, CHAMPUS does not place any restrictions on the definition of accepted, good medical practice. Consequently, the service for which CHAMPUS pays reflects the defini- tion of good, accepted medical practice within the community. DOD is purchasing a treatment, rather than the particular method used; there- fore, the average cost to the Government by product class can represent the price paid for that service. Prices of CHAMPUS payments were obtained by dividing the expenditures for each product class by the number of units (e.g., hospital day, doctor visit) purchased. The price therefore reflected the average cost for a hospital day, doctor visit, or prescription. Changes in the unit price occurred because of changes in the mix of the indivi- dual items purchased within a product class. For example, shifts between low- and high-average-value prescriptions would show as price changes. The definition of the unit being purchased, in this case, was "prescription drugs" and not a specific drug. It was assumed that the item purchased was a treatment, in this case a drug, and that the treatment was appropriate for the particular disease. Correspondingly, for physician outpatient care, the unit of measure was the doctor visit. The purpose or length of the visit was not 91 important since the unit being measured was the number of visits to a doctor. Therefore, changes in average cost of a doctor visit were treated as price changes to the DOD. The price for administrative fees was the average cost of processing a claim to DOD. The average price for processing a claim was derived by dividing quarterly data on payments to contractors by the number of claims processed. Price change for purchases of other medical ser- vices was assumed to move similarly to price change for CHAMPUS expenditures. 6. Quality No quality adjustments were made to the prices calculated for pur- chases of medical care. The average prices reflected the treatment or service purchased within the product class. Since all treatments reflected the definition of "accepted, good, medical practice," the quality of treatment was the same over time. 7. Sources t Cost and Budget Analysis Section, Office of CHAMPUS, Pentagon • Directorate of Management Services, Office of CHAMPUS, Fitzsimmons General Hospital, Denver, Colorado Missiles 1. Definition The missiles category consists of ordnance items commonly referred to as guided missiles, as well as related spare and repair parts, modification kits, certain launching devices, guidance radars, and shelters. Also included are contractual services that support the procurement of guided missiles and that are procured from the missile procurement appropriations. Purchases for the DOD space satellite programs are included. Purchase, repairs, and maintenance of sub- marines used to launch ballistic missiles, ship board launching systems, deployment costs, and research and development of guided missiles are excluded. Also excluded are purchases by or from industrially funded activities and stock funds. The construction portion of the Minuteman force modernization program is included in the construction category. Purchases of missiles, components, and associated equipments are classified as durable goods, and support services as services in the NIPA's. This category accounted for 2.5 percent of total DOD pur- chases of goods and services in CY 1977. 2. Concepts The unit purchase appropriate to measure the price change of missiles was determined by the contractual procedures used by DOD to purchase 92 different missile systems. While some missiles were purchased as complete units--e.g., the Army Dragon and the Air Force Maverick--most missiles were purchased as components. These components, such as the missile body, guidance and control unit, and rocket motor, were assembled into a final product by the main contractor or by a GOGO facility. In all cases, end items purchased from non-DOD sources were priced. Not all components were delivered to DOD during the same time period; therefore, a deflator for the complete system at any point in time would reflect the mix of items delivered in that period. Also, some of the components were produced or assembled by an industrially funded activity and were specifically excluded. Therefore, a deflator developed for the complete system based on a sample of commercially purchased components would be incomplete. In addition, in some cases, a complete missile system is composed of several items other than the actual guided missile--for example, launchers, test equipment, trackers, practice rounds, and protective shelters. These items are not generally purchased for each missile. A sample of these items purchased from non-DOD sources was also selected and priced. Introduction of new missile models into the series required the major assumption that the expected price of the 200th new missile was a valid overlap price to value the quality difference. This quality difference was then adjusted for inflation and added to the appro- priate old missile current series price, thereby linking the new missile series onto the old. Two key determinations in adjusting for quality were: was the new product a quality improvement? Was it at a higher cost? The Trident missile, for example, clearly costs more and outperforms the Poseidon. In these cases, the quality difference was valued and the two weapon systems were linked component by component. When quality improved at a lower cost, the improvement could not be measured and so the two series were compared directly. Linking the Trident and the Poseidon price series will involve a determination of the quality difference valued at an overlapping period--the one in which the first production contract was signed. This is the significant time period because the decision to sign the first production contract commits DOD to buy the new missile instead of the old missile. By using the expected price of the 200th missile, the learning curve effect on unit price expectations will be implicitly taken into account. Because of the learning curve, initial Tridents will be delivered at prices considerably higher than later ones. Consequently, the simple difference in overlap prices at the time of the first Trident delivery would be an overvaluation of the quality difference. Such an incorrect valuation of the quality difference would impart a downward bias to the price index and overstate the real quantity of defense purchases. 3. Sample Purchases of missiles were initially stratified into two subcategories: 93 long-range ballistic missiles and other missiles. The first subcate- gory was further stratified into the Intercontinental Ballistic Missile (ICBM) and Submarine—Launched Ballistics Missile (SLBM) product classes. All ICBM components were identified, and the follow- ing were selected: A. Attitude control group B. Guidance and control group C. Control exhaust nozzle D. Attitude injector E. Signal data converter F. First-stage rocket motor G. Second-stage rocket motor H. Third-stage rocket motor I. Post-stage reentry vehicle J. Onboard motor These components represented over 95 percent of the total value of ICBM components. The sample for pricing the SLBM was: A. Missile bodies B. Guidance and control units These components represented about 45 percent of the total value of SLBM components. The second subcategory was stratified into Army, Navy, and Air Force purchases of short-range ballistic missiles, nonballistic missiles, and associated equipment. For each military service, every missile system with an annual rate of expenditure larger than $5 million during FY 1974-76 was selected. In total, 13 different missile systems were selected. Within each missile system, every recurring specifica- tion with a total dollar annual volume of expenditure larger than $10,000 was selected. Nonrecurring items, such as special purpose equipment, were not priced. The exclusions had varying effects on the total amount of dollars actually being priced. For example, in the Dragon system, over 90 percent of the dollars expended were directly priced, while in the Hawk system about 50 percent of the dollars were directly priced. 4. Weights The current-dollar value of Navy missiles and Air Force other missiles was estimated for each quarter based upon a blowup of deliveries of the priced components. A two-stage blowup method was used to account for unpriced components. In the first stage, the sum of the unit prices for each sampled item within a system each fiscal year was related to the corresponding unit appropriation value. The resulting appropriation factor was used to blowup the priced items to the total 94 missile system. In the second stage, a blowup (max factor) was needed to estimate the current-dollar value of unsampled missiles included in a military service's appropriation. A max factor was derived for each fiscal year; it was the sum of appropriation values of sampled missiles divided into the corresponding total appropriation values. The max factor was then multiplied against the unit value result of the first- stage blowup. The product was an implicit current-dollar value per unit of delivered components. In any quarter, the sum of the product of delivered units and current-dollar values per unit became the estimate of the value of missile deliveries. The assumptions made in using the two-stage blowup were: all appropriated funds were spent, and unsampled items were delivered on the same schedule as sampled i terns . Quarterly weights for contractual services related to missile pro- grams, purchases for the satellite programs, and spare or repair parts and modification kits were derived from available expenditure data. 5. Prices Prices of components that constitute missile systems frequently are not identified unitl the rewards, penalties, and engineering changes are determined and negotiated at contract completion. A variety of prices may exist at any time for each physical specification; e.g., billing price, target price, initial contract price, estimated price at completion. Conceptually, the ideal price is one that reflects the ultimate total payment for the same physical specification in different time periods. Estimated prices at completion, derived from contract control documentation, were used whenever available. Even though these prices were only estimates at any point in time, they were the best available because, in addition to cost changes, they reflected profit adjust- ments caused by cost overruns or underruns. The nature of missile procurements and configuration changes made a combination of reports vital to ensuring that the same physical speci- fication was priced over time. Estimated prices at completion were derived from CPR's by prorating underruns, profit, and overhead across other contractor costs. Cost information reports were used to identify nonrecurring costs--not shown in CPR's--that, if included, would have distorted the recurring specification's price change. For missile components bought under firm fixed-price contracts, and not involving configuration changes, prices were taken directly from the contracts. Some components of the missile appropriations were deflated by related price series. Missile modifications, spare parts, support equipment, and facilities were deflated by price series developed from missile procurements that were directly priced. Space program expendi- tures (e.g., satellites), funded by the missile appropriations, were deflated by the R. & D. deflator. Classified missile expenditures were deflated by the overall DOD deflator less compensation. Total Army missile services were deflated by hourly rates and hours associated with service contracts for the Lance and TOW systems. 95 Minuteman services were directly deflated using hourly rates for recurring specified skills being purchased. Navy and other Air Force missile services were deflated by the Army missile service deflator. 6. Quality The developed price series relate to a recurring physical specifi- cation. To price the same physical specification over time, the price series were adjusted for configuration changes and changes in GFE. ECO's were helpful in evaluating the purpose of configuration changes and providing detailed cost estimates before negotiation. Contract modification documentation was an invaluable source of information about definitized costs of various configuration changes. Quality adjustments were made for physical changes that met the following criteria: A. The change was not made to correct a design deficiency or manufacturing failure; and B. The change affected range, speed, or accuracy; or C. The change affected safety of personnel; or D. The change affected length of service, need for, or ease of repairs; or E. A configuration change involved increased security, increased reliability, or integrity of components; and F. The change affected cost. No adjustment was made for changes in color, style, convenience or production compatability differences. 7. Sources Air Force § Minuteman System Program Office, Norton Air Force Base, California t ASD, Wright-Patterson Air Force Base, Ohio Army • Missile Readiness Command, Huntsville, Alabama Navy • Strategic Systems Project Office, Arlington, Virginia 96 NAVSEA, Arlington, Virginia • NAVAIR, Arlington, Virginia Other equipment 1. Definition The other equipment category consists of equipment, supplies, and materials that are procured with funds from the "Other Procurement" appropriations of Army, Navy, Marine Corps, and Air Force and that are not included, by definition, in any other category. Also included are supplies and materials purchased by industrially funded activities from non-DOD sources. Items in this category consist of many diverse products, such as airfield lights, pumps, compressors, photographic equipment, light craft (boats), and pollution control equipment. Specifically excluded are purchases from or by stock funds. Purchases of other equipment are classified as durables in the NIPA's; they accounted for 3.1 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts The DOD contracting procedure determined the unit to be priced in this category. Some pieces of equipment were purchased as complete units, while others were purchased as components. In both cases, the particular item that was separately purchased was the item priced over time. The items were usually well defined end items and presented no problems with regard to the concept of what was being priced. The items generally did not change drastically from period to period and could be readily priced using specification pricing techniques. 3. Sample The category was stratified into five major subcategories: Army purchases, Air Force purchases, Navy purchases by SPCC and the major commands, and purchases by industrially funded activities. SPCC purchases consisted primarily of spares and repair parts for the Navy: major commands purchases were new equipment. The procedure for the item selection for the Army, Air Force, and Navy SPCC subcategories was basically the same. Data were available from automated contract information systems that generally contained information on prices, quantities, delivery schedules, etc. Development of the item sampling frame required the deletion of items purchased only once, as they could not be priced over time. Next, the value of deliveries for each item was developed by multi- plying contract unit price times quantity delivered and summing all records for identical NSN's. To ensure representativeness, delivery values for the period FY 1974-76 were included. On the assumption that price movements for similar items were the same, items were then 97 grouped by four-digit FSC. Items within each FSC were ranked highest to lowest, based on their total value of deliveries for the period. A sample size for each subcategory was: Air Force--75 items Army- -75 items Navy- -75 items Within each subcategory, the number of items available for selection were allocated among each FSC on the relative proportion of FSC delivery value to total delivery value. Within each FSC, items were selected PPS; the size was the relative value of deliveries. Selection of the item sample for the Navy major command subcategory was a two-stage process. In the first stage, programs (subheads) were selected from "Appropriation Status by Fiscal Year Program and Sub- accounts," NAVCOMPT Form 2158. The selection was on the basis of PPS; size was expenditures for FY 1974-76. A sample of 15 programs was selected from "Other Procurement." In the second stage, a judgmental sample of items within each pro- gram was selected from the "Navy Apportionment Book." The general criteria were (1) size of item procurement and C2) repricing possibilities, Time and resource limitations did not allow sampling of purchases by industrially funded activities. 4. Weights Quarterly weights for each of the subcategories were based upon actual disbursements as recorded in financial data reported by the military services and the industrial funds. Weights for the FSC's were based upon the value of deliveries for each derived from the various procurement history files. Item weights were also based upon calculated delivery values from the procurement history files. Weights for unsampled items within a FSC were distributed proportionally to the sampled items within that FSC. This procedure assumed that prices of unsampled items moved similarly to those of the sampled items within the FSC. 5. Prices The NSN was used to track pricing specifications in this category. The contract price for each NSN, adjusted for all modifications, was used as the item price for each sampled item. The average quarterly price for each item was derived by summing quarterly item deliveries and quarterly delivery values for each NSN and dividing the delivery value by the quantity delivered. Data on contract prices were avail- able either from item managers within the military commands or from the automated data systems. Price change of purchases by industrially funded activities was assumed to move similarly to that of other items in this category. 98 6. Quality Determination of quality change was based on determination of a specification change. Military specifications at the NSN level are so detailed that is is unlikely that an item could have been changed without a change in the specification. If a specification change occurred, the resulting change in cost was determined. If increased quality was associated with higher cost, or decreased quality with lower cost, the unit price was adjusted accordingly. In general, the major source of data for quality adjustment was the ECO which, for every physical change affecting a piece of equipment, stated a dollar increase or decrease in the price of the item. The formula for adjusting data for quality change was as follows: Let P b = base price P = unit price in period x P x + -i = unit price in period x + 1 Y = price change associated with quality change Z = pure price change The " : P x + 1 - P x ■ Y + Z To adjust for quality change, price relatives were developed such that: P P + Y r x r x P b New P b 7. Sources Air Force • "J041 Procurement History File" provided by each Air Force Logistics Center • "Financial Status Report" HAF-ACBCM)7130, Pentagon Army t "Military Standard Contract Procedure Report" provided by each commodity command • "Appropriation Status by Fiscal Year Program and Subaccounts, 1 DD Comp(M)-1002, Pentagon 99 Navy • "Procurement History File" and the "Contract Status File," SPCC, Chambersburg, Pennsylvania • NAVSEA item managers and project offices • "Appropriation Status by Fiscal Year Program and Subaccounts," (M) 1002, NAVCOMPT Form 2158, Pentagon Other services 1. Definition The other services category consists of all DOD purchases of con- tract services that are not specifically included in other categories. Included are rentals of ADP equipment, printing and reproduction, direct and indirect hire of foreign nationals, engineering services, contract computer programming services, and consulting services. Purchases of other services are classified as services in the NIPA's; they accounted for 3.0 percent of the total DOD purchases of goods and services in CY 1977. 2. Concepts Difficulties arose in identifying specifications for the various types of services purchased. Because the purchases were so varied, and because military procurement offices purchased independently of one another, there existed no consistent terms of contract for a particular service. In many cases, therefore, detailed pricing speci- fications applicable to all sampled items could not be identified. For example, sometimes contracts stated a price for an entire task and sometimes they stated a price on a dollar per hour basis. Services in the first case generally could not be priced over time. Services in the second case could be priced over time, subject to the condition that contracts for the following years were stated in similar terms. As a result of these inconsistencies, specifications for each sampled contract had to be defined separately. 3. Sample The wide variety of services in this category, the geographic differences in purchasing patterns, and the large number of military installations, within the United States as well as abroad, created problems in constructing a sampling frame. Detailed data on expendi- tures by type of service were not available centrally. Resource limitations prevented the derivation of data other than aggregate financial totals; thus, an item sample was not selected. Based on aggregate financial data, however, five preliminary product classes were established: 100 A. Consulting and engineering services B. Direct and indirect hire of foreign nationals C. Printing and reproduction services D. ADP services including rental of equipment E. Miscellaneous services These product classes, as well as the specific items included within each class, are subject to revision as the full sampling frame becomes established. An initial evaluation of several maintenance types of contracts indicated that a constant-dollar series can be prepared by using data from certain standard sections of service contracts. 4. Weights For the same reasons as discussed for the sampling frame, weights were also difficult to derive. It does appear, however, that weights at the product class level may be available by military service on an annual basis. Aggregate expenditures are available from each of the military services. 5. Prices Pricing was not attempted for the entire category since an item sample was not selected. To derive a measure of constant-dollar purchases, each of the product classes was deflated as follows: A. Consulting and engineering services. — These services are somewhat similar to R. & D. activities. As a temporary measure, therefore, current-dollar data were deflated using the wage and salary portion of the R. & D. deflator series. B. Direct and indirect hire of foreign nationals. — Master labor contracts were not available in time to be utilized for this product class; therefore, average wage rate changes for Japan and Germany were used. These two countries accounted for over 60 percent of total DOD expenditures in this subcategory. In addition, changes in exchange rates were treated as price changes. C. Printing and reproduction services. — Current-dollar expendi- tures were deflated using prices developed from U.S. Army purchases from the Government Printing Office for specific classes of publications. D. ADP services and miscellaneous services. — No directly relevant data were available; therefore, the deflator for installa- tion support services, which was based on a test sample of maintenance contracts, was also utilized to deflate expendi- tures for these product classes. 101 Deflators derived for most of the product classes in this category are based on non-DOD sources. These procedures were not entirely satisfactory; however, given time and resource constraints, the series offered the best estimates of price change for these product classes. Development of the full sampling frame and expansion of pricing to all product classes is currently under way and will be implemented as quickly as available resources permit. 6. Quality Quality adjustment for other services was related to changes in contract work statements. Any work statement change that affected cost was interpreted as a quality change to the service performed. Since general specifications were not used in this category, changes were related to contracts in the previous year for a specific type of service at a given installation. Where a work statement change occurred, the modification or adjustment was identified and a cost of the change was determined. The new contract was then adjusted to the same work statement of the previous contract and a price change for the same specification was calculated. This contract then became the basis for comparison for the contract in the following year, and the same procedures were used to determine quality change. Thus, an identical specification was not priced over time. The requirements for this procedure were that the new contract could be adjusted back to the previous contract and that costs could be assigned to the work statement changes. 7. Sources Air Force • "Command Operating Budget Status Report" RCS-HAF-ACF(M0)7148, Air Force Finance Center, Denver, Colorado Army • Department of the Army, Comptroller Army - Operation Maintenance Program, Pentagon • Office of the Comptroller, The Adjutant General, U.S. Army, Pentagon Navy • Navy Comptroller's Office, Operations Division, Office of Budgets and Reports, Arlington, Virginia Government Printing Office • Office of the Comptroller, U.S. Government Printing Office, Washington, D.C. 102 Research and development 1 . Definition The R. & D. category consists of all contracts for the performance of research, development, test, and evaluation by the private economy for DOD. R. & D. performed within DOD, such as at DOD laboratories, is excluded. Except for contract R. & D., purchases by DOD labora- tories are deflated in other appropriate categories. Purchases of R. & D. are classified as services in the NIPA's; they accounted for 8.1 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts The output of an R. & D. effort is virtually impossible to define. The service purchased is constantly changing—there is a product, but a product that cannot be repriced over time. Evaluation of contracts indicates that there is no typical R. & D. task statement; as a result, there is no standard method of defining an acceptable measure of the output for R. & D. An alternative to pricing the output of a category is to price the inputs. This approach requires developing fixed specifications for inputs to the R. & D. effort and pricing these fixed specifications over time. Comparing the average price of a constant mix of input specifications over time provides a rough estimate of price change for the output. This is not an entirely satisfactory approach, because, in general, changes in the productivity of labor, changes in types of materials, and substitution of capital for labor cannot be included. In addi- tion, certain inputs (e.g., profits, depreciation) are very difficult to price or measure for an accurate estimate of the constant-dollar series. Input pricing, however, is the only feasible approach for estimating changes of the output prices of R. & D. In order to price R. & D. by fixing input specifications, each contractor's activity was segregated into three subcategories: direct labor, direct materials, and overhead. For direct labor and direct materials, units were defined that could be specification priced at a detailed level. The price for direct labor was wage rates expressed as hourly rates plus the associated fringe benefits. The rates did not reflect changes in employment mix or shift due to quantities of labor purchased at higher or lower wages; therefore, these factors did not influence the price. The only factors that did influence the price were the hourly rate and the increase or decrease in the cost of fringe benefits. For direct materials, standard items (i.e., raw materials, purchased parts) could be specification priced. Purchases of materials (com- ponents) requiring an R. & D. effort to produce were excluded from direct materials because the movement of their prices was more'likely to be similar to that of total contractual R. & D. 103 Overhead components were not priced. Instead, any change in the ratio of overhead charges to total costs of direct labor and materials in constant prices was defined as a price change. In other words, the quantity of overhead was directly proportional to the quantity of direct labor and materials. This procedure assumed that changes in the ratio of overhead to real material and labor did not contribute to changes in the level of real R. & D. purchased by DOD in any period. The real purchases of R. & D., therefore, reflected the movements of constant-dollar direct labor and direct material inputs with a fixed overhead ratio in support of the other two inputs. 3. Sample Data on material and labor inputs as well as overhead rates for R. & D. are not available from within DOD; therefore, they were obtained from private contractors. Contractors, rather than type of R. & D. performed, were sampled. This procedure was more efficient and yielded a wide range of relevant input data. The publication, "500 Contractors, Contractors Receiving the Largest Dollar Volume of Military Prime Contract Awards for RDT&E" was used to identify R. & D. contractors. Publications for 3 fiscal years (FY 1974-76) were used to increase the representativeness of the sample over time. Because of resource limitations, 25 contractors were selected. Eight of the top 10 contractors in each fiscal year were examined and selected with certainty. The other 17 contractors were selected using PPS techniques. To test the feasibility of the approach, data were collected from a test subsample of eight contractors. Recent approval of the data collection procedures by the Office of Management and Budget, will allow collection of data from the full sample of 25 contractors. 4. Weights Contractor weights were determined by using the value of prime contracts awarded to the top 500 contractors for FY 1974-76. The 8 contractors selected with certainty received approximately 45 percent of total contract awards, and the full sample of 25 contractors received approximately 63 percent of the total value. Each subsample contractor was assigned a weight equal to its share of contract awards during FY 1974-76. This weight was held constant for the entire period. The balance of DOD expenditures for contract R. & D. were allocated proportionately to the subsampled contractors. This pro- cedure assumed that the price change for inputs of all R. & D. con- tractors was the same as that for the 8 subsample contractors. 5. Prices Each contractor's expenditures for direct labor and direct materials were disaggregated into product classes, items, and ultimately into individual specifications. The 8 subsample contractors were asked to 104 provide direct labor data reflecting very narrow and detailed speci- fications (i.e., electrical engineer IV, draftsman level II) for employees directly working on DOD R. & D. projects. Each supplied wage rates and fringe benefit costs for a sample of employee speci- fications; these costs were used to create direct labor price indexes. These indexes were weighted together to develop a direct labor deflator, which was applied to the contractor's current-dollar labor expendi- tures to derive constant-dollars. Price indexes for direct material inputs (excluding materials requir- ing an R. & D. effort) to DOD R. & D. projects were created by using applicable PPI's. Initially, the subsample contractors were asked to provide actual price and quantity data for the materials used in R. & D.; however, the idea was abandoned when it proved difficult to obtain the data. Most subsample contractors provided a breakdown of direct material purchases by product groups as well as quarterly relative weights for each group. In addition, reporting contractors also supplied information on which PPI's they believed best represented price change for each group. The overhead factor, for each contractor, was derived by calculating the ratio of overhead to the sum of direct labor and direct materials for the base period. In each quarter, the sum of direct labor and direct materials (in constant-dollars) was multiplied by this overhead factor to determine overhead in constant-dollars. An overall IPD for each contractor was derived by aggregating the quarterly value of direct labor, direct material, and overhead, in current- and constant-dollars, and dividing the current-dollar sum by the constant-dollar sum. Some subsample contractors provided price and current- dollar data that applied only to a division, plant, or particular geographical location or a particular type of R. & D. activity; however, they assured BEA that the data were representative of the total effort for DOD. Deflators for these data were assumed to represent the total effort and were treated as total contractor deflators. Each contractor's overall quarterly deflator was applied to the appropriate quarterly current-dollar weight and the resultant constant- dollar series aggregated. The deflator for contract R. & D. as a total activity was derived by dividing total quarterly current-dollar expenditures by total quarterly constant-dollar expenditures. 6. Quality No quality adjustments were made: (1) The direct labor component prices were for detailed types of labor with constant specifications over time; (2) the PPI's used in deflating direct materials had already been adjusted for quality changes; (.3) overhead was defined as a constant ratio to the quantities of materials and labor purchased. 7. Sources % The principal data source for pricing contract R. & D. was 105 private contractors. The identity of these contractors and the data supplied by them are considered proprietary informa- tion by BEA. • "500 Contractors, Contractors Receiving the Largest Dollar Volume of Military Prime Contract Awards for RDT&E," Department of Defense, Office of the Assistant Secretary of Defense (Comptroller), Directorate for Information Operations and Control § "Status of Funds by Functional Title," Office of the Assistant Secretary of Defense (Comptroller), Directorate for Program and Financial Control • PPI's, BLS Retired military pay 1. Definition The retired military pay category consists of all DOD payments to retired military personnel and their survivors. These payments are considered to be transfer payments in the NIPA's; therefore, they do not enter GNP as Government purchases. To derive a constant-dollar series for DOD outlays, however, it was necessary to deflate these payments. 2. Concepts The concept of constant-dollar retirement pay presents serious problems. It is difficult, if not impossible, to imagine a quantity of retirement. Similar difficulties occur in defining and measuring what constitutes quality change. From the perspective of an employer, the main function of a retire- ment system is an inducement to hiring and retaining personnel. It is one of several fringe benefits added to the basic salary of an employee, During the period of active employment, the employer benefits from the work of the employee. The employer gains virtually nothing from the payments to retired personnel; thus, he sees the establishment and maintenance of a retirement system as an added cost of employment. From the perspective of an employee, the major benefit of a retire- ment system is received in the future and not during the period of active employment. But, the existence of a retirement system does have a definite bearing on an applicant's decision to accept or keep a given job. While the cash flow from the retirement system occurs to the employee during retirement, the system has a certain cash value to the employee in the present. Given a retirement system, an employee need not save as much current income as he would if no system existed; thus, the employee is, in fact, saving, even though the savings may not come out of current income. The discounted present value of 106 expected future benefits could be included as part of the current compensation received by an employee. A typical retirement system, as described above, leads to a particu- lar method for deflating current contributions to the system. Current- period contributions should be added to the wage and salary payments of an employee as a fringe benefit and deflated as part of current compensation. This procedure is the one that was followed for civil- ian DOD employees. The Federal contribution to the Civil Service retirement fund, for example, was deflated as part of the current salary of a GS employee. The DOD military retirement system, however, is not set up in this manner. Retirement checks are not disbursed from a contributory fund to former employees; instead, the amount paid out is funded by annual congressional appropriations. To maintain comparability with civilian retirement pay, the preferred method would be to calculate actuarially the present value of the future retirement benefits in each time period and add this to the current wage and salary payments to military personnel. Estimates of the amount of contributions needed to fund an actuarially sound retirement system could be calculated and added to the current salaries of military employees. These amounts, however, would not correspond to actual DOD outlays for military retired pay in any given period. The complexities involved in these calculations, as well as the fact that the data would not relate to current outlays, led to the abandonment of this approach. For the present study, an alternative methodology was developed that reflected any change in the rate of retired pay to an individual as a price change. Since retirement pay for each retiree is based on rank and length of service while on active duty, individual pay differs-- e.g., a retired general receives more than a retired sergeant. These differences in retired pay (at a given point in time) can be treated as differences in the quantity (or quality) of retired personnel. This assumption is consistent with the treatment of working military personnel, since the same quality factors are used to determine the pay while working and retired pay. Given this framework, the appro- priate definition of price change for retired military pay is a change in the rate for a given retired individual. Generally, the only change made to an individual's level of retired pay, after retirement, is the cost-of-living adjustment that is incorporated periodically. The adjustment is applied uniformly to all retired personnel; there- fore, only changes in the cost of living are necessary to reflect the appropriate measure of price change for retired military pay. 3. Sample Since cost-of-living adjustments are applied uniformly to all retirees, sample selection was not necessary. Only the periodic cost-of-living adjustment factor was necessary to develop the deflator for total outlays. 107 4. Weights Expenditures for retired military pay by each of the military services were obtained from the major military offices. Weight data were not required at a lower level of detail. 5. Prices Periodic changes in the rate of retired pay due to the cost-of- living adjustment factor were used as the measure of price change. Since each rate change is expressed as a percent, these percents were chained together to form the price index. The base period average rate was set at 100. Succeeding period changes were multiplied by the prior period index to develop the succeeding period's index. For example, the base period index was 100; if the next period's cost-of- living adjustment factor was 5 percent, it was multiplied by 100 to yield a deflator of 105.0. For the third period, if there was a 10 percent cost-of-living adjustment, the index of 105.0 would be multi- plied by 10 percent to yield a deflator of 110.5. Rate changes were assumed to apply to the entire month in which the change was effective. Thus, on a quarterly basis, a rate change effective in September would have one-third of the increase effective in the third quarter of the calendar year and the remaining increase effective in the fourth quarter. 6. Quality The criteria for quality implicit in this measure were the same as those for active duty military compensation; i.e., military rank and length of service. Since these criteria are set by active duty military service, it was not necessary to adjust the retirement pay for quality change. For any individual, quality change cannot occur since these factors are exogeneous to the retirement situation. 7. Sources Air Force § Air Force Finance Center, Denver, Colorado Army t Army Finance Center, Indianapolis, Indiana Marine Corps • Marine Corps Finance Center, Kansas City, Missouri Navy • Naval Finance Center, Cleveland, Ohio 108 Ship construction 1. definition The ship construction category consists of all new ship construction, conversions, and associated equipment directly funded from the Navy shipbuilding and conversion appropriation. Navy Yard conversions and purchases by or from stock funds are included in other categories. Approximately 5 percent of total purchases of ships and associated equipment (e.g., technical support, architectural drawings) is classi- fied as services in the NIPA's; the remainder is classified as durable goods. This category accounted for 3.1 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts Because of two characteristics of new ship construction, the pricing method used was different from that used for other weapon systems. First, ship construction is included in the NIPA's on a work-put-in- place basis, rather than a delivery basis. Second, many engineering changes are made on any given ship. Thus, even two ships of the same class, constructed consecutively, can be yery different by the time the second ship is launched. These characteristics make output pricing both difficult and inappropriate. Most attempts to construct prices for ship construction in the past have relied on the input method of pricing. Since the output could not be priced, the material and labor inputs were used to develop estimates of the output price change. This method, in general, has two major drawbacks: (1) to measure quality change and (2) any changes in productivity are extremely difficult to measure and incorporate into the estimate of price change. To circumvent both these problems, several ships under construction or just completed were selected and the physical designs were "frozen." The "frozen" designs became the specifications that. were repriced over time. Exact specifications were developed in conjunction with selected private shipyards, to ensure that the data on labor hours, labor rates, materials, and overhead costs related to appropriate designs that could be reconstructed over time. Once the basic data were developed for the base period for the "frozen" ships, they were then reestimated in succeeding time periods. The designs were held con- stant and the changes in labor hours, labor rates, materials and overhead necessary for producing identical ships in the succeeding time period were estimated. Because designs were identical in the two time periods, the aggregation of the second set of costs, compared to those of the base period, yielded the change in price for the identical amount of ship construction 1n the second period. This deflator can then be used to deflate work in process in shipyards. The "frozen" ship approach was taken because it had the fewest conceptual and practical problems. It also had two key conceptual advantages: inputs were specifically related to a defined output, and 109 current-period price changes were reflected in the output price deflator. The first advantage meant that the price effects of changes in production were measured implicitly. Since the inputs to produce the identical output were measured over time, input mix changes (because of productivity, labor, material substitutions, or any other cause) were reflected in the mix of the inputs priced and, therefore, in the output price that was estimated. Thus, productivity increases (or decreases) were reflected in the fixed output requirements by increases [or decreases) in the total number of worker-hours required to produce that output. The second advantage, that of reflecting current-period price changes in the output price, was important because" the NIPA's include naval ship construction on a work-put-in-place basis. The work-put-in-place approach has historically been taken because shipbuilding companies normally keep records of sales and profits on a put-in-place basis. An alternative approach focusing only on delivery price would not reflect current-period price changes during the construction period. A direct relationship between price changes of work put-in-place and what was actually measured by a "frozen" ship deflator, required 2 signficant assumptions. First, what was priced, i.e., a complete ship, was not what was actually constructed in any specific time period. Obviously, the material and skill mix required for each phase of construction varied from period to period; whereas, in the "frozen" ship approach the same mix was repriced in each period. Actual work put in place and the "frozen" ship would coincide. Differences- in price change because of this were implicitly assumed to be insignifi- cant in the deflators. Second, many configuration changes were made in each period, but they were not included in the pricing of the "frozen" ship. Price change for the configuration changes was assumed to move in the same way as that for the basic work done on the "frozen" ship design. It is believed that these assumptions did not seriously affect the deflator; therefore, the "frozen" ship approach was a practical solution to a very difficult pricing problem. 3. Sample New naval ship construction was stratified into three product classes: surface combatants, submarines, and surface support ships. Judgmental samples of ships and private shipyards constructing these ships were selected for each class. Surface combatants were repre- sented by guided missile frigates (FFG-7 class) and destroyers (DD-963 class). The submarines were represented by high speed nuclear attack submarines (SSN-688 class). The surface support ships were represented by fleet oilers (AO-177 class) and replenishment oilers (AOR-7 class). The seven shipyards providing data accounted for over 90 percent of all naval new ship construction and conversions performed in private shipyards. 110 4. Weights The current-dollar value of work put in place was estimated from expenditures. Gross disbursements were aggregated into surface combatants, submarines, surface support, ship conversions, and navy yard conversions. The true value of work put in place may be slightly different from the value of expenditures developed, because Navy expenditures generally lag behind the value of work in place. Expendi- tures may also reflect refunds, credits, or accounting corrections; these factors may somewhat distort the values from the true one of work put in place. Despite these shortcomings, expenditures are a reasonable and practical estimate of the value of work put in place. A value for purchased services was derived by taking a percentage of total expenditures based on relationships for selected programs in the shipbuilding and conversion appropriation. 5. Prices The price for each "frozen" ship design was a simple cost summation (hours times wage rates), overhead, and materials for the physical specification that was being priced. Shipyards were asked to provide labor hours, labor rates, and over- head for a "frozen" ship design, at the first level of detail within the WBS. The WBS for classifying the various ship construction phases includes such categories as hull construction, propulsion, electric plant, general outfit, and furnishings. Estimates for the base period and current-periods were made within this structure. Labor hours, labor rates, materials, and overhead were estimated for each WBS category or finer detailed levels, depending on the records kept in each shipyard. Although shipyards were generally cooperative, they could not always provide the desired data. Labor rates were relatively straightforward and easily provided. However, the change in labor hours proved more difficult. Changes in labor hours could have resulted from productivity gains or losses, Environmental Protection Agency requirements, Occupational Safety and Health Administration requirements, or many other factors that affected work loads. In spite of these difficulties, the shipyards were able to estimate the change in labor hours by WBS category sufficient for the purposes of this study. Overhead rates were generally those prevailing in particular time periods. However, they were not identical to those in new ship bids because the bid rates incorporate overhead estimates for future years. The overhead rates used were those prevailing for current-period Navy work put in place. As well as reflecting the changing environmental and safety requirements, the overhead rates in each period also reflected changes in general business conditions of the shipyards. Data on profits for the "frozen" ship proved impossible to develop. The long construction periods made it difficult to translate quarterly cost changes of work put in place into price changes. Price implicitly includes a profits component. Profits, however, relate to the entire 111 ship (i.e., ship contract), but the ship's actual price remains uncertain until delivery is made and the rewards, penalties, engineer- ing changes, and any lawsuits are settled. The consequent uncertainty about price and profits make it difficult to measure price change by tracking current costs. In addition, shipyards were extremely reluc- tant to discuss profits. As a result, any explicit incorporation of profits in "frozen" ship prices was abandoned and total ship price fluctuations caused by variations in profit levels were assumed to be relatively insignificant. The direction and magnitude of profits on these indexes remains a major measurement problem that will have to be resolved. An additional problem was related to material prices. Although the precise materials in a ship could be identified through the shipyards, good estimates of price change could not be prepared. Since materials are bought infrequently by shipyards, periodic pricing is difficult. In addition, some materials are bought in lots at the same price and time for several ships. Repricing a consistent material mix will misstate the value of work put in place if equipment is bought in lots or at the same price for many ships. Sometimes, actual material contract prices were constant over long periods because the shipyard paid a premium to buy out price escalation provisions from subcontractors Most material costs were associated with major equipment items, such as propulsion turbines and reduction gears, and general classes of goods such as pumps. The PPI did not contain data directly applicable to purchases of these items from marine vendors. In addition, BLS could not provide detailed estimates of price change for marine vendors without a very expensive expansion of the PPI price collection pro- gram. As a practical matter, therefore, existing PPI's for each "frozen" ship design were reweighted. Materials data collected from shipyards were used to weight the PPI's within each WBS category. The contract prices that shipyards actually negotiate with sub- contractors and the Navy are frequently escalated by the movement of PPI's. Consequently, until a contract is renegotiated, the price change for many materials, especially important pieces of equipment, will be similar to the movement of particular PPI's. In these instances, weighting the PPI's provided the correct measurement of price change. This approach for materials was an improvement over the usual approach because the unique material composition of each "frozen" ship had its own set and weighting of PPI's. In addition, the relative importance of any PPI can vary when its price changes relative to the average. For example, if steel prices increase faster than electrical machinery prices, steel will form a larger percentage of the "frozen" ship's price in subsequent periods. Ship conversions are frequently performed by the same private ship- yards that build new ships; however, the mix of materials and labor is different from new ship construction. Because of resource constraints, ship conversions were not directly priced. Expenditures, however, were stratified along the lines of new construction (i.e., surface combatants, submarines, surface support) and then deflated by reweighting new construction price changes for labor, materials, and overhead. 112 This approach attempted to recognize the input mix differences of new construction and conversion and to take advantage of shipyard price change data. 6. Quality The "frozen" ship approach, by definition, eliminated quality change by freezing the physical specifications being priced between time periods. Many engineering changes and design changes do occur in ship construction, however, and this will render the "frozen" ship design obsolete. It is expected that a "frozen" design must be replaced approximately ewery 5 years. Then, a new design must be selected and an overlap period priced for both ships, to provide a link period for the new design. 7. Sources • The principal data source for pricing ship construction was private shipyards. The identity of these shipyards and the data supplied by them are proprietary information to BEA. • NAVSEA, Arlington, Virginia • PPI, BLS Stock funds: Defense Stock Fund 1. Definition This category consists of purchases by the DSF. The DSF finances the acquisition of inventories of consumable materials and supplies for resale to the military services and other authorized customers (e.g., other Federal agencies) and also for stockpiles to be used in the event of mobilization. The six DLA supply centers are responsible for all items purchased by the DSF. The types of products purchased include large and small equipment items, petroleum products, expendable parts, repair materials, and a variety of troop-related items, such as food, medical and dental supplies, and clothing. DSF purchases were divided into four subcategories: clothing and textiles, bulk petroleum products, subsistence, and other supplies, materials, and equipment. A. Clothing and textiles. --This subcategory consists of clothing items, textiles and other items used in the manufacture of clothing, personal equipment, and household items. The Defense Personnel Support Center of DLA purchases over 95 percent of total DOD purchases of clothing and textiles; the remainder is purchased by the individual military ser- vices' stock funds. 113 Standard clothing issued to military personnel (initial Issue) is considered to be compensation-in-kind in the NIPA's; other special purpose clothing, personal equipment, and purchased textiles are classified nondurable goods. B. Bulk -petroleum products.— This subcategory consists of all bulk petroleum products purchased by DOD. The DLA, through the Defense Fuel Supply Center, is responsible for the integrated management of these purchases. The Defense Fuel Supply Center purchases over 95 percent of all DOD purchases of bulk petroleum products; the remainder is purchased by the military services' stock funds. C. Subsistence. --This subcategory consists of all food and food- related items purchased by DOD for issue to military per- sonnel. Food purchased by DOD and resold to commissary stores, and in turn resold to active duty and retired military personnel, is not recorded as Government purchases in the NIPA's, but as personal consumption expenditures. The DLA, through the Subsistence Directorate of the Defense Personnel Support Center, is responsible for management of these items. Food is issued to military personnel in messes, on field maneuvers, in military hospitals, and in DOD correctional facilities. In the NIPA's, the value of food issued in messes (about 50 percent of the total) is classified as compensation-in-kind and deflated with total compensation of military personnel. The remaining 50 percent of this subcategory is classified as Government purchases of non- durable goods and deflated as a separate series. D. Other supplies 3 materials, and equipment. — This subcategory consists of all other troop support equipment, expendable parts, and repair materials, purchased by DSF. Svawnary.— Approximately 80 percent of the total dollar value of DSF purchases are classified as nondurables in the NIPA's, and the remain- ing 20 percent as durables. DSF purchases accounted for 5.8 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts The DSF, as do other stock funds, finances the acquisition of materials and supplies for resale to the military services. It does this by maintaining inventories and selling items to a user, or possibly to another stock fund. Normally, customers pay for purchases with appro- priated funds, and the stock funds use these funds to replenish their inventories. Stock fund prices should cover replenishment of inven- tories. As a result, there are no annual appropriations for stock 114 funds; rather, stock fund managers are authorized to enter into con- tracts and purchase inventory items in expectation of resale. This authorization is based upon estimated demand figures submitted by stock fund customers in their annual budgets. Ideally, annual DSF purchases equal annual sales. The nature of stock fund transactions creates problems for deflating DSF purchases; namely, many supplies and materials purchased in one period may not be sold to a customer until some later period, and the price charged to customers is different from that paid by the stock funds. Stock funds pay the contractual price charged by the vendor. Depending upon the contract, this price may or may not include first- destination transportation. The standard price charged to customers for the identical good in the time period consists of (1) the basic price of the good based on the latest representative buy, (2) a standard first- and second-destination surcharge, (3) a standard surcharge to ensure against retail and wholesale loss, and (4) any other special surcharges. This standard price, which is generally changed only once a year, reflects purchase prices and conditions in a prior period. Use of standard prices to deflate current-dollar pur- chases by the stock funds would therefore (1) reflect price changes for prior periods, (2) show price changes only annually, and (3) reflect price changes based on other than price-determining char- acteristics; e.g., second-destination transportation and special loss surcharges. Development of current- and constant-dollar series for purchases made from directly appropriated funds would utilize purchases from the stock funds and not purchases by the funds. In this case, the standard price becomes the correct purchase price, in that it is the price at the point where price change measurement takes place. The various surcharges and prices constituting the standard price become the price-determining characteristics. The major objective was to develop current- and constant-dollar series for purchases by DOD from non-DOD sources. Stock fund pur- chases from stock funds (intra-DOD purchases) were excluded from consideration. Special treatments had to be used for the clothing and textiles and bulk petroleum products subcategories. Clothing and textiles. —In many cases, DOD purchases of clothing and textiles are clearly defined items, such as shoes, shirts, uniforms, helmets, which are delivered in a form ready for immediate use. These items presented no special pricing problems. In addition, DOD also purchases material, and cut, make, and trim services, the combination of which results in a finished product. For example, DOD purchases wool of a particular type, weight, and color, to ensure uniformity in military clothing. It then provides this wool as GFM to a cut, make, and trim shop, and contracts with the shop to produce a finished product (e.g., coat, man's, wool, serge, AG-44, Class 2), which is returned to DOD. Often the material is purchased in one time period and provided to the shop at some later period. 115 Clearly, DOD purchases two separate and clearly defined outputs- material and cut, make, and trim services—and not one finished product. The purchases were therefore priced separately. The items were classi- fied into homogeneous groups, specifications were sampled, and items corresponding to the selected specifications were priced over time. Pricing purchases separately made it extremely difficult to develop a price series for purchases of finished goods. First, the materials and cut, make, and trim services of similar finished goods were grouped separately. Second, there was no preselection to ensure that the material and services for a particular finished good were both selected, Third, in many cases, the material and services were purchased in two different time periods. Bulk petroleum products. --Prices paid for petroleum products vary between geographic regions. If disbursements for the same quantity of petroleum increased because procurement shifted from a lower to a higher priced area, a price increase was registered. The assumption was that no matter where petroleum products were being purchased, each unit of a given physical specification was the same. 3. Sample Samples were selected independently for each subcategory. Clothing and textiles. --The Defense Personnel Support Center manages about 23,000 individual clothing and textile items, each with its own NSN. These NSN's are grouped into approximately 7,500 generic items. One generic item could consist of many NSN's because each NSN is for a particular size of an item, whereas the generic item describes the item regardless of size. Where size was a price-determining character- istic, an individual specification (NSN) was used. Where size was not a price-determining characteristic, the generic item was used. DOD expenditures for clothing and textiles were stratified into six relatively homogeneous product classes, under the assumption that, generally, prices for similar items move together. These product classes were: A. Textiles B. Outerwear C. Special-purpose clothing D. Footwear E. Hosiery, handwear, and underwear F. Other Outerwear was further divided into two subclasses: 1. Outerwear involving cut, make, and trim services 2. Outerwear not involving cut, make, and trim services Generic items in each product class were grouped into several primary sampling units on the basis of item homogeneity. One item from each unit was selected on a PPS basis, where size was the total dollar 116 value of contract awards for an item in FY 1975. In units where the randomly selected item did not provide an acceptable frequency of observations, the item with the largest dollar value of purchases of the remaining items was selected. A total of 49 specifications was selected. In FY 1974, the total dollar value of the sample was 67 percent of the universe. Bulk petroleum ■products. --Bulk petroleum products were stratified into five homogeneous product classes, and, using total FY 1974 dis- bursements, the items within each class were ranked highest to lowest, One specification from each product class was selected on a PPS basis. They were: Product code NSN Nomenclature JP4 JP5 9130002568613 Turbine fuel, aviation, grade JP4, MIL-T-5624 9130002732379 Turbine fuel, aviation, grade JP5, MIL-T-5624 DFM 145 9140002732377 9130001791125 Diesel fuel, marine, MIL-F-16884 Gasoline, aviation, grade 115/145, MIL-G-5572 MG1 9130001601818 Gasoline, automotive, combat type I, MIL-G-3056 These items account for approximately 95 percent of bulk petroleum purchases. Subsistence.— Purchases of food and food related items were strati- fied into two groups: perishables and nonperishables. Perishables were further stratified into 10 product classes, and nonperishables into 18 classes. The time and resources were not available to select a probability sample from each class. Instead, a two- tier sampling procedure was employed. First, all product classes whose annual procurement value was larger than 1 percent of the total were selected, Nine product classes were selected from perishables (over 99 percent of total perishables), and seven classes from nonperishables (86 percent of total nonperishables). Together, these classes were over 96 percent of the total. Based on resource availability, 75 specifi- cations were selected judgmental ly. Approximately 20 percent of total food purchases was directly priced. Other supplies, materials, and equipment. — For each supply center, the items for which they had management responsibility were grouped into four-digit FSC groups; the assumption was that, generally, prices for similar items move together. Although a center may purchase an item for which it does not have management responsibility, only the managing center transactions were sampled. Sample sizes were deter- mined by available resources. For each center, PPS techniques were 117 used to select two replicated single-stage samples for each FSC. Size was defined as the value of DLA's reimbursable issues to its customers over the FY 1975-76 period. It was assumed that, over time, the mix of items DLA sells to its customers was the same mix DLA buys. 4. Weights Weights were readily available for bulk petroleum and subsistence, but not for clothing and textiles and for other supplies, materials and equipment. Quarterly weights for bulk petroleum products were derived from actual disbursements as reported by DLA. Purchases at cost were adjusted for discounts taken, transportation, and the change in accounts payable, to arrive at a total expenditure figure. The weight of each sampled item was based upon the disbursements of each sampled item in relation to the disbursements of all sampled items. The total value of food and food-related items purchased by DLA was available quarterly from DLA headquarters. The value of food furnished to military personnel in messes was derived from data provided by the Office of the Secretary of Defense. The remaining portion of this subcategory is classified as nondurable purchases in the NIPA's; it consists of total DLA purchases less food furnished to military personnel in messes. Quarterly weights for product classes and products in the subsis- tence category were based on the distribution of the value of con- tracts awarded, as published quarterly in "Military Contract Awards over $10,000." Quarterly sample weights were based on the relative procurement of each sample item to the total of all sampled items in each product grouping. Derivation of weights for clothing and textiles and for other supplies, materials, and equipment was different and proceeded as follows. The total value of deliveries of items for any particular period was not available in an aggregate form. However, since payment usually followed delivery within 30 days, total expenditure data were used as the total weight for each subcategory. (These data include transportation and repair expenses charged to the stock fund.) Value of deliveries and total expenditures were not available by FSC; therefore, the FSC distribution for obligations, adjusted for timing, was assumed to be the same as the FSC distribution for expendi- tures. Adjustments were made for the time between when contracts were let and deliveries began. Adjustments were also made for the length of contracts and the frequency of deliveries. For clothing and textiles, deliveries normally began 6 months after the contract was signed and were made monthly over a 6-month period. The distribution of expenditures by FSC for a given quarter was there- fore based on a weighted average of obligations by FSC for two and three quarters earlier. For other supplies, materials, and equipment, production lead time was generally 3 months and deliveries were made in the fourth month. The expenditure distribution by FSC for the current quarter was therefore based on obligation data for the prior 118 quarter. Percentage distributions of the derived expenditures were used as product class weights. For clothing and textiles, item weights were held constant and based on FY 1975 contract obligations for each item. For other supplies, materials, and equipment, item weights were based on the value of deliveries of the sampled items. Within each FSC, unsampled items were distributed in proportion to the sampled item. 5. Tviaes The contract price for each sampled item was used in all subcate- gories except bulk petroleum products. The current price for any sampled item for any quarter equaled the weighted average price for all deliveries of that item in that quarter. The base price equaled the weighted average price for all deliveries in the base year. For the subsistence subcategory, no further processing was necessary because the appropriate price for deliveries was available from the Subsistence Procurement Reporting System. For the clothing and textiles subcategory and the other supplies, materials, and equipment subcate- gory, however, price data on a delivery basis were not readily acces- sible; to relate price data to weight data, contract prices were lagged over the same time period as the FSC distribution of obliga- tions. The procedure was facilitated by the fact that most contracts were firm fixed price, and modifications did not usually affect price; however, adjustments were made when contract modifications did affect price. Price measurement was complicated by the fact that many items were purchased both f.o.b. (free on board) origin and f.o.b. destination. For f.o.b. destination purchases, the first-destination transportation was arranged and paid for by the stock fund. This transportation charge was included in the transportation expenses portion of total stock fund expenditures, but was commingled with second-destination; i.e., from the depot to the customer CArmy, Air Force, etc.). It was therefore extremely difficult to isolate transportation expenses for each purchase. An alternative procedure was developed to avoid the product mix type of price change resulting from f.o.b. origin and destination changes. For items consistently purchased both f.o.b. origin and destination, the current average quarterly price was computed by multiplying the prior quarter average price by the weighted change in both origin and destination prices from the prior to the current quarter. For items purchased either f.o.b. origin or destination, but not both ways at the same time, two distinct series were maintained. For bulk petroleum products, unit prices for each sampled specifica- tion were the quotients obtained by dividing total quarterly disburse- ments by total quarterly quantities. Figures for FY 1976 were avail- able in the "Contract Status Obligation Summary" by NSN; summaries for FY 1974 and FY 1975 were not available and so expenditure data were calculated from the Contract Price History Report, and the detailed buyer listing from Defense Fuel Supply Center. The report showed 119 contract prices, price modifications, and timing of modifications; the listing showed quantities shipped and the shipment quarters. Petroleum contracts were written either f.o.b. origin or destina- tion, depending upon costs, logistics, expediency, etc. The use of aggregate disbursements and quantities to develop average prices implicitly assumed that the mix of origin-destination contracts was constant. Most contracts with large refiners were f.o.b. origin, because MTMC and MSC determined that DOD-procured transportation is cheaper than that provided by contractors. Contracts with small refiners, who rely on close proximity to DOD customers for their competitive advantage, were f.o.b. destination. The mix of large- small refiners supplying DOD has not changed significantly during the past several years. 6. Quality The major criterion used to determine whether or not there had been a quality change was to determine whether or not there had been a physical change. The specifications for each NSN are so detailed that it is highly unlikely that any physical change could have occurred without a specification change. All specification changes were examined to determine if a change in quality had occurred, and the associated costs were determined. If increased quality was associated with higher cost, or decreased quality with lower cost, the price was adjusted according to the formula: Let P. = base price P = unit price in period x P x + 1 = unit price in period x + 1 Y = price change associated with quality change Z = pure price change Then, P x + 1 - P x - Y f Z Price relatives were determined so that: Px Px + Y P b New P b 7. Sources t Defense Personnel Support Center, Philadelphia, Pennsylvania t "Awards by Federal Supply Class" DLA Headquarters, Cameron Station, Virginia 120 t Defense Fuel Supply Center, Cameron Station, Virginia • Defense Construction Supply Center, Columbus, Ohio t Defense Electronics Supply Center, Dayton, Ohio t Defense General Supply Center, Richmond, Virginia • Defense Industrial Supply Center, Philadelphia, Pennsylvania Stock funds: Military services 1 . Definition This category consists of purchases by the military services' stock funds. The funds finance the acquisition of inventories of consumable materials and supplies for resale to the military services and other authorized customers (e.g., other Federal agencies) and for use in the event of mobilization. Purchases include large and small equipment items, expendable parts, and repair materials. All purchases by these stock funds are classified as durables in the NIPA's; they accounted for 2.7 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts Normally, customers pay for purchases with appropriated funds, and the stock funds use these receipts to replenish their inventories. The amount charged by the stock funds—the standard price—should cover replenishment of their inventories. As a result, there are no annual appropriations for stock funds; rather, stock fund managers are authorized to enter into contracts and purchase inventory items in expectation of resale. This authorization is based upon estimated demand figures submitted by stock fund customers in their annual budgets. The nature of stock fund transactions creates problems of deflation; namely, many supplies and materials purchased in one period may not be sold to a customer until some later period, and the price charged to customers is different from that paid by the stock funds. Stock funds pay the contractual price charged by the vendor. Depending upon the contract, this price may or may not include first-destination trans- portation. The standard price charged to customers for the identical good in the same period includes: (1) the basic price of the good based on the latest representative buy (2), a standard first- and second-destination transportation surcharge, C3) a standard surcharge to ensure against retail and wholesale loss, and (4) any other special surcharges. This standard price, which is generally charged only once a year, reflects purchase prices and conditions in a prior periods Use of standard prices to deflate current-dollar purchases by the stock funds would therefore (1) reflect price changes for prior periods, 121 (2) show price changes only annually, (3) reflect price changes based on other than price-determining characteristics; e.g., second-destina- tion transportation and special loss surcharges. Development of current- and constant-dollar series for purchases made from directly appropriated funds would utilize purchases from the stock funds and not purchases by the funds. In this case, the standard price becomes the correct purchase price, in that it is the price at the point where price change measurement takes place. The various surcharges and prices constituting the standard price would become the price-determining characteristics. The major objective was to develop current- and constant-dollar series for purchases by DOD from non-DOD sources. Stock fund pur- chases from the private sector therefore came within this scope, but appropriation-financed purchases from the stock funds did not and were excluded from consideration. Each military service stock fund is organized along wholesale and retail lines. The wholesale funds procure items that are unique to the military service; these funds are the major interface with the private sector. The retail funds also procure some items from the private sector, but mostly purchase from the wholesale fund. Resource limitations prevented the pricing of retail fund purchases; therefore, they were deflated by using the data for wholesale funds purchases. 3. Sample Wholesale stock fund purchases were divided into subcategories by military service. Each subcategory was then stratified by inventory control point because each inventory control point has management responsibility for procuring a large class of similar items. The total sample size for each military service stock fund was based on estimated workload in relation to resources. Allocation of the number of sample items among the various inventory control points was based upon the relative values of procurement. Each item purchased by stock funds is coded with an NSN. As stated previously, the specifications attached to these NSN's are so detailed that they can be used as the specification level for pricing. For each inventory control point, all items were ranked according to the total dollar value of deliveries for FY 1974-76. A sample of NSN's was selected from each inventory control point on a PPS basis where size equaled the total value of deliveries. 4. Weights The total value of purchases from the private sector for all military service stock funds was the aggregate weight for this category. The weight for each supply center was the value of its item purchases. The weight for each sampled item was the relationship of the dollar value of deliveries for each sampled item to the total dollar value of the sample within each inventory control point. The weight of the unsampled items was allocated proportionally to the value of the sampled items within each inventory control point. 122 5. Prices Air Force stock fund. --Price relatives were computed from contract data in the "J041 Procurement History File." Actual delivery data were unavailable; therefore, assumptions were made on the deliveries between the first and last delivery dates. On each contract where the first and last delivery dates for an item were identical, the calendar quarter in which that date fell was used as the quarter of shipment for the total quantity purchased under the contract. Different first and last delivery dates within the same quarter were treated similarly, Where these two dates were in different calendar quarters, it was assumed deliveries were made ratably between these quarters. The value of deliveries and the assumed delivered quantities were summed for each sample NSN for each quarter. A weighted average quarterly price, by NSN, was derived by dividing the value of deliveries by the delivered quantities for the quarter. Army stock fund.— Price relatives were computed from contract data in the "Military Standard Contract Procedure Report." For a given quarter, total delivered value and total quantities for each sampled NSN were cumulated. The weighted average quarterly price, for each NSN, was obtained by dividing the total delivery value by total quan- tities delivered for the quarter. Navy stock fund. --Price relatives were computed from contract data in the "Contract History File" and "Contract Status File." The ship- ment date was used to determine the quarter of delivery for purchases under each contract. In some cases, this date was unavailable and the schedule delivery date or contract was used. For a given quarter, total contract delivery value and total quantities delivered for each sampled NSN were cumulated. The weighted average quarterly price, for each NSN, was derived by dividing the value of deliveries by the delivered quantities for the quarter. 6. Quality The major criterion used to determine whether or not there had been a quality change was to determine whether or not there had been a physical change. The specifications for each NSN are so detailed that it is highly unlikely that a physical change could have occurred without a specification change. All specification changes were examined to determine if a change in quality had occurred, and the associated costs were determined. If increased quality was associated with higher costs, or decreased quality with lower cost, the price was adjusted according to the formula: Let P^ = base price P = unit price in period x P + i = unit price in period x + 1 123 Y = price change associated with quality change Z = pure price change Then - P x + 1 - P x n Y + Z Price relatives were determined so that: p x P x + Y p b New P L D 7. Sources Air Force • "J041 Procurement History File," Headquarters, AFLC, Wright-Patterson Air Force Base, Dayton, Ohio Army • "Military Standard Contract Procedure Report," Automated Logistic Management Systems Agency, St. Louis, Missouri Navy • "Contract History File" and "Contract Status File" from SPCC, Mechanicsburg, Pennsylvania and Aviation Supply Offices, Philadelphia, Pennsylvania Transportation of things 1. Definition The transportation of things category includes all DOD expenditures for the transportation of things and for the care of the things while they are being transported. It includes rental of trucks and other transportation equipment from commercial suppliers and Government motor pools, and reimbursement of Government personnel for authorized movement of household goods, privately owned vehicles, and house trailers. This category excludes transportation paid by a vendor, regardless of whether or not the cost is itemized on the bill for commodities sold. All purchases of transportation of things are classified as services in the NIPA's; they accounted for 2.2 percent of total DOD purchases of goods and services in CY 1977. 124 2. Concepts The purchased services are those necessary to move, or to facilitate the movement of, a certain commodity or class of commodities between two points. DOD uses various instruments and procedures to calculate the amount payable to the private transporters. It most frequently uses detailed rate schedules (e.g., ICC tariffs) that specify the amount to be paid to transport a certain commodity or class of com- modities between two points, the method of transportation to be used, and all of the auxiliary transportation services to be provided. In other cases, DOD uses contracts, each more or less unique, that specify the transportation services to be provided and the method of payment. In both cases (rate schedules and contracts), the transportation unit purchased is specified and priced. The transportation of things category was subdivided into five subcategories corresponding to the primary modes of transportation: air, rail, sea, truck, and terminal services. Rate schedules were generally used for purchasing rail and truck transportation and termi- nal services. Specific trips and related services were the pricing specifications, and prices were developed from the applicable rate schedules. The unit purchased was thus a specific trip, including origin, destination, and mode (rail or truck), or a specific terminal. Price change for transportation of things purchased in this manner was, therefore, determined by a change in the tariff for a specific trip. Individual contracts were used for purchasing air and sea trans- portation. The transportation unit purchased was determined by the contract. For example, MAC usually purchases ton miles, LOGAIR and QUICKTRANS purchase plane miles and commercial landings; MSC pays a specified amount per measurement ton by specific route for dry cargo shipments and a specified sum per unit of deadweight capacity for petroleum shipments. MAC and MSC are industrially funded activities that sell transporta- tion services to the various DOD departments. These sales are intra- DOD transactions, however, and therefore were excluded from considera- tion. In many cases, MAC and MSC subcontract particular transporta- tion services to the private economy. These purchases of transporta- tion are included as Government purchases and are priced in this category. Other purchases by MAC and MSC (e.g., fuel, compensation of employees) are included in other appropriate categories. 3. Sample Total DOD purchases of transportation of things were stratified by mode into 5 subcategories, 3 of which were subdivided into product classes: A. Air 1. MAC: Category A, category B, special assignment airlift mission 125 2. LOGAIR (Domestic rapid cargo transportation service provided on a scheduled basis via contract let by Air Force) 3. QUICKTRANS (Domestic rapid cargo transportation service provided on a scheduled basis via contract let by Navy) B. Sea 1. Break bulk (dry cargo) 2. Containerized (dry cargo) 3. Petroleum GOCO 4. Long-term consecutive voyage charter [petroleum) 5. Time charter (petroleum) 6. Bareboat charter [petroleum) C. Rail D. Truck 1. Domestic 2. Personal property (domestic and foreign) E. Terminal services Because a relatively small number of contracts were involved in some product classes, all contracts were included in the air subcategory and in the petroleum GOCO, long-term consecutive voyage charter (petroleum), time charter (petroleum), and bareboat charter (petroleum) product classes of the sea subcategory. Typical routes in each of the remain- ing two product classes of the sea subcategory were selected by MSC. Nine high frequency routes for break bulk (dry cargo) and three major commodities transported in each route were selected. For con- tainerized (dry cargo), the three major commodities on each of the 11 heaviest volume routes were selected. MTMC selected a sample of trips for the rail and truck subcategories. Based on frequency, 15 intercity domestic trips were selected for the rail subcategory and 5 trips were selected for the domestic truck class. (Beginning in the second calendar quarter of 1977, the total sample size was increased to about 50 trips.) For personal property, 12 heavy volume domestic trips and 12 heavy volume international trips were selected. Three terminal facilities were selected by MTMC to represent the terminal services category. 4. Weights Purchases from the private economy for transportation services were available from MAC and MSC. Expenditures for LOGAIR and QUICKTRANS were available from the Air Force and Navy, respectively. Purchases from the private economy for transportation of personal property 126 (domestic and foreign) and for terminal services were available from MTMC. Quarterly expenditures for rail and truck transportation on commercial and Government bills of lading were received from each of the military services and other DOD agencies. Within the rail sub- category and the domestic freight truck product class, the individual rates selected were weighted equally. 5. Prices For each specification, a price per unit was derived: A. Air 1. MAC: Average price paid per ton mile for each of three product classes 2. LOGAIR and QUICKTRANS: Price paid per plane mile and price paid per commercial landing B. Sea 1. Break bulk (dry cargo) and containerized (dry cargo): Average price paid per measurement ton for pier-to- pier ocean transportation 2. Petroleum GOCO, long-term consecutive voyage charter (petroleum), bareboat charter (.petroleum) and time charter [petroleum): Average price paid per deadweight ton of capacity C. Rail and truck: Average price paid per trip D. Personal property: Average price paid per hundred-weight [100 pounds) to move personal property from point of origin to point of destination, for domestic and for foreign E. Terminal services: Average longshoreman's straight time hourly pay for personnel provided by the contractor 6. Quality A change in the quality of the transportation of things would be determined by the changes in transit time (if it is considered impor- tant to the user) and in the condition of the goods upon their arrival at the destination. No data were available on these changes over the period; therefore, no quality adjustments were made. 7. Sources • MAC, Cost Analysis Division, Scott Air Force Base, Belleville, Illinois 127 t MTMC, Office of the Comptroller, Falls Church, Virginia • MSC, Statistics and Analysis Division, Office of the Comptroller, Washington, D.C. • LOGAIR, Transportation and Support Division, AFLC, Wright-Patterson Air Force Base, Dayton, Ohio • QUICKTRANS, Navy Material Transportation Office, Norfolk Naval Station, Norfolk, Virginia • Office of the Assistant Secretary of Defense (Comptroller), Pentagon • Directorate of Accounting Operations, U.S. Army Finance Center, Ft. Benjamin Harrison, Indianapolis, Indiana • Office of the Comptroller, U.S. Navy, Pentagon • Fiscal Analysis Division, Budget Directorate, U.S. Air Force, Pentagon • Fiscal Division, U.S. Marine Corps, Arlington Annex, Arlington, Virginia Travel and transportation of persons 1. Definition The travel and transportation of persons category consists of all DOD expenditures for transportation of Government employees, uniformed military personnel, and their dependents. Also included are per diem allowances, and other expenses incident to travel that are paid by the Government either directly or by reimbursing the traveler. Travel away from official stations and local travel and transportation of persons in and around the official station of an employee are included. Purchases of travel and transportation of persons are classified as purchases of services in the NIPA's; they accounted for 1.5 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts The purchased services are those necessary to move, or to facilitate the movement of, DOD civilian and military personnel from point of origin to point of destination. Included is the reimbursement of civilian and military personnel for certain specified expenses they incur while in travel status. Many expenditures in this category are made under published rate schedules (e.g., ICC tariffs) that specify the amount to be paid to transport personnel between two or more points according to the mode of transportation utilized and the 128 portation services provided. Published rate schedules also specify the amounts to be reimbursed to DOD civilian and military personnel for certain auxiliary expenses (e.g., per diem) they incurred while in travel status. In other cases, transportation is provided via con- tracts, each more or less unique, but all specifying the services to be provided and the method of payment. In both cases (rate schedules and contracts), the transportation unit purchased was specified and priced. The proper unit of purchase for travel expenses incurred under published rate schedules was the unit specified in the appropriate tariff. For example, in domestic air travel, specifications included the point of origin, the point of destination, and the type of service purchased (e.g., first class or coach). A price change was determined by a change in the rate schedule under which the service was purchased. The unit of travel purchased in the case of special contracts was determined by the individual contract. For example, in several international passenger transportation contracts, the appropriate unit of measure was the rate per passenger mile, since the specific origin and destination are not price-determining characteristics. Transportation services furnished by MAC to the military services or other DOD agencies are intra-DOD transactions and were excluded from this category. Most international passenger transportation purchased by DOD is through MAC, but is subcontracted by MAC to the private economy. These purchases are included in this category. Expenditures by MAC for transportation provided (e.g., gasoline, spare parts, compensation of employees) are included in other appropriate categories 3. Sample Total DOD travel and transportation of persons expenditures were stratified by mode of transportation (e.g., air, bus, rail) and by type of expense incurred and reimbursed to personnel by DOD. Each of these four subcategories was subdivided into product classes: A. Air 1. MAC: Category A, category Y, category B, special assignment airlift mission 2. Category Z (non-MAC international) 3. Domestic B. Bus 1 . Intercity C. Rail 1 . Intercity D. Reimbursable travel expenses 129 1. Limousines and taxicab fares 2. Rental cars 3. Dislocation allowance 4. Mileage (civilian, military temporary duty, military permanent change of station! 5. Per diem [civilian, military temporary duty, military permanent change of station! Relatively few specifications were required to price MAC purchases and international air transportation purchases under category Z; therefore, all were included in the sample. Domestic air and inter- city bus and rail transportation are purchased under legal tariffs. For these, the following number of trips were selected by MTMC (on the basis of trip frequency) for pricing: domestic air--100, intercity bus--100, intercity rail --4. Under reimbursable travel expenses, sampling was not necessary because of the limited number of rates involved for: dislocation allowance, mileage, and per diem. Resources did not permit sample selection for the limousine and taxicab fares and rental cars classes. 4. Weights MAC expenditures for international air transportation contracts were readily available from that command. The military services and other DOD agencies provided quarterly total expenditures for reimbursable travel expenses, which represent the weight for the remainder of this category. Expenditures for intercity transportation of persons were stratified by mode (bus, air, rail) on the basis of data from MTMC. Within each mode, the individual trips sampled were given equal weights to the total. For reimbursable travel expenses, a representative sample of 150 travel vouchers were selected for each of the permanent change of station and temporary duty categories for FY 1976. Expenditures from these vouchers were distributed into the 5 product classes of this subcategory. These distributions were used to determine the weights for permanent change of station and temporary duty reimbursable expenditures. 5. Prices For each specification, a price per unit was derived: MAC: Average price paid per passenger mile for four product classes Category Z (non-MAC international): Average price paid per seat mile Domestic air, intercity bus and intercity rail: Tariff price per specific trip selected 130 Limousine and taxi cab fares: CPI for taxi cab fares Rental cars: CPI for public transportation Dislocation allowance: Monthly basic quarters allowance rate Mileage: Specified rate per mile for each of the three specifications Per diem: Specified rate per day for each of three specifications 6. Quality Quality change in this category is related to the speed and comfort of the traveler. Quality change during this period of time was insigni- ficant; therefore, no quality adjustments were made. 7. Sources t MAC, Cost Analysis Division, Scott Air Force Base, Belleville, Illinois • MTMC, Office of the Comptroller, Falls Church, Virginia t Office of the Assistant Secretary of Defense (Comptroller), Pentagon • Directorate of Accounting Operations, U.S. Army Finance Center, Ft. Benjamin Harrison, Indianapolis, Indiana • Office of the Comptroller, U.S. Navy, Pentagon t Fiscal Analysis Division, Budget Directorate, U.S. Air Force, Pentagon • Fiscal Division, U.S. Marine Corps, Arlington Annex, Arlington, Virginia • Per Diem Travel and Transportation Allowance Committee, Pentagon Utilities 1. Definition The utilities category consists of DOD purchases of electricity, gas, water, steam, hot water, and sewage disposal. Utilities produced within DOD are excluded from this category, and inputs to these activities [e.g., compensation of employees, supplies) are included in other appropriate categories. DOD purchases of utilities are classified as services in the NIPA's; they accounted for 1.0 percent of total DOD purchases of goods and services in CY 1977. 131 2. Concepts The correct concept for defining price change in this category was not immediately evident. Two methods of price change were considered: (1) based on the average unit cost and (2) based on the rate schedule. The unit cost method defines the unit price as the ratio of the total expenditures for any given utility service in a period to the total units of that utility service purchased during that period. Changes in this unit price over time would reflect price change for utilities. For example, the unit price for electricity for any given period would be the total electricity expenditure divided by the number of kilowatt hours consumed during that period. This method assumes that each unit purchased (e.g., kilowatt hour) is identical. The unit cost method, however, may not accurately reflect changes in prices actually charged by utility companies. Two characteristics of the legal rate structure account for this. First, generally more than one factor is used in determining the price of a particular service. For example, electricity prices are generally based on a demand (capacity) charge, measured in kilowatts, and a consumption (use) charge, measured in kilowatt hours. In addition, high load factors are taken into account, and basic fuel adjustment rates can change. Consequently, measuring a unit price per kilowatt hour does not reflect the full specification of price-determining characteristics unless all other factors remain constant. Second, utility charges are usually based on rate schedules in which price per unit drops by steps as consumption increases. For example, a natural gas company may charge a customer $1.00 per thousand cubic feet (MCF) for the first 1,000 MCF, $0.80 per MCF for the next 40,000 MCF, $0.60 per MCF for the next 450,000 MCF, $0.50 per MCF for all consumption over 500,000 MCF. Thus, the average cost per MCF to a given customer could change, even if the basic rate rate schedule had not changed, simply because the customer had been consuming a smaller or larger quantity of natural gas in a specified time period. Levels of consumption could change, for example, due to the seasonal fluctua- tions in energy needs, an increase or a decrease in the population of a military installation, the phase-in or phase-out of an activity or plant at an installation. The unit cost method would reflect these changes in consumption patterns. To avoid these two problems, and to maintain consistency with other deflation procedures used in the NIPA's, the unit cost method of pricing was rejected and the rate schedule method was adopted. The rate schedule method of pricing required the development of a fixed specification for each utility service purchased by each installa- tion in the sample. The specifications take into account all the price-determining factors applicable to a particular utility service. In addition, they fix the quantities purchased so that changes in the quantities purchased are not registered as price changes. These specifications were fixed, where possible, by averaging the amounts, of the price-determining factors of the utility service consumed by the installation during the middle month of the four quarters of the base 132 year. These fixed specifications were then priced according to rates in effect during the middle month of each quarter beginning with August 1973. Although the fixed specification method results in the pricing of a fixed market basket of purchases, it is the same method used by the CPI and PPI, which have been used to deflate the utility services in other parts of the NIPA's. 3. Sample Total utility expenditures for FY 1974 by the military services and other DOD agencies were stratified by States and countries. These States and countries were then ranked in descending order of total expenditures on utility services. The 20 States and countries con- stituting the top 75 percent of the total utility expenditures were selected with certainty. The remaining States and countries were then further stratified into groups consisting of the 10 Census Bureau regions, Europe, Pacific, and Latin America. From each of these 13 groups, one additional State or country was chosen on a PPS basis. One group had been completely covered by the original selection, therefore, twelve States and countries were added to the sample. From each of these 32 States and countries, individual installations were selected with PPS based on the total expenditures of the installa- tions within each State or country. Three installations were selected from each of the 20 States and countries from each of the remaining 12 States and countries. Installations chosen more than once were not replaced, but their weights in the final index were multiplied by the number of times they were selected. The total sample selected consisted of 61 installations (10 chosen more than once): 24 Army, 19 Navy, and 18 Air Force. Each was asked to supply information on all utility services it purchased from non- DOD sources. 4. Weights Total fiscal year expenditures by installation and type of utility service were available from each of the military services and from other DOD agencies. These data were aggregated into the 32 State and country combinations. The percent distribution of this total was applied to DOD total expenditures for each utility service, to arrive at the aggregate State and country weights. Weights by individual classification were calculated as the percentage of the sample installa- tions within a particular State or country to the total for that State or country as calculated previously. This procedure yielded annual current-dollar weights by installation by type of quality service for the selected installations in the sample. Although the sources for the annual weight derivations did not contain quarterly data, three methods were investigated in an attempt to develop quarterly weights. The first was to distribute annual 133 expenditures over the four quarters of each year by using contract obligations data. Contract awards for each type of utility service for the Army and Air Force were available from the "Contract Awards Report," which contains quarterly data for all contracts over $10,000; however, the Navy does not report quarterly data. Also, contracts are awarded annually and are clustered during the year; therefore, this method of quarterly distribution was not satisfactory. The second was to distribute annual expenditures for each utility uniformly across four fiscal quarters. This method was unacceptable because of the seasonal shifts in demand for various activities. The third method was to use quarterly weights that yielded relatively smooth changes in the constant-dollar series over the year while controlling to the annual current-dollar totals. The series were allowed to fluctuate for expected seasonal increases in demand owing to heating and air conditioning expenses. This method was used, but was not completely satisfactory; alternative methods are under investi- gation and will be implemented in the revised series. 5. Prices As previously mentioned, utility rate schedules list different prices for different levels of consumption; however, the rate schedule pricing method required the development of a fixed specification and quantity for each utility service purchased over time. With con- sumption constant, it was possible to measure the effects of changes in all of the price-determining factors (rate schedules, fuel adjust- ment charges, load factor adjustments) on utility prices; changes that occurred in the average unit price because of changes in the level of consumption could be disregarded. The pricing specifications for each installation were derived by averaging the amounts of all price-determining factors of a particular utility service consumed by that installation during the middle month of each of the four quarters of the base year. This specification was then repriced from the rate schedule in effect during the middle month of each quarter from FY 1974 forward. Specific adjustments (e.g., fuel adjustment, load factor, purchased gas adjustments) were always taken into account, and taxes and facility charges were included where possible, but consistency of coverage by installation over time was the main criterion. 6. Quality Changes in quality of utility services are yery difficult to deter- mine. The quality of a kilowatt hour of electricity or MCF of natural gas would relate to the power derived from them. In most tariffs, prices of natural gas are stated in a form relating the British thermal units per MCF; therefore, as the quality of gas changes, so does the price charged to the customer. The price paid is a price per British thermal unit, and the data are quality adjusted. Quality adjustments were not attempted for electric power or the other utility services in this category. 134 7. Sources Air Force § "Civil Engineer Cost Report," RCSHAFPRE(SA)7101 , and "Family Housing Cost Report," RCSHAFPRE(SA)7102, Air Force Accounting and Finance Center, Denver, Colorado Army • "Facilities Engineering Annual Summary of Operations," USACE, Washington, D.C. Navy t "Utilities Procurement Report," NAVFAC, Alexandria, Virginia Other DOD agencies • Operations Directorate, Office of the Secretary of Defense, Pentagon Vehicles 1 . Definition The vehicles category consists of all DOD purchases of vehicles and associated spare parts and equipment; construction material handling equipment and special -purpose vehicles for missile programs are excluded. Purchases by or from the stock funds are excluded, as are vehicle components purchased from industrially funded activities (e.g., the gun tube for the M60 tank purchased at Watervliet Arsenal). Purchases by industrially funded activities are included in other appropriate categories. Vehicles are classified as durables in the NIPA's; they accounted for 1.2 percent of total DOD purchases of goods and services in • CY 1977. 2. Concepts Vehicles purchased by DOD are of three major types: combat, tactical, and commercial. Combat vehicles have specific combat roles and are built entirely to DOD specifications. Tactical vehicles are basically commercial vehicles with some enhancements required by DOD to fulfill a tactical role; e.g., the M880 1%-ton truck is a standard pickup type of truck that has been militarized to meet its tactical function. Commercial vehicles are standard production-li.ne vehicles with no military- inspired changes. The contracting procedure used by DOD determined the unit to be priced. Combat vehicles, such as the M60 tank, are purchased as 135 components and then assembled into a final product by the main con- tractor. Some of the components are purchased by the main contractor, and some are separately purchased by DOD and supplied to the main contractor as GFM. Consequently, selected components of a number of combat vehicles, rather than complete units, were priced. Tactical and commercial vehicles, however, are usually purchased assembled and were priced as complete units. Since DOD purchases combat vehicles as components rather than complete units, a deflator for the complete unit would be inadequate. First, components purchased from industrially funded activities are excluded. Second, the nature of combat vehicle procurement will result in erratic shifts in component deliveries due to changes in procurement mix. This product mix problem is compounded by the high percentages of foreign military sales purchases. These purchases lower the unit price to DOD, and are normally included in the same contract with DOD requirements. Although foreign military sales deliveries are excluded from the data used, changes in the level of activity can significantly affect the price paid by DOD and the delivery schedule to DOD. 3. Sample Generally, all DOD vehicle requirements, whether for the individual services or foreign military sales, are managed by the U.S. Army Tank - Automotive Materiel Readiness Command. The only exceptions are commercial sedans and light pickup trucks, which are procured by the General Services Administration. Time limitations and data -unavail- ability precluded development of deflators of purchases of commercial vehicles from the General Services Administration, therefore, the sampling universe was limited to vehicles purchased by the command. Total DOD purchases of vehicles were stratified into two subcate- gories: (1) combat vehicles, and (2) tactical and commercial vehicles. Combat vehicles were further stratified into three product classes: (1) medium tanks, (2) armored personnel carriers, and (3) recovery vehicles. Tactical and commercial vehicles were not stratified further. All vehicles within both subcategories were ranked by the total value of contract obligations for FY 1974-76. One combat vehicle from each product class was selected by using PPS techniques. Obligations for the three selected vehicles accounted for over 90 percent of total obligations for all combat vehicles during FY 1974-76. Specifications of separately purchased components for each of the sampled vehicles were selected based on their relative unit cost. Components purchased from an industrially funded activity were not considered for selection. The sample for combat vehicles was: A. Combat tank, M60 A1/A3 105MM gun 1. Vehicle 2. Laser range finder 3. Solid state computer 136 4. Engine 5. Transmission 6. Track 7. Gunner sight B. Recovery vehicle, medium M88A1 1. Vehicle 2. Engine 3. Transmission 4. Track C. Personnel carrier, M113A1 1. Vehicle 2. Engine 3. Transmission Due to the difficulty in obtaining complete price and delivery data on all tactical and commercial vehicles, 14 vehicles were selected on the basis of the availability of data and the periodicity of purchase. The sample of tactical and commercial vehicles was: A. Truck, dump 19000 GVW 4X2 B. Truck, dump 28000 GVW 4X2 C. Truck, tractor 4X2 D. Truck, refuse 39500 GVW E. Tractor, tank, water l^T, 400 gal F. Truck, wrecker 24000 GVW 4X2 G. Truck, wrecker 34500 GVW 4X2 H. Semi -trailer, low bed, 60 ton M747 I. Semi-trailer, electronic van, 6 ton, M373A2 J. Truck, % ton cargo 4X4 M880 K. Truck, 1% ton cargo 4X2 M890 L. Truck, h ton utility 4X4 M151A2 M. Truck, 2% ton cargo M35A2 N. Truck, 8 ton 4X4 cargo GOER 4. Weights Quarterly weights for the two subcategories (.i.e., combat vehicles and tactical and commercial vehicles) were based upon an estimate of the total value of deliveries of the vehicles and their associated equipment. Since progress payments in this category were relatively small, the estimate was based on disbursements from the Weapons and Tracked Combat Vehicles, Army, and Other Procurement, Army, appropria- tions. For each calendar quarter, the value of deliveries of each sampled item in relation to the value of deliveries of all sampled items in that subcategory was the weight of that item. This procedure assumed that, within each subcategory, the value of deliveries of 137 unsampled items was proportional to the sampled items. 5. Pviees Contract prices for each sampled specification, including all con- tract modifications, were used. The average quarterly item price was the weighted average price of all items delivered in the time period. The base price was the weighted average price of all items delivered in the base period. Deliveries of each sampled item were determined from data submitted by the appropriate project office for combat vehicles, and from the "Monthly Production Status Record" for tactical and commercial vehicles. 6. Quality The major criterion used to determine whether or not there had been a quality change was to determine whether or not there had been a specification change. Any specification change that resulted in a physical change to a vehicle was evaluated to determine if it affected the mission of that particular vehicle or component. Where a quality change did occur, the cost of the change was determined, and the price was adjusted appropriately. The major sources of information on quality changes were ECO's, which are required for all engineering changes made to an item with a military specification. ECO's generally include information on the nature of the change and its impact on the vehicle and an estimate of the cost of that change. Information on specification changes from period to period was available only for combat and tactical vehicles. Commercial vehicles are purchased under performance specifications, rather than detailed technical specifications. Specifications on commercial vehicles are determined by vehicle manufacturers together with their customers; there are no special military or DOD specifica- tions. Technical specifications therefore, changed from year to year and the associated dollar costs were not available. from DOD. 7. Sources • U.S. Army Tank - Automotive Materiel Readiness Command, Warren, Michigan Weapons 1. Definition The weapons category consists of DOD purchases of weapons and their components from non-DOD sources. Included are: small arms weapons, such as rifles, machine guns, automatic rifles, pistols, smoke generators, flamethrowers, and associated equipment; artillery weapons, such as heavy, medium, and light artillery, anti-aircraft artillery, recoilless rifles, mortars and associated fire control equipment; towed howitzers; and selected aircraft armament and components, such as 20mm machine 138 guns provided to airframe manufacturers as GFM. Components of large artillery weapons and nontracked combat vehicles are largely purchased through industrially funded activities and, therefore, are not included in this category. Also excluded are purchases from or by the stock funds. Weapons and their components and associated equipment are classified as durables in the NIPA's; they accounted for 0.1 percent of total DOD purchases of goods and services in CY 1977. 2. Concepts The contracting procedure used by DOD determined the unit to be priced. Some weapons are purchased as complete units, while others are purchased as components that are assembled into the final product. In both cases, the particular item separately purchased by DOD was priced over time. An index could not be prepared for weapons pur- chased as components because the data would be incomplete. First, some components are purchased from industrially funded activities and are excluded since they are intra-DOD transactions. Second, the mix of components delivered in any given time period may not be propor- tional to the makeup of any given weapon. Some weapon components may be delivered in one package during a given quarter, but other compo- nents of that same weapon may be delivered periodically over a period of several quarters. Thus, in any given time period, the deflators for deliveries of weapons components would not necessarily correspond to the composition of the complete weapon. 3. Sample Weapons, components, and associated equipment were stratified into two product classes: (1) small arms and artillery, and (2) aircraft armament. The majority of these weapons were procured by the Army Armament Materiel Readiness Command; therefore these purchases were used as the universe for sample selection. Since few weapons were procured during the period, a judgmental sample was selected from each product class. Periodicity of purchase and data availability were the main criteria for sample selection. Data for the GAU-8 gun (used in the A-10 aircraft) were available from the Air Force; therefore, the weapon was included in the aircraft armament sample. The weapons sample consisted of: A. Small arms and artillery 1. Grenade launcher, 40MM M203 2. Rifle, 5.56MM, Ml 6 3. Tripod mount, Ml 22 4. Machine gun, 7.62MM, M60 5. Tripod mount, 50 CAL N6, M3 139 B. Aircraft armament 1. Gun, auto, 20MM, M61A1 2. Gun, auto, 20MM, M197E1 3. Delinking feeder, M89E1 4. GAU-8/A gun 30MM 4. Weights The weight for each product class was the total expenditure for all items within each class. The weight for each sampled item was its relationship to all other sampled items within the product class. This procedure assumed that deliveries of unsampled weapons were proportional to the value of deliveries of sampled weapons. 5. Prices The price for each sampled item was available, and for a given quarter was equal to the weighted average price for all deliveries of the sampled item in that quarter. The base price equaled the weighted average price for items delivered in the base year. 6. Quality The major criterion used to determine whether or not there had been a quality change was to determine whether or not there had been a specification change. The detailed nature of military specifications made it highly unlikely that quality could have changed without a change in specifications. All specification changes were examined to determine if a change in quality had occurred and costs associated with changes were determined. If increased quality was associated with higher cost, or decreased quality with lower cost, the price was adjusted according to the formula: Let P. = base price P x = unit price in period x P x + 1 = unit price in period x + 1 Y = price change associated with quality change Z = pure price change Then: P x + , - P x = Y + Z 140 Price relatives were determined so that: P x P + Y X p b New ? b 7. Sources "Procurement of Equipment and Missile Management Accounting and Reporting System," Army Armament Materiel Readiness, Command, Rock Island, Illinois • "J041 Procurement History File," AFLC, Wright-Patterson Air Force Base, Dayton, Ohio • "Cost Performance Reports," ASD, AFSC, Wright-Patterson Air Force Base, Dayton, Ohio 141 •-: O «=t lo oc r^ cc r*> •s* «tf O «tf LO C cr LT CM CM 0> LO ro r-. >- r-- CO CO 00 c OOOlC r^~ o r^ c CM CO i— <^t t^ i — CO LO LO c_> o~> 5 ro lo i — o~ LO LO CM LO CTi CM CO LC LO LO CT> C^ LO «3" i — LO «3" «vt r- CM CT> C CTi i — ^1" LC O CM CO CM cr> co ^3- cr LO O CO i — CM lo >- r^ CM co *3- <3- r-» i— CM CO C CO CO LO c 00 00 LO LC r^ CO CM CTi LO c_> cr> c~ i — lo o c\j «*■ "vl" r— «tf LO i — CM lO co -^j- r^ cv «3" CM CT> «tf" *d" LC ro cm co r-v CM O CO LT ^t CO CO oc LO O CO c\ r— r~- - r^ ** CO CO CTi C ■ — cm O CO CT> CM CO CO i — c\. CO O r— CO CM co r^. CM <3- 1 — r^- cm co o i— r-- <^- cr> r-«. cm lo cri >3- lo co CM LO CO LT ONon^j- «d- >- r^ ur co en r— CO CM LO LO LC cr> r^. i — lo cm «^- i — cr r~» LO CM CT> LO O CT> Olr-Or- CM i— O i— i— CT> ■ — CM r-MNC CMO^ON LC CT^ CM LO CM CM Cn LO •=* CTi CO CM CO CO r— CM 0C NOr-OON CO >- r~» LC i-OOCOh- CM «d" r— r- LO CTi LO ^J- <* cn n it «3- cm r--~ >^- lo O CT> c o o cr> c r— O O C OOlOr- OOi— o ooooo c oooo o o o c o o o c oooo ooooo CM >- r^ c o o o c o o o c: o o o c: oooo ooooo O CT> c o o o c o o o c ) • 4- C ) • a ) •I- • 4- . • c 1 > ^ > o • »r- U S- > o a i • E $. 4-> 4-> or ) L0 O ■ a i c ' L0 OVr- > (/ 1 CD > « CD S- •r- ro 4-> 4 > o_ L0 re i SZ 4J e s. o • C 4- > 4-> S_ CD c ) 1— . CD -r- o . CD J ■ CD c: 4- r- 4- a. U CL C L LO > O (Dt O L0 __ c -r- r I CD . 4-> cd •!- -a E E «=t to ■ IO 3 1/ ) C ro r- > c • E cr •1 — • cd a 1 o O S- c i • o c E CD CD C : > E c ) "O 3 •• • •i- 4-> 1- . 'r- c ) o +-> c ) s- . Q--r- c S- w i/ ) 4-> E 4 •r- C U -r- CD • -r- > ro 4-> "O X i ro -a > O ro +- ) -t_> .r- .p- +. > to ■ 3 t L0 c c : 4-> a lo 4-> t- U ft 1 O fO C n ( to cr a ) -C C Z5 Z i &- (O CD LO 4_ +j .,- (/ ) =3 E O ■- i— CD CD c ) U O 4- M- O -i— CD 00 fO «r- E C S- S- f- ro r— s- u Q-i— 4-> i— E s- e rs a ) +-> 4-> +-> re ( O t- S- S. ro -^ _i : oo cd •!—(_> o O 3 E C L to O U +■ > t- «/> cu a 1 CD cl U L > E > r— •■- Q. S- E E E E CL CD C > -D wrx LO -r- O C ) ro ro t- _E ro •■- E O C » O CD i— C : CD -r- 4-> 4- > CD -C 4J 4- > S- S- 4-> CD CD «oc «3 : c_ i <_ > r_ ) c ) u j *- < s: s c > c ) cc : v ) cr ) • 3 E CD Cl CL o 4- , S- o CD CD 4-> ro a S- o ro E ^z a ro CD CD OO z. Cl E O O o _E OO 4-> E CD E O CL E O , x: en 4-5 O 4- *0 O -o o E o •I™" +-> Cl S_ o I/) CD T3 4-> CD -a CD no CD O CD 4-> O 143 r— i — CM CM C^lDOlOl tnor-io *i-mui r-. oo i— cm >=r cm en r~ r— ■=1- co cm >a- IDClOlO r-lOOW ID r— r— ID O CM ■* ID "* cm a- CM i— co r-. ^r CM CO LO O CM o •a o CM CM i— i— r~ r-. id >3- 3- en co in r-- co r-. in co oo co OlONN CTl CO ,<— i— XI II CM r- CM CM I — cm cm oo r- io«d-CMC0 mi — inin ID «* CM CO I — ai w en 0) •— v •i- >- > <_; s- ai 00 en en ID CO ID <* CO I-. O en «d- cm in i — r-- i — 2? o E -a csjioiD -i- t- +J W o +-> ra 10 O O +-> C O- CO 10 a> 3 S- lO CT CO 0) CO 10 (O i — U •!- S- i. •i- i/l CO CO -o io jz x: 10 10 >> co s- 10 f0 s= +-> CO •■- CO C 4- 1— > O CO ■!- co -r- -o E •o +-> o ■O 3" » c i. 10 in 10 +-> T3 "CJ 10 c c x: c 3 3 +-> a. i>o 00 co s- co Q. ■r- IO x: +J +-> s. o 1— C in CO ■■- u O c > 1— -r- a. fO cd 10 3 -C c u •1^ s- ■<-» 3 c a. o u Q O 1 (-1 >1 s_ s_ o o M- U> CD co +J s- m S- o 4- •1-1 CD m Q S tu >> cj J3 cj > •r- S- r- ai a. i/5 E i-i "O C I * CT> COOOlLfl O O CO 1^ r~. «d- i — en COCMOIN CTv (NJ LT> CO i — i — i — «=r cm co ■— i — ^ONCO Ni-fflOl «tir>m o en o CM cm co en *s- i — cm oincvjN "tWLDLO en en i — co o CM CM I — CM CO LO CO CO LOOtCO i — CO LT) O O^OCOCO en en co co i— cm r— en ld in CO CM co ^- i— en LflN CDNLO t— o c c cd cu c > o o +-> o s_ •■- •!- C CJ T- CD CD s- CO fO c +■> CU t- cn o OVi- c: +■> CJ ro C 03 O •— i— S_ C 3 CD u 3 E a. S- E E E ■■- E O O < Q.-I- •i- > 3 S_ co cr a) CD CD co IO i — O -r- S- S- i- 00 CD CD O CO JZ -C CD -r- +-> +-> s s: o o > O CD •!- oj-r-o E -a +-> CJ TJ 3 ••'•• C S- co CO (O +■> "O T3 • r— 4_ O- -C C 3 3 S- 10 CD CO U O 4- 4- o CU CO i- U a. 4J c ro -* _^ CO cd •r— CJ o cu a. o CJ c > i — ■F— ex CO *i— o n •O +-> S- s- +J CD CD or c/0 oo CO h- 1— _) > S o a. E 145 CO CL -o S- E O O00Oi<*N'!j <* i— i— CM o o r^. cm «0 O O r-« r«- cm cm «3- CTi O O O i— CM CTi CM -o >> CU O CTi 00 CX) O CM en o o o OOfJiOi -M E > E r— i- i™ - i -- i -- i -- i— r— ' l_ ro fa s: suppo :es, •r™ « > 4- CO i_ 4-> CU • o S- CO CO U O s3- OJ CM Ol r— CO CTi ^~ CO O CD O ro o CO 5- • ••••« Q-i— -i- cu O orvcoMinn i— «sj- i— CM LO O CO CO rtJ E CO u_ O O i— CO CO LO OiOOO Oi— •— o s- 3 o to i— ^— r— ^— I— - t— I— •r- -M •!- -C cu i- fl O > u u •r~ a ro ro S- -r- •a; cu s- 3 > s- +j -o O- 1- C E OJ "D O ro Q CO e o o ro 4-> o >> s- c V cu cu cu s- ro O *3- O 00 00 CO r^. oi en lo CO 1^ LO "^" S- -E E o ■*-> >> ro +■* ro 4- T- E o "* vo rv oj cc 00 CX> C3^ r- i— ^- r»«. c a. o E 1_ o o t— oo ro co CX> C> Ol C O O O i— CO i- CO •!- c o o co ro co ■»-> >,o E ro IB^r- O CD i— II •i- E -E C}- «CM ■*->•!- O d) co r»» ro S- 3 O +-> CT> a cu co E i— •I- &> O) cu 4- E - U E >- •i- •!- CO •i- o o p— Oi-rooicoLn co o-i LO LO CO CC T3 05r— S- D. ro O C (O o_ E •t-> Or-NfONN CO CM i— i— CO «d" CO CO E a; t- o o O o cy> o «— co cnooc o o o cr i- +-> O \— !■■• r-~ ^™ i -- i~~ I™ - 1— ^~ r— r am n— +J ». cu •r- ci- co +J o +-> ro CU ro •i- 4- i- •«- E i— ro a ■»-> Cl S- s- •!- -a E o i— i S- •r- i— cu ro •!- .c: O CO 1 5 1 CU ro •!- -a 4- c 3 =3 S- co cu r— -a 3 (J C 4- CO c: ro 1 E ^— +J C cu 03 O CU E rC 1— 1 1 — 1 EEC 1— > t— 1 1— 1 > ' 1— 1 1 — 1 > a. t. > HH Ml — 1 1 — 1 1— 1 1— 1 1 — i co -i- cu , . CU 3 > o cro . . • • o cu o cm co *3- co co r-> CM CO r-^ i—. r*^ i^ r^ r» r^ r^. CT> CTi Ol CTi Cn O" > CT> en r ~ 1 cu -t-> >->->->->-> >- >- o 1 <~ ) (_ > c_ ) c_ > c_ ) c_ ) c_ ) c_ ) z 146 o -o CU to 3 a> c tO •!- JC c a o $- o a. i Q 0J o to o (13 cu o Q- U > - « o to +J c CD £R s- co o I S- CO •■- Navy and Marine Corps cm r>- lo r*- . en cm 00 ■— co 1— CTt 00 cn 1— r^ r-» CD co lo en en en co lo «d- CM CO CO «3- ►—i 1 — 1 1—1 1— 1 cn 5- C_) LO en >- C_) LO cn >- CY 1977: -o +-> c S- Q- to 3 CU tO O •i— •> > to S- +-> CU • J- to to CL to CO CD S- i- -4-> "O c c DO* c o f0 +-> s- c CU CD CU S--CE (O +J (O a. o E to S_ •< c tO 03 tO (T3 (t3 U -a o en r- U S- 3 cu to CU c « 1- to CDi— C #. CU L|_ to -M n3 CU to i- -r- E o +-> S- -r- T3 •r- 1— CU fO •<- _C cr o cu o o sz CD o a cu +-> o 147 00 CD CO u O- S- o en cu ■M Q fO O O 25 -Q 3 J- CO o <♦- s- o CO 'I— ) 02:0 +-> o fO >>>— i— -O II <+- CM aj •• r-^ O CO cx> -4-> r— cu c - c o o Q. E +-> O •i- o u •i- c i— o CLT- E +-> I— I •!"- c 1 3 I CO ai -O rO 00 •i- +J u o S- 3 CD "TD E O E S- o a. o •a a> on ■0 4J4J (O C 1- fO 0) i— Q. Q- O c S- o O T- c o I +-> 0) E c u i- A3 CD i- > +-> o c C5 O fT3 CO o OONOOtJUO o co id i— co cr> o o i— co <5t <3- LD CO LO CM «^- i— i— r^ i-» r-. cm co co co cm i — cncrioo OOi— i— O O CO CO CM CTi o en o r-^ co co o o co Ln en o r— r— i— r— r— CM o o o o cm co «3- cr> CO CT) i— CM CO r-. CM CO CTiCTiOO OOf— i— O CM f— CM CM CO "3" CO O O i — CO LO LO CO CT> CM CTi "^ LO CM CM l^>r^CMCM CO CO CM CM CTiCTiOO OOi — i — cu CJ •r— > CD 00 a fO Q. eo £ CD CO CO 03 -a o c +-> CJ zs o S- Q- C0 QJ -o Z3 cm co «* en co r-» r-^ r-^ r-~ r^ r-» r^. CT> Ol CT> 0~> CJ1 CT> CM CT) 1 — oououo >- o CO 1-^ CT) >- C_) 148 4- O XJ cu to 3 QJ C to «r- -C C O O S- o Q O O S- o to s- o 4-> (0 CU Q O CD CU +-> fO u _Q oo o o s- i— O II T-3CNJ s: en CD >> O -Q >- •r- O S- «* ex. to +J +-> c •i- a; o c •r- o i— O- 9- E E o i— i CJ> >?f +-> 1 •1 — CQ C 3 CD E r— b -a <: fO (O to •I- +-> u o S_ 3 CD "O E O E J- O a. C_5 CD Ul "O +-» -4-> CD c o cd s- > +■> o c o o a o i c =3 i (T3 n3 o ld ^- r^ o «3- cm o O co >*j- r-» to Or-CTlN CO LO tO «5f- CM CM «3" >vT MflfOi — ^- «3" 00 CO CM «3- ■?»- lo lo «^- LO CT> to oo in to to i— r^^d-cn to CM i — CM i— 00 O CM CO LO LO LO CO O r— LO OOCMCMtO cm cm co **■ «>j- lo to to ocneno i— ooo i— i — i — i — i — i — i—i— CMi — i — CM CMCMCMCM 00 i— CTi CM >5j- CO co r-v oo i— CM r— o r^ CM «3" CM r— i — i — CM CO CO «* i— 00 CM CO LO <^J- O CO i— CM tO LD Lf) LO LO LO to to *3" en >- c_> LO en to en o en o to cd u •r— > S- CD l/i J^ u a. -o c CU -Q E CU to to ro -o (0 o to sz o o 13 o S- Q. to (O o O 3 s- to 3 Q_ S_ O Q •>-> CD u to $- O O -i- o co n r>» i— 00 f— i— r- U. C o O I— r- 00 r— CO Q CTi CTi O CM CM i— i— S- T- O O O CT> O r— O CT> CT> O o o o o •r- > i"~ r~— f— ■ r"* r— " r- ^— f— ^— p— n— >o O -O ■— <+- II •>CM W-PIS s- cm O CJr- (O o.>- I— •!- CJ> CD O" Q UJ CD (J O -r- •i- E t- O O- S- ■M U •i- CD O i— i— i fO I I CO c o •r" 4J CO fO O O -r- •i- C -M C O C 3 ^ ^ E O Q. O CD i- u i— zs CD Q- -O CD c -a 3 C O (O s_ <0 +-> o I CO CD X) fO O O O i— i— CO CT» «3- 00 00 CM «3" CO CO CT> O CTi CT> Ol O CT> CT> O Q CM 00 O O O O OunCMCO^Tv CT> *fr 00 00 CO "5^- Lf) LO 8o8££ CT» CM CT> Cp CTi CT> O O CM CO CTi CT> O O O O • ••••• CM CO «* Lf) CD r-» r>, r-* i^«. r-» r<* r-«. Ci CT> CT> CT> CT> CTi CM CTi CO 1^ CTi 1 — 1 — uuuuuu >- o >• C_3 150 l/l C o • 1— ■(-> fO o •i — c -o 3 CU g 3 E C o •r- o 4-> C «4- o O o CO 1 CU > (O J_ jz o o C7> S- 3 4-> Q. fO O Q JD o 3 r*t CO O $- s~ o o O i— 4- •l-J II ro CM CO SI r^. S- cr> o >»— +-> JD <0 >- f— •>o 4- +j Cu> c C o •r™ •r— O c •i— o ^~ i. Q.-M E u i— i O) p— i UJ 1 -o LT) c 1 fO CO Q) ^™ JD ro CU u CO s- o O 'I- U_ E O S_ •!- •i- > co c o •r- -t-> U5 fO O O -r- •f- i= ■«-> C O E 3 s- cy E u Q. O **- ir> woin o o o o n«3 ^O N O CT> CT> CTi CO O Q «— CVJ-^OO OOOO i— t— CM CM CO CO <£> st CO CM CO C\J CM <£> CO CT> i— «* O f— *j- m m crt OOOO en cm r-* co O — r- r- i— o un co WCMr-r- VO «3- h< sj- CM CO CM CO r— i— ' CM CT) r-^WIs CT> r— VO Ol "vl- «tf- r^, CO sf LO LT) r^. OOOO CO CM CO CM CO •=3- CTi >- o Lf) fx. CT> >- CJ> CX> >- o en o 151 o o Q M- O E O •r— +J (0 in c s- C o o fO o +-> •r— r— ns i— II ^~ •r- CM 4- > r^. O) •r* - Cn Q O r ~ cu -o >- a c <-J •r- 03 i- Dl. ^ +J fO •r" -»-> O •i — Q-5I E i — i i <£> CO QJ «3 > •r" +-> o Olf)i-OCMN O O i— C\J CO «3- roocviiD NCOO^t OiO^OO COr-I^O O O O r— O O »— 1^. 1^- Lf) orvrocofoc^ O O i— i— C\J CM o o\ *? r-. CONtDr- i— cncnovo oooi— OCM«^NOC0 O O i— C\J CM CO •it^i-c- minoi- 00 CO CTV 5£ CTV CT^ CT> O LO LT) r~- i — O O O i— CM CO ^" Lf) V£> C*«. r»>. r>. r^ r^ r^ r«. 0"> 0~> Ol O") Ol CXi C_) CJ> O O O o CM r-. o-> >- o en >- 152 Q O Q c > O o^oocoro O) O Lf) p-% i— oo cm r->. lo «a- vo «3- O f— ** CD o cm «a- oo CM CM CM CM OOr-lD CO CO CO CO 1^. 00 CD P-. CO CO CO «3" c o •r- T3 4-> Q- C E O O O O $- O r- 4- CD c to c &- o o O to o 4-> S- i— 03 CD II «— CL, CM 4- h-. - O •!- O •I- > S- -r- D_ O +J -a •r- c o •r- £ CM CM «* Lf) Lf) Lf) Lf) CO co co lo tn Lf) Lf) Lf) «?f r-r-NN r>% r*» r-. cm MCMWrs CM CM CM CM r-«» r>- r-~» in CM CM CM CO «3 O i— «d- r— CM 00 CM CM 00 co ko cr> o "53- NWO r— i— CO 00 00 01O<* i— i— CM CM Lf) Lf) Lf) r— CM CM CM CO CO r— CM O CO co«d- I 00 a> r— -O 03 •3- CD Lf) CTi cd CD >- o >- o c_> 5- o 153 4- O V) ^ O s~ O o CD J- CD o +-> <+- 03 O 00 -Q s- 23 o oo +■> fU s- o r— o o 4- *"-3 1 — >CT O JD i — •r* 5- #% >- D- c o o -»-> *l — •r— +-> O o •r— 3 i— S- Q.-M E oo i I E o o l^ 00 fO •^ S- E +-> O CD (T3 Q.4-> N E Z3 'r- (0 E E E S- o •■- S- cn •<- s: CD o +j -a s- (j a; o Q. 3 -E F S- -M +-> CD 1/5 4- U E O S- O o O 4- oo^-cocrn o o o ■— o o oooo oooo o o o o oooo oooo oooo o (O CJ ■*-> Z3 •r- t- f— 4J •I- V) 2: e o CJ -t-> o O 00 O O r- O O CJ *t CM «3- CO Or-Mn^iD CM CO CO CM S(OON r- oo cm cm en en o o «d- 00 <£> O OOr-M O CM Cn CM Cn r*. O CM CM i— i— O O r— CM CO St ID CM CO CM i— UT) r— *& 1^ r>« CO CM CM «J-00lfiCfl en en o o o o r— t— CM CO «3" ID r^ r^ r^. r** r~- r-^ r^ en en en - o CO - o 154 <4- O CO 01 CO Q O O %- o I - 2 o CO Ol J- QJ -•-> o ■P nJO (O O i— i— -O II 4- 3 C\J QJ CO r^ O i- i— 01 o o-n>- •i- <0 C_> 2& o »> •i- E ■— O Q.T- i o 3 S- 4-> CO o 00 t_> QJ Construction portion of the Minuteman force modernization program 00*£>«a- ROCOCO CO CO CO i— oo r-» oo co oor^>!j- <^-r^cr>i— i— «^-OLD sfiooo O O O i— i— OJ r— i— ■— O O O O O i— i— Military construction o«^-lo«si- lonms r^roLOr— co ^t- «3- cr> co<£>r-s<£> «— tr> r»» r». c^vD^om coooooo OJCVJCSJi— CM CM CO CO CO «tf- «* "vl" CO CO CO LT> to +■> o 1— p^c\j>^-c\j i— vdoo oor-«.ooo i— un«d-n- <* uivo n) i — cx> r-» co r»- co >* cm co en co p»» C\J CM C\J r— OJOJCOfO CO ^t" «3" **" Cn ID ID tD CY 1974: I II III IV CY 1975: I II III IV CY 1976: I II III IV CY 1977: I II III IV 155 -D CD 4-> (_> CD r— ■o CD c GO 3 u_ S- o -^ M- o o to +-> J- oo o +-> CD fD to r— E «+- CD O CD 4- O O CD f— Q II CD CM O CD 1^. •^« SZ s- +-> i — D_ «f- >- +-> o O •r" O to •r" to O CO x> 1 3 QQ OO CD fD CD O c CD +-> to I/) -Q =3 OO -o c CD CT)p— C •■- •r- -t-> JZ X 4-> CD O +J C_) +-> o 13 o s- CD i — to o +j S- O +-> 3 CD -O Q. O $- Z3 CD O 00 OO CT> CTi <£> O CO CM OO LT) r^ o co i— i— i — c\j oo ^3- cr^ r^ 00 CTi 00 CM kOVOCMM cnoioo inooico i— OO CM i— O 00 00 "d" CTi 0O O O CM oo oo «?t- oo CT> ld o VO O 00 -vj- ooooooo ■=a-r^coi— cncnoo o o o i— o ■— to 00 o oo i— i— CM CM 0O 0O inOiVDO 00 CTi CTi r— 1^ CO r— 0O CTi CT> O O 00 LT) O CTi O r— CM 0O CM 0O «3" LT> <£> (^ r-~. r-~ r-^ r-«» r-^ r-^ Ol Ol CT> Ol Ol Ol CM 1 — oouuuu >- CO cy> >- o 156 "D -o CO i — c Q) o CO o S- 1 o 4- X.» c CO 3 S- U_ o -M .x: fO o ^~ o 4- +J o (UtflO o 1 — s- 4- r— Q- 0) >- •r- CD •r- +-> C_> Q-4- e o CO i =d- Bulk petroleum products •^-i — oicvj «3- co i^~ ld ldi — i— si- o> «3- ir> co lo o r— co lo^i— (X) i— co co cri cr> ld r-. co OOLOCmOi CO CO CT> CO CT> CO CM i— cm co co «3- cm cm cm cm c\jc\jc\jc\j cm cm co co co co co co CY 1974: I II III IV CY 1975: I II III IV CY 1976: I II III IV CY 1977: I II III IV 157 <4- O >, OO s- CD O 00 CD ra cu sz ■•-> u fO i- o 3 jQ Q. 3 co Q o i- Q o 'O ±- TO O 2" M- >> oo -Q S- o «« -M 00 O ro +-> o r^ C f— M- 0) II cd C CM Q o r-. Q-Cn CD B<— u o •r - o >- s- CO D- 00 E -M ai •r" +j O oo •r- >> r— co Q- b in i — i CU r— 1 •r— I oo V) en •r^ i 5- en 5 a; ai r— z JD ra rO I +-> -m cu to C C -r- a; •■- r— i— C S.i- w •r (O fO 0) +J E J2 <— C J3 -r- O 3 "O 00 (J to CD 00 i~ -C T- dl"D u E -t-> c c C ro 3 i— i ro to s- > CU i— > -C -r- ro ■*-> 00 Z O 00 CU (J 00 t- CU O S- ■— LL. CU T- .c to S.-P w •r- O >r- > CU 00 H c l/l S- •1 — •r— CT> LO CM CO Cn CT> Cn CT> O 00 LO CO O CO O tO LO CM 00 r~~ O O O O CM CM lO O CO CO torsco "3- LO Lf> CM O O O oo r-. OCTii-OCTn O 00 O CM -sd- LO o cn cn cn r-» r~ r-pocorN tocor-^oo LO CO CO CO O 01 Cn Ol OOOCTlN CT> O O to «* CO «* <* o lo i — to i — r-. O WCTi CTiOO enrvroro r— oo en >3- r— o o oo o o o en co to LO r— cn cn cn cn cm co <* lo to r^. r^ r-~ r-~ r-- r-- r-» en en en en en en uuuuou CM en >- to co en >- o 00 cu ■r- 4-> n T- 00 > CU :r- •i- -M +J a •r- ro r— •r- -a u cu ro "D 4- C 3 -a 4- s= • ro >> E r— ro +-> i— S- C ro cn CU T- O E S- s- Q.-M Q. •i- 00 Z3 3 C cr-o o cu c •1" •r- +-> -»-> ro i- c N O -i- •r- Q. C 0-XJ s- 3 CU cu 00 U -o 3 o '■a fc 00 o ■l-> s- CU i- Q. o ro s- C2- 00 o +J <4- s- c •r- cu c: ro C ro Q- O t CU Q. CU S- E 4-> O 3 -a o C c rr- ro "D s: C CU ro CU S- ^: ro « -«-> Q. 00 00 CU M- O O «N -^ 00 > C C 5- o O CU •r— •i- 00 +-> +J s- ro i— o U fO Q_ •r- 3 4- -t-> C •i- U o -a E -M a E 3 CU o S- i— o +J •r- to to s- c oo cu o •i- sz o E -M o CU CU -C ■a -a -l-> 3 C i— ro CU O -a C en 3 •i- c i^ •1 — <_) +-> s- C O CU •r - C CU E 4J 00 T- O CU CD c o c Q CU 00 CU 1 o o CU -M O • 158 re I +-> +-> CU to E E •.- CU -r- i— r E S- r— to •r- «3 <0 Ol c 2 ■r- O 3T! to <_> oo CD to s- je •r— JJt) o & +J c c cm 3 i — i re -a =5 £ If- •!- o -*-> c to o a; o co /a i o >> S- J- 3 O a. en ai O -p o (tJ o u J- 13 o to s- to o S- -i-j o re +->z: o re O <— >><— <*- x> II CU CM CO Cft CU +■> r— u e £ £ O a. o o o CO s- cu >> CU r— > .£ -r- ro 4-> to 2 O ifl CD U to S- i- CD o O) i — J- -£ "r- -l-> e o O) re c_> r— •r- >> o >!j- co LT> C\J cooler-- r- cm cm lo VDOlT- r--« r^ r-» r— ai o cvhd Ol Old O Oi — i — i— LD O CM tO CO "53" "d" CO CM CO CO t^ LT> Lf) LT> Lf) Lf) IT) «>f r- «*00r— «* tO O CO 00 tO tO CM (T) CM «* O O C31 CTi i— CM CTi r*. CD Lf) oooiocm 00COCOCO Lf) CM CM Lf) «* tO r— O r-i — COLO r— I— CD LO r-CTi«d-i— o^->*Ln r~- cr> cm i — r— CD Lf) r-«. r- to to *3- r» **• lt) CO r- o> o co co r-. oo r--. to r~ en CM CM CM «* Lf) CM t^~ 00 r— •* «* O O CO «* LO O CM r— LD CTiOOOr- 00 CT) 00 CD "3" CM on cr> o oo l — tO Lf) OY rvvocors CM f- «3" 00 r-~. i— . r»<. r^ to r^ to co NCM LOW CO Lf) ^f CM o to to «d" CO CT> ■— Lf) co co en en Lf) •* r— LO - C_) Lf) cr> >- o to en >- o en >- <_) to cu ■1— ■»-> •> ■!— to > CU -i- •1- ■♦-> -M o •i- re r— •r- -a U CD ro T3 c»- £ 3 -a m- e • ro >> E r— re -M r— s- £ ro cn CO *i~ E J- O Q-+J Q. •i- CO 3 3 E CT-O O ai c •r— *r- +-> +J re S- C M O »r- •i — Q. E Q.T3 S- zs a; CU co a -o ^ o «"D fc: to O -l-> J- CU s- a. o ro S- a. to o ■»-> H- s- c T- 0) C ro c re 0J Q- CD t- E +J o 3 •o <-> C c *r~ re tj ^' c a) re cu s- .E ro « -M Q- to co a> 4- o O ** •!" to > E E S- O o a> •r- •r- CO 4-> +j S- ro i— o v ro Q. •r- 3 <+- +j E •■- a O X) re •i — o s- -4-> E +-> s- E O at re 1 — r— H- >- CD o Q CD O •r- s- D_ o Q- E i o t— CO CD jD ro Government furnished equipment for new ships OlTl^OOOM CO r— Or— O CO Lf> CO ororoivoi'^ cr»ooo Oi— coco O O i— i— i— CM CT> O O O OOOO New ship construction at private shipyards O ^ CM O IT) Ol pv. O i— CM NOOM OfOfOlflr-N OiOOO i— CO «t3" LO Or-rOlfilsCO CT>000 i— i— i—i— ro O r- OCOCOr-ON I s * r— i— i— CM [-«. CO 00 OOiLOOCTiCO CT^OOO NOOr- CM oocM«^-«d-Ln cnooo ooi— r— • • ■ • CMro«d-LOcor» cm co r-* r-» r*. r-«. r-. r-» r^ £* CT> CX> C7> CT> C7> CT> CT> C7> >->->->->->- >- >- O O C_> (_> O O O <_) cu E a. •r— CT CD CU -C CO ■r— E C- 4- C (Li E C s- CD > O c o •r - CO S- C o CJ ro co "O S- ro >> Q. jr co 0) +J ro > •r— s- o. +-> ro a> c o XJ • •a CO cu C "O O 3 CO o S- X cu cu > c cu o s- O ro cu +J ■G 160 -D e o o oo Q. GO 00 CD to 10 J- >- O O Ljl -E E CO 4J CD E E 5 o e o 00 •1 — T3 +J J- u A3 3 >> S- Q. -l-> •r* to .e e 00 o o CU 4-> D. rO •i — > -E •r" 00 S- Q. 2 CU ■P s: (O 4-> O I o CO cu r— f0 CO Lncvj<* cnj cr> o-> i — ono^i- i— r^ r^. cr> a^cvjcoN O i— ■— i— co lo i— o r— i— C\J CNJ 00 r— 00 00 r-^SIr-rr- CNJ O i— "5J" CNJ CO CNJ CNJ r-COCOr- cnj ^- r^ r-» O CO CO o 00 CO LO O ro i^ co cnj CNI CNJ co «* o cnj r^. o Lf> LO LO LO 00 O i— LO «^- i^ r^ co 00 00 00 cr> ■vf LO >5f <3" LO i— LO i— LO LO LO LO en >- C_) LO en >- LO en o 4-> E CU E CL ■r— ZZ CT CU ■D cu E 3 +-> E CU E E $- CU > o CJ3 E O 00 cu > E o (J ■D E fO 00 -a s- >> Q. •r— x: 00 cu +-> (0 > Cl +-> E CU O S- en >- o cu o 2: 161 CO 4- O 00 O) 00 03 x: o S- 3 D_ Q O Q S- oo o en 4- •r— oo x: i- h- O o •M 4- o fO o r— t— II 4- c CM r— 0) 03 u +-> >- •r~ S- o S- o Q_ D- 00 -(-> c: • r— ra <_> s- I O) OJ co i— t/> 09 OJ O CM O CO CM r- CO CO CM CO id id •** cd C O •i — t— O ID CD «3" O CO CM LD ^f CD i— i— CO r— E > O i— CM CO CD CO CD CD O O r— r— r— CM S- S- i— p— r— r - r— r™ qj a; h~ CO -^ O CM O «d- 1^ r— CD CO CO CO CVJf-Nr- o 3 O CO O CM CD CM r^ o o o CO CO CO CD s- O O CM CO CO «3" CD O O O O O O O 1— O CD 00 VD i— O CM O O CD O O CO CD *r™" 03 O CM CD ID CD O CD O O O r-r-MN cc O O r— CO ID r>* CD O O O o o o o CM O i— CD CM «d- CD CO O O CD r— «3" 1^ id 03 qj O CM O r— CM CD ID CO 00 CO CO i — CD <3" CO 00«1- WNCO CD O CD O O O CD O - o o r^. co co cd CO r- O O rv o «3- r^ r— r^ r^ f^ r— CD CM Csjcvj^- r^ O O O CM i— i »— i > • ••••• m % % cm co «a- id co r^ CM CO r-^ r^ r^ r^» r^ r-» r>. r^ CD CD CD CD CD CD CD CD 1 — r— >->->->- >~ >- >- >- oooooo O o O) c E o •r— ■r™ Q. +J a. 03 •r— 4-> x: i- 10 o CL r— co 03 c •p— 03 QJ 00 E s- S- E a> •r- O •f— 03 o s- 4- E 03 O o CJ CO 4- t— QJ 03 oo c •r— 03 o O x: 'r- i~ o +-> QJ S- 03 £ 3 +-> £ Q- S- O o U CO CL s 00 £ < C o ce: 03 l- f— S- 4- ^ +-> C_) X> i— i 03 QJ rj a; 10 o- CO 03 x: T3 4- u C o s- 03 3 V) Q. ft _J 3 S- a. QJ #\ oo -a -o c c -o 03 03 QJ F E +-> E E 03 O O i— CJ> CJ> a> -M • +J 4- oo 4- T3 •r* s- •i— £Z )— ai r— 03 S- •r" o3 •r- S- a> CJ) <: S_ CO C 03 •p- >. (J >> s. s- s- o 03 r— 03 -a -M 03 -»-> QJ •r- 'r— •F— > r— U r— QJ •r" $- "T" •*-> S ai E S 00 CO t- oo 00 QJ O QJ QJ "O u "O -a 3 3 3 r— ^> r^ r~- a o U CJ c: s- c E »— i 4- (-H i— i CM CO 162 to CD to rO sz u JU 3 Q_ ■o OJ O 3 8 e -!-> J- c o o <*- o to 1 s- o to o +■> DO (0 E i— ^» •r- II t~ E CM cd i— rv Q CT> 4- i— a> o o >■ •r— c o s- o D_ «^" +J 4-> ro •p- +-> U i- *r— o p— O- Q. CO E c ►— < ro S- i h- I CQ CD -Q rO co r— co ro S- i- CD a> l— oo u r0 a; c\j rC O CO E «3- «3|- CO CO CM CM CO O CM CM C\J CO IV CT) CM CM • — rv lt) «sa- rocvjro* rv rv rv rv CO Lf) co o CM i— O O CM CM CO CO rvr^oicr> i — rv «=3- cm ro o rv ro rv rv o i— en cm co en si- cm lo rv r— i— CM CM CM CM O «3" co co co ro IV r— IV CM co co co «si- CO O Sj" LO CO si- <5T si- coroLnrs rv co en en CT) CO CO Lf) cm rv Lf) cn CO CO CO CD O >— CM CM CO CO CD CO CM CO CO «d" co en cn rv St «vf- Lf) CO CO O O Lf) rv rv rv co CM LT) CO O IV IV i— CM i— «3" i— i— If) Lf) CO CO r-ff)OlS CM CM CO Lf) CO i— O CTi CO CO LO Lf) i-r-CMN VOMSN co co en en «3- id cm en CO CO ^f Lf) rv iv r— r- r-i-OllS CM 00 «Sf Lf) O O i— i— rv cm lo lo f— CM CM CM Lo «sa- CM CM IT) O CO sfr O 00 00 sj- «^- CO CO «5j" CM i— CO IV LOOOlCTl - o CO rv en >- o IV rv cn >- O ro +-> CD to E S- S- E CD •r- O T- 03 O S- S- o 5 o ro •r-* O s- CD E TJ CD to ro E O s- a; to to i— i ro _l 3 i- Q- CD •> (/) -D -O E E "O ro ro CD C_) O CD i— ro i— CD to E to to CD O CD CD -do -a -a 3 3 3 a o o o E S- E E (— I 4_ |_H I 1 CM CO 163 M- O to CD to CO JZ o s- 3 O- e Q o O 00 Q CD i. O- O M- M- o to S- e o o +-> • 1 — o fO 4-> o r— <0 f — <4- 4-> II QJ i- C\J Q o r». Q-CTi CD to i — O e •r™ 03 >- s- S- <_) Q_ r— 4-» T3 •i- E u ro a. i — i CO S- I r— I CO CD r0 CD r— r— -Q to 03 i— CD OfOCVJSOW o o o o i— i— ID LD to CD to S- > E o o co o co r-^ o o o o oddd =3 ro CD O O O i— CNJ CO o o o o o o o o -O S- Q. E +-> X ■!"■ CD CD cm O CO LO O LO O r— O O LO ID ID r— i— r— •i — O CO 00 oo c\j en co i— en oo 00 00 00 00 03 OO^OWPTt OOOiOi Cn Cn Ci Cn cm o r^ co cr> to id ^f CNJ CO 00 oo *d- r^ en CO 3 o id en ^d- en cnj r^ en i— i— cnj co r» r^ CD o o o r— co r^ en en o o o o o o o en o to 00 CO r— i— 1^. CO ^J- LO r— CO S~ •i— O CO Cvi LO ID tO CTi CTv O i— t-^t-roto < O O i— CNJ CO "* en en o o o o o o OONNOW ID ID «3" P^s oo o ld r^ fO +-> O CNJ tO CO 00 r— encnoo O CNJ i— CO o O O O i— CNJ «3- CMTiOO o o o o h- 1— 1 1—1 1— 1 1— 1 > 1 1 l-H > cnj co >=3- lo to r^ CNJ CO r-, r^ i^ t*^ r^ r^ r-v r» CT> Cn C7> Cn Cn Cn Cn en >->->->->->- >- >- C_> C_> O O O o o O to CD T3 e E •l~ fO to =3 #s o E E CD •t— •r - <— "D 9\ S- to >> J- X 03 CD -p e — o •r— E >> "O CD e +J fO f0 > e •r- fO S- •r— Q. ^-* •r— #» > CD •r" CJ O E ro Q S O o Q r ~ O ro 4-> E to O • +-> •i - 00 E +-> CD CD ro OO E O E CD o CD to r— a. S- to X 3 •r— cu .a -a E oo •r— «> 3 CD to o S_ S- a> ro E to O ro CD i— ■o r— i — 3 ro O) ^— +J u U E OO E CD •r— t— 1 S- b 164 to cu to 03 -a JZ cu o 3 i- E 13 •i — D_ 4-> E Q o O (J Q S- o to M- E O lO to J- s- o cu +-> D- o 03 o ■ — <+- 1 — il- O II CU CNJ Q E r-« O CTi CU •i — i — CJ +J • 1 — 03 >- S- +J o Q- o 4-> Q. •r— to O E •r- 03 i— s- Q.I— E i — i i -a E "3 CNJ CU 1 > 03 ca I— cu i — -Q n3 J- Reimbursable travel , expenses =3- cm ro *d- -3- r-nN cm ro ro ro cm >=a- «=d- «^- <^f «3" ID ID to CO C3~> i-- VD CO CM CM "vf- ID WO CM O CM CM < co en o O O O i— i— t— o o i— r— CM CM o ro r— 00 CM CM ID WD CO CO WO WO VDIX3NIS i- ro o o~> ro CT» ID i— 1^ <3" CO «— WO ID CO ID WO rx-jcn •=*■ id O O r— i— ID ^" KO ID CM CM CM CM 00OO1M cm ro ro ^d- CO ID ID O ■vl" «vf- ^J" ID r— +-> o 1— ro oo i— C30 r^ <3- CT> O ro r— CM o CO WO "3" CO co co cr> o o O O i— ID ID WO CTi CM i— CT> i— cm ro ro CT> O O CO CO "vl- "vl" >* i — i i— t i— i i— i >- C_> ID <3N >- C_> WO 1^ CTi >- C_> 1^ en >- c_> to CU "D E E •r— 03 to 13 r\ o E fc CU ■P" •1 — »— ■o n S- to cu -O. CL 03 a n cu S- 03 o «4- CU CU E E E CU O i— to o S- •■- CU -E Q. CU > >> i- "D 03 CU +-> E •r- s r- o E >> •o cu E -M 03 03 > E «r- 03 i- «r- Q. > CU •■- O U E 03 O <: o o Q i— O 03 +-> E to O • +-> -i- to C+J CJ CU 03 lO o cu o to i — s- to 3 •<- .o -a cu to 03 U E CU CL X Q) to o cu E 03 to CU -a i— i— 3 03 CU .— -MO U E tO E CU -r- i-t S- E 165 *4- O l/> CD I/) 03 •C u Q- Q O Q s- o s- o +-> A3 CD Q CD o ■r- s- Q. O) a. o o _ CM .o cr> to >- 0) o +-> 4-> •r- •!■" O i— •r- "r- i— -M E CO co CD 03 to 0) ■#■• +-> •r— r— OOr-M00r- «3- «d" CM CM CO «3- O CO 4-> OLnoNO^sr CO CD O CM CM If) U3 r co o CO OONNi-lD CD CTi CD O O O O i— r— i— i— i— C\J CM >> ■«- > •^ o orvfo^coro CO CO If) CO VO If) CM If) •f— i- OINCMr-r-P) OOOVOr- CO If) VD OOlflCnOM CTi CT> O O O O O i— u i— r- i— r— CM CM r— i" - r— i— r*» r - i CO CTi co r» cm if) «5f- CO CD O fO ■»-> ONCVJWNn CO CD O i— CO If) If) «d- O OO^SClr- CTi CTi O O O O O i— 1— • • • • ■ • ■ • « • • • • • • • * • • • • • • • • 1— 1 ►—! 1—4 »— 1 > i— i i — i > • ••••• CM CO *J" If) l£> 1-^ CM CO i~— r — r-. r — r-^ r~- r^ r^ CD CT> CT) CD CD CD CTi CD i — i — r— i— i — i — p— ' >->->->->->- >■ >- o cj o o c_> o o o 166 it- CD lO CD to fO JZ o J- 3 Q_ Q -a o cu Q 3 e s- •^ o +J 4- c o (/) o s- o 1 ■»-> re CD O r— Q.O 4- >»— CU h- II o CM s- r-^ CD o a-> U •r-jr— • r— fO s- s: 5- D_ O >> -•-> X •i— o A •r- V) r— (0 Q. •r- E -M M *r- ^~ | •r- 1 -»-> ID co 1 — CO CD ^— -O fO to CD •r— ■P •r— •r- +J S- CD -C o cm r^ *d- lo Ps^J-^Di — LO LO CT> «3" OCOCON CO CT> i— i— O O i— i— ^■IfiOlr- CM CM CM n CTl i— «3" i— ro «^- CO LO O O CM OO CM CM CM CM lo LO r^ oo LO LO LO r-» CM CM CM CM ■n- u •r- s- +J u CD LU r— O CO O WCONN M^mn CM i— CT> O r— 1^ r^ LO co «3- lo r- en oo >3- co oo oo cr> cr> cotvinrv CTi CT> O O i— i— i— CM CO CM CM CM CM CM (0 +-> O 1— IT) CO i — i — VDONCM ^fr-00O «3- O «=J- LO CM CO LO LO CM CM 00 CM r^. r^. oo co onomo oo co en o cm o i— ro O i— CM CM CM CM CM CM i — i i — i i— i 1— H "3" CTi >- c_> LO r-» r— 5- C_> LO CT> >- C_) i^ cr> 5- C_> 167 Appendix C: Observation Group Members and Meeting Agenda The observation group organized to review the methodology and to evaluate the data and procedures in this study consisted of: Barry Blechman Brookings Institution James Dei Rossi Interplan Corporation John Despres Arms Control and Disarmament Agency John Layng Bureau of Labor Statistics Donovan Leyden Office of Management and Budget Jerome Mark Bureau of Labor Statistics Clifford Miller Office of the Assistant Secretary of Defense Joel Popkin National Bureau of Economic Research Irving Rottenberg National Bureau of Economic Research Richard Ruggles Yale Unversity T. Arthur Smith Management Analysis, Inc. Arnold Strasser Office of Management and Budget John Turner U.S. Department of the Treasury Agendas for the meetings consisted of: October 19, 1976 1. Welcome and background 2. DOD participation 3. Organization and goals 4. Stock funds and ammunition 5. Aircraft and missiles 6. Utilities, transportation, and communications 7. Shipbuilding 8. Research and development 9. Compensation 10. Adjournment May 17, 1977 1 . Opening remarks 2. Status report 3. Discussion of concepts a. Compensation b. Military retired pay c. Construction d. Health services e. Contract maintenance services 4. Treatment of certain types of purchases a. Government-owned contractor-operated plants b. Government- furnished material and equipment 169 c. Industrial funds d. Warm production base 5. Army ammunition 6. Army and Air Force missiles 7. Air Force aircraft 8. Combat vehicles 9. Adjournment June 27, 1978 1 . Opening remarks 2. Status report 3. Discussion of concepts a. Contract maintenance services b. Automatic data processing services c. Aircraft quality adjustment d. Aircraft new product procedures 4. Examination of results 5. Future plans 6. Adjournment 170 Appendix D: Appropriation Categories Funds for DOD military functions are appropriated under eight major titles, as follows: Military Personnel Retired Military Personnel Operation and Maintenance Procurement Research, Development, Test, and Evaluation Military Construction Family Housing Civil Defense These titles generally contain several specific categories to which funds are appropriated. The FY 1976 budget contained 58 categories under these titles. Following are brief defi nitons of the major appropriation titles and a list of the categories included under each. Military Personnel The appropriations under this title finance the pay and allowance of officers, enlisted personnel, cadets, and midshipmen; the subsistence of enlisted personnel; permanent change of station travel; and other military personnel costs. Categories Military Personnel Air Force Army Marine Corps Navy Reserve Personnel Air Force Army Marine Corps Navy National Guard Personnel Air Force Army Retired Military Personnel The appropriations under this title finance the pay of all personnel on DOD military retired lists. Requirements cover (a) payments to retired officers and enlisted personnel of the Air Force, Army, Marine Corps, and Navy; (b) retainer pay of enlisted personnel of the Fleet Reserve of the Navy and Marine Corps; and (c) survivors' benefits. Category Retired Pay, Defense 171 Operation and Maintenance The appropriations under this title finance the day-to-day costs, except military personnel costs, of operating and maintaining the Armed Forces, including the Reserve components, and related DOD support activities. These funds include amounts for pay of civilians, contract services for maintenance of equipment and facilities, fuel, supplies, and repair parts for weapons and equipment. Categories Operation and Maintenance Air Force Army Marine Corps Navy Defense Agencies Air Force Reserve Army Reserve Marine Corps Reserve Navy Reserve Air National Guard Army National Guard National Board for the Promotion of Rifle Practice, Army Naval Petroleum Reserve Claims, Defense Contigencies, Defense Court of Military Appeals, Defense Procurement The appropriations under this title finance the acquisition of capital equipment, such as aircraft, missiles, ships, combat and support vehicles, weapons, torpedoes, and communication equipment; air, ground, and ship munitions; major items for support of the capital equipment when it is in use; industrial facilities necessary to produce that equipment; and major modifications of equipment in inventory where modernization can be achieved without buying new equipment. The capital equipment financed by these appropriations is principally procured from private contractors or produced in Govern- ment arsenals, shipyards, and plants. Categories Aircraft Procurement Air Force Army Navy Missile Procurement Air Force Army Other Procurement Air Force Army Navy 172 Procurement Defense Agencies Marine Corps Procurement of Weapons and Tracked Combat Vehicles, Army Ammunition Procurement, Army Weapons Procurement, Navy Shipbuilding and Conversion, Navy Research, Development, Test, and Evaluation The appropriations under this title finance the development, test, and evaluation of new and improved weapon systems and related equip- ment carried out by the Air Force, Army, Navy, Defense agencies, and Director of Test and Evaluation. They also provide for scientific research supporting defense functions and operations. Work is per- formed by industrial contractors, Government laboratories, universi- ties, and not-for-profit organizations. Categories Research, Development, Test, and Evaluation Air Force Army Navy Defense Agencies Director of Test and Evaluation, Defense Military Construction The appropriations under this title finance the acquisition, con- struction, installation and equipment of temporary or permanent public works, military construction, naval installations, and facilities for the armed services and other DOD agencies. Categories Military Construction Air Force Army Navy Defense Agencies Air National Guard Army National Guard Air Force Reserve Army Reserve Naval Reserve Family Housing, Defense The appropriations under this title finance housing expenses for the military services and other DOD agencies for construction, including acquisition, replacement, addition, expansion, extension, and .altera- tion, and for operation, maintenance, and debt repayment, including leasing, minor construction, principal and interest charges, and insurance premiums. 173 Category Family Housing, Defense Civil Defense The appropriations under this title finance expenses necessary for carrying out civil defense activities including financial contribu- tions to States, studies and research to develop measures and plans for civil defense; continuing shelter surveys, marking, and equipping surveyed spaces. Categories Defense Civil Preparedness Agency Research, Shelter Survey, and Marking 174 Glossary of Abbreviations ADP Automatic data processing AFLC Air Force Logistics Command AFSC Air Force Systems Command ASD Aeronautical Systems Division BEA Bureau of Economic Analysis BLS Bureau of Labor Statistics CERCOM Communications and Electronics Material CHAMPUS Civilian Health and Medical Plan of the CPI Consumer Price Index CPR Cost performance report CSIF Communications Services Industrial Fund CY Calendar year DLA Defense Logistics Agency DOD Department of Defense DSF Defense Stock Fund Readiness Uniformed Command Services ECO Engineering change order ESD Electronic Systems Division f.o.b. Free on board FSC Federal Supply Class FY Fiscal year GFE Government furnished equipment GFM Government furnished material GNP Gross National Product . GOCO Government-owned, contractor-operated GOGO Government-owned, Government-operated GS General Schedule ICBM Intercontinental ballistic missile ICC Interstate Commerce Commission IPD Implicit Price Deflator LAP Load, assemble and pack MAC Military Airlift Command MCF Thousand cubic feet MSC Military Seal if t Command MTMC Military Traffic Management Command NAVAIR Naval Air Systems Command NAVELEX Naval Electronics Systems Command NAVFAC Naval Facilities Engineering Command NAVSEA Naval Sea Systems Command 175 NIPA National Income and Product Accounts NSN National Stock Number PPI Producer Price Index (formerly Wholesale Price Index) PPS Probability proportional to size R. &. D. Research and development SLBM Submarine launched ballistic missile SPCC Ships Parts Control Center USACE U.S. Army Corps of Engineers USPS United States Postal Service WBS Work breakdown structure WG Wage Grade WL Wage Leader WS Wage Supervisor